Change search
Refine search result
123 1 - 50 of 110
CiteExportLink to result list
Permanent link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Rows per page
  • 5
  • 10
  • 20
  • 50
  • 100
  • 250
Sort
  • Standard (Relevance)
  • Author A-Ö
  • Author Ö-A
  • Title A-Ö
  • Title Ö-A
  • Publication type A-Ö
  • Publication type Ö-A
  • Issued (Oldest first)
  • Issued (Newest first)
  • Created (Oldest first)
  • Created (Newest first)
  • Last updated (Oldest first)
  • Last updated (Newest first)
  • Disputation date (earliest first)
  • Disputation date (latest first)
  • Standard (Relevance)
  • Author A-Ö
  • Author Ö-A
  • Title A-Ö
  • Title Ö-A
  • Publication type A-Ö
  • Publication type Ö-A
  • Issued (Oldest first)
  • Issued (Newest first)
  • Created (Oldest first)
  • Created (Newest first)
  • Last updated (Oldest first)
  • Last updated (Newest first)
  • Disputation date (earliest first)
  • Disputation date (latest first)
Select
The maximal number of hits you can export is 250. When you want to export more records please use the Create feeds function.
  • 1.
    Ahmed, Muhammad
    et al.
    Department of Computer Science, Technical University of Kaiserslautern, 67663 Kaiserslautern, Germany; Mindgrage, Technical University of Kaiserslautern, 67663 Kaiserslautern, Germany.
    Hashmi, Khurram Azeem
    Department of Computer Science, Technical University of Kaiserslautern, 67663 Kaiserslautern, Germany; Mindgrage, Technical University of Kaiserslautern, 67663 Kaiserslautern, Germany; German Research Institute for Artificial Intelligence (DFKI), 67663 Kaiserslautern, Germany .
    Pagani, Alain
    German Research Institute for Artificial Intelligence (DFKI), 67663 Kaiserslautern, Germany.
    Liwicki, Marcus
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Embedded Internet Systems Lab.
    Stricker, Didier
    Department of Computer Science, Technical University of Kaiserslautern, 67663 Kaiserslautern, Germany; German Research Institute for Artificial Intelligence (DFKI), 67663 Kaiserslautern, Germany .
    Afzal, Muhammad Zeshan
    Department of Computer Science, Technical University of Kaiserslautern, 67663 Kaiserslautern, Germany; Mindgrage, Technical University of Kaiserslautern, 67663 Kaiserslautern, Germany.
    Survey and Performance Analysis of Deep Learning Based Object Detection in Challenging Environments2021In: Sensors, E-ISSN 1424-8220, Vol. 21, no 15Article, review/survey (Refereed)
    Abstract [en]

    Recent progress in deep learning has led to accurate and efficient generic object detection networks. Training of highly reliable models depends on large datasets with highly textured and rich images. However, in real-world scenarios, the performance of the generic object detection system decreases when (i) occlusions hide the objects, (ii) objects are present in low-light images, or (iii) they are merged with background information. In this paper, we refer to all these situations as challenging environments. With the recent rapid development in generic object detection algorithms, notable progress has been observed in the field of deep learning-based object detection in challenging environments. However, there is no consolidated reference to cover the state of the art in this domain. To the best of our knowledge, this paper presents the first comprehensive overview, covering recent approaches that have tackled the problem of object detection in challenging environments. Furthermore, we present a quantitative and qualitative performance analysis of these approaches and discuss the currently available challenging datasets. Moreover, this paper investigates the performance of current state-of-the-art generic object detection algorithms by benchmarking results on the three well-known challenging datasets. Finally, we highlight several current shortcomings and outline future directions.

  • 2.
    Ahmed, Tawsin Uddin
    et al.
    Department of Computer Science and Engineering, University of Chittagong, Chittagong, Bangladesh.
    Hossain, Sazzad
    Department of Computer Science and Engineering, University of Liberal Arts Bangladesh, Dhaka, Bangladesh.
    Hossain, Mohammad Shahadat
    Department of Computer Science and Engineering, University of Chittagong, Chittagong, Bangladesh.
    Islam, Raihan Ul
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Computer Science.
    Andersson, Karl
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Computer Science.
    A Deep Learning Approach with Data Augmentation to Recognize Facial Expressions in Real Time2022In: Proceedings of the Third International Conference on Trends in Computational and Cognitive Engineering: TCCE 2021 / [ed] M. Shamim Kaiser; Kanad Ray; Anirban Bandyopadhyay; Kavikumar Jacob; Kek Sie Long, Springer Nature, 2022, p. 487-500Conference paper (Refereed)
    Abstract [en]

    The enormous use of facial expression recognition in various sectors of computer science elevates the interest of researchers to research this topic. Computer vision coupled with deep learning approach formulates a way to solve several real-world problems. For instance, in robotics, to carry out as well as to strengthen the communication between expert systems and human or even between expert agents, it is one of the requirements to analyze information from visual content. Facial expression recognition is one of the trending topics in the area of computer vision. In our previous work, a facial expression recognition system is delivered which can classify an image into seven universal facial expressions—angry, disgust, fear, happy, neutral, sad, and surprise. This is the extension of our previous research in which a real-time facial expression recognition system is proposed that can recognize a total of ten facial expressions including the previous seven facial expressions and additional three facial expressions—mockery, think, and wink from video streaming data. After model training, the proposed model has been able to gain high validation accuracy on a combined facial expression dataset. Moreover, the real-time validation of the proposed model is also promising.

  • 3.
    Bai, Yifan
    et al.
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Signals and Systems.
    Kanellakis, Christoforos
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Signals and Systems.
    Nikolakopoulos, George
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Signals and Systems.
    SA-reCBS: Multi-robot task assignment with integrated reactive path generation2023In: 22nd IFAC World Congress: Yokohama, Japan, July 9-14, 2023, Proceedings / [ed] Hideaki Ishii; Yoshio Ebihara; Jun-ichi Imura; Masaki Yamakita, Elsevier, 2023, p. 7032-7037Conference paper (Refereed)
    Abstract [en]

    In this paper, we study the multi-robot task assignment and path-finding problem (MRTAPF), where a number of robots are required to visit all given tasks while avoiding collisions with each other. We propose a novel two-layer algorithm SA-reCBS that cascades the simulated annealing algorithm and conflict-based search to solve this problem. Compared to other approaches in the field of MRTAPF, the advantage of SA-reCBS is that without requiring a pre-bundle of tasks to groups with the same number of groups as the number of robots, it enables a part of robots needed to visit all tasks in collision-free paths. We test the algorithm in various simulation instances and compare it with state-of-the-art algorithms. The result shows that SA-reCBS has a better performance with a higher success rate, less computational time, and better objective values.

    Download full text (pdf)
    fulltext
  • 4.
    Bai, Yifan
    et al.
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Signals and Systems.
    Lindqvist, Björn
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Signals and Systems.
    Karlsson, Samuel
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Signals and Systems.
    Kanellakis, Christoforos
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Signals and Systems.
    Nikolakopoulos, George
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Signals and Systems.
    Multi-Robot Task Allocation Framework with Integrated Risk-Aware 3D Path Planning2022In: 2022 30th Mediterranean Conference on Control and Automation (MED), IEEE, 2022, p. 481-486Conference paper (Refereed)
    Abstract [en]

    This article presents an overall system architecture for multi-robot coordination in a known environment. The proposed framework is structured around a task allocation mechanism that performs unlabeled multi-robot path assignment informed by 3D path planning, while using a nonlinear model predictive control(NMPC) for each unmanned aerial vehicle (UAV) to navigate along its assigned path. More specifically, at first a risk aware 3D path planner D∗+ is applied to calculate cost between each UAV agent and each target point. Then the cost matrix related to the computed trajectories to each goal is fed into the Hungarian Algorithm that solves the assignment problem and generates the minimum total cost. NMPC is implemented to control the UAV while satisfying path following and input constraints. We evaluate the proposed architecture in Gazebo simulation framework and the result indicates UAVs are capable of approaching their assigned target whilst avoiding collisions.

    Download full text (pdf)
    fulltext
  • 5.
    Banerjee, Avijit
    et al.
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Signals and Systems.
    Mukherjee, Moumita
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Signals and Systems.
    Satpute, Sumeet Gajanan
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Signals and Systems.
    Nikolakopoulos, George
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Signals and Systems.
    Resiliency in Space Autonomy: a Review2023In: Current Robotics Reports, E-ISSN 2662-4087, Vol. 4, p. 1-12Article, review/survey (Refereed)
    Abstract [en]

    Purpose of Review: The article provides an extensive overview on the resilient autonomy advances made across various missions, orbital or deep-space, that captures the current research approaches while investigating the possible future direction of resiliency in space autonomy.

    Recent Findings: In recent years, the need for several automated operations in space applications has been rising, that ranges from the following: spacecraft proximity operations, navigation and some station keeping applications, entry, decent and landing, planetary surface exploration, etc. Also, with the rise of miniaturization concepts in spacecraft, advanced missions with multiple spacecraft platforms introduce more complex behaviours and interactions within the agents, which drives the need for higher levels of autonomy and accommodating collaborative behaviour coupled with robustness to counter unforeseen uncertainties. This collective behaviour is now referred to as resiliency in autonomy. As space missions are getting more and more complex, for example applications where a platform physically interacts with non-cooperative space objects (debris) or planetary bodies coupled with hostile, unpredictable, and extreme environments, there is a rising need for resilient autonomy solutions.

    Summary: Resilience with its key attributes of robustness, redundancy and resourcefulness will lead toward new and enhanced mission paradigms of space missions.

    Download full text (pdf)
    fulltext
  • 6.
    Barman, Sourav
    et al.
    Noakhali Science and Technology University, Noakhali, Bangladesh.
    Biswas, Md Raju
    Noakhali Science and Technology University, Noakhali, Bangladesh.
    Marjan, Sultana
    Noakhali Science and Technology University, Noakhali, Bangladesh.
    Nahar, Nazmun
    Noakhali Science and Technology University, Noakhali, Bangladesh.
    Hossain, Mohammad Shahadat
    University of Chittagong, Chittagong, Bangladesh.
    Andersson, Karl
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Computer Science.
    Transfer Learning Based Skin Cancer Classification Using GoogLeNet2023In: Machine Intelligence and Emerging Technologies - First International Conference, MIET 2022, Proceedings, part 1 / [ed] Md. Shahriare Satu; Mohammad Ali Moni; M. Shamim Kaiser; Mohammad Shamsul Arefin; Mohammad Shamsul Arefin, Springer Science and Business Media Deutschland GmbH , 2023, Vol. 1, p. 238-252Conference paper (Refereed)
    Abstract [en]

    Skin cancer has been one of the top three cancers that can be fatal when caused by broken DNA. Damaged DNA causes cells to expand uncontrollably, and the rate of growth is currently increasing rapidly. Some studies have been conducted on the computerized detection of malignancy in skin lesion images. However, due to some problematic aspects such as light reflections from the skin surface, differences in color lighting, and varying forms and sizes of the lesions, analyzing these images is extremely difficult. As a result, evidence-based automatic skin cancer detection can help pathologists improve their accuracy and competency in the early stages of the disease. In this paper, we present a transfer ring strategy based on a convolutional neural network (CNN) model for accurately classifying various types of skin lesions. Preprocessing normalizes the input photos for accurate classification; data augmentation increases the amount of images, which enhances classification rate accuracy. The performance of the GoogLeNet transfer learning model is compared to that of other transfer learning models such as Xpection, InceptionResNetVe, and DenseNet, among others. The model was tested on the ISIC dataset, and we ended up with the highest training and testing accuracy of 91.16% and 89.93%, respectively. When compared to existing transfer learning models, the final results of our proposed GoogLeNet transfer learning model characterize it as more dependable and resilient.

  • 7.
    Blaszczyk, Martin
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering.
    Autonomous Quadcopter Landing with Visual Platform Localization2023Independent thesis Advanced level (professional degree), 20 credits / 30 HE creditsStudent thesis
    Abstract [en]

    Multicopters such as quadcopters are a popular tool within industries such as mining, shipping and surveillance where a high level of autonomy can save time, increase efficiency and most importantly provide safety. While Unmanned Aerial Vehicles have been a big area in research and used in the mentioned industries, the level of autonomy is still low. Simple actions such as loading and offloading payload or swapping batteries is still a manual task performed by humans. If multicopters are to be used as an autonomous tool the need for solutions where the machines can perform the simplest task such as swapping batteries become an important stepping stone to reach the autonomy goals. Earlier works propose landing solutions focused on landing autonomous vehicles but the lack of accuracy is hindering the vehicles to safely dock with a landing platform. This thesis combines multiple areas such as trajectory generation, visual marker tracking and UAV control where results are shown in both simulation and laboratory experiments. With the use of a Model Predictive Controller for both trajectory generation and UAV control, a multicopter can safely land on a small enough platform which can be mounted on a small mobile robot. Additionally an algorithm to tune the trajectory generator is presented which shows how much weights can be increased in the MPC controller for the system to remain stable. 

    Download full text (pdf)
    fulltext
  • 8.
    Borngrund, Carl
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Embedded Internet Systems Lab.
    Dump truck object detection dataset including scale-models2020Data set
    Abstract [en]

    Object detection is a vital part of any autonomous vision system and to obtain a high performing object detector data is needed. The object detection task aims to detect and classify different objects using camera input and getting bounding boxes containing the objects as output. This is usually done by utilizing deep neural networks.When training an object detector a large amount of data is used, however it is not always practical to collect large amounts of data. This has led to multiple different techniques which decreases the amount of data needed. Examples of such techniques are transfer learning and domain adaptation. Working with construction equipment is a time consuming process and we wanted to examine if it was possible to use scale-model data to train a network and then used that network to detect real objects with no additional training.This small dataset contains training and validation data of a scale dump truck in different environments while the test set contains images of a full size dump truck of similar model. The aim of the dataset is to train a network to classify wheels, cabs and tipping bodies of a scale-model dump truck and use that to classify the same classes on a full-scale dump truck.

  • 9.
    Borngrund, Carl
    et al.
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Embedded Internet Systems Lab.
    Hammarkvist, Tom
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering.
    Bodin, Ulf
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Embedded Internet Systems Lab.
    Sandin, Fredrik
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Embedded Internet Systems Lab.
    Semi-Automatic Video Frame Annotation for Construction Equipment Automation Using Scale-Models2021In: IECON 2021 – 47th Annual Conference of the IEEE Industrial Electronics Society, IEEE, 2021Conference paper (Refereed)
    Abstract [en]

    Data collection and annotation is a time consuming and costly process, yet necessary for machine vision. Automation of construction equipment relies on seeing and detecting different objects in the vehicle’s surroundings. Construction equipment is commonly used to perform frequent repetitive tasks, which are interesting to automate. An example of such a task is the short-loading cycle, where the material is moved from a pile into the tipping body of a dump truck for transport. To complete this task, the wheel loader needs to have the capability to locate the tipping body of the dump truck. The machine vision system also allows the vehicle to detect unforeseen dangers such as other vehicles and more importantly human workers. In this work, we investigate the viability to perform semi-automatic annotation of video data using linear interpolation. The data is collected using scale-models mimicking a wheel-loaders approach towards a dump truck during the short-loading cycle. To measure the viability of this type of solution, the workload is compared to the accuracy of the model, YOLOv3. The results indicate that it is possible to maintain the performance while decreasing the annotation workload by about 95%. This is an interesting result for this application domain, as safety is critical and retaining the vision system performance is more important than decreasing the annotation workload. The fact that the performance seems to retain with a large workload decrease is an encouraging sign.

  • 10.
    Cacciarelli, Davide
    et al.
    Department of Applied Mathematics and Computer Science, Technical University of Denmark, Kgs. Lyngby, Denmark; Department of Mathematical Sciences, Norwegian University of Science and Technology, Trondheim, Norway.
    Kulahci, Murat
    Luleå University of Technology, Department of Social Sciences, Technology and Arts, Business Administration and Industrial Engineering. Department of Applied Mathematics and Computer Science, Technical University of Denmark, Kgs. Lyngby, Denmark.
    Tyssedal, John Sølve
    Department of Mathematical Sciences, Norwegian University of Science and Technology, Trondheim, Norway.
    Robust online active learning2024In: Quality and Reliability Engineering International, ISSN 0748-8017, E-ISSN 1099-1638, Vol. 40, no 1, p. 277-296Article in journal (Refereed)
    Abstract [en]

    In many industrial applications, obtaining labeled observations is not straightforward as it often requires the intervention of human experts or the use of expensive testing equipment. In these circumstances, active learning can be highly beneficial in suggesting the most informative data points to be used when fitting a model. Reducing the number of observations needed for model development alleviates both the computational burden required for training and the operational expenses related to labeling. Online active learning, in particular, is useful in high-volume production processes where the decision about the acquisition of the label for a data point needs to be taken within an extremely short time frame. However, despite the recent efforts to develop online active learning strategies, the behavior of these methods in the presence of outliers has not been thoroughly examined. In this work, we investigate the performance of online active linear regression in contaminated data streams. Our study shows that the currently available query strategies are prone to sample outliers, whose inclusion in the training set eventually degrades the predictive performance of the models. To address this issue, we propose a solution that bounds the search area of a conditional D-optimal algorithm and uses a robust estimator. Our approach strikes a balance between exploring unseen regions of the input space and protecting against outliers. Through numerical simulations, we show that the proposed method is effective in improving the performance of online active learning in the presence of outliers, thus expanding the potential applications of this powerful tool.

    Download full text (pdf)
    fulltext
  • 11.
    Carlbaum Ekholm, Erik
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Signals and Systems.
    Towards Robust Localization: Deep Feature Extraction with Convolutional Neural Networks2020Independent thesis Basic level (degree of Bachelor), 10 credits / 15 HE creditsStudent thesis
    Abstract [en]

    The ability for autonomous robotics to localize themselves in the environment is crucial and tracking the change of features in the environment is key for visual based odometry and localization. When shifting into rough environments of dust, smoke and poor illumination as well as erratic movements common in MAVs however, that task becomes substantially more difficult. This thesis explores the ability of the deep classifier CNN architecture to retain detailed and noise tolerant feature maps out of sensor fused images for feature tracking in the context of localization. The proposed method is enriching the RGB image with data from thermal images which is fed into a AlexNet or VGG-16 and extracted as a feature map at a specific layer. This feature map is used to detect feature points and is used to pair feature points between frames resulting in a discrete vector field of feature change. Preliminary complementary methods for the selection of channels are also developed.

    Download full text (pdf)
    fulltext
  • 12.
    Chhipa, Prakash Chandra
    et al.
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Embedded Internet Systems Lab.
    Chopra, Muskaan
    CCET, Punjab University, Chandigarh, India.
    Mengi, Gopal
    CCET, Punjab University, Chandigarh, India.
    Gupta, Varun
    CCET, Punjab University, Chandigarh, India.
    Upadhyay, Richa
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Embedded Internet Systems Lab.
    Chippa, Meenakshi Subhash
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Embedded Internet Systems Lab.
    De, Kanjar
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Embedded Internet Systems Lab.
    Saini, Rajkumar
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Embedded Internet Systems Lab.
    Uchida, Seiichi
    Human Interface Laboratory, Kyushu University, Fukuoka, Japan.
    Liwicki, Marcus
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Embedded Internet Systems Lab.
    Functional Knowledge Transfer with Self-supervised Representation Learning2023In: 2023 IEEE International Conference on Image Processing: Proceedings, IEEE , 2023, p. 3339-3343Conference paper (Refereed)
  • 13.
    Chhipa, Prakash Chandra
    et al.
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Embedded Internet Systems Lab.
    Rodahl Holmgren, Johan
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Embedded Internet Systems Lab.
    De, Kanjar
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Embedded Internet Systems Lab. Video Coding Systems, Fraunhofer Heinrich-Hertz-Institut, Berlin, Germany.
    Saini, Rajkumar
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Embedded Internet Systems Lab.
    Liwicki, Marcus
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Embedded Internet Systems Lab.
    Can Self-Supervised Representation Learning Methods Withstand Distribution Shifts and Corruptions?2023In: 2023 IEEE/CVF International Conference on Computer Vision Workshops (ICCVW 2023), Institute of Electrical and Electronics Engineers Inc. , 2023, p. 4469-4478Conference paper (Refereed)
  • 14.
    Chhipa, Prakash Chandra
    et al.
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Embedded Internet Systems Lab.
    Upadhyay, Richa
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Embedded Internet Systems Lab.
    Saini, Rajkumar
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Embedded Internet Systems Lab.
    Lindqvist, Lars
    Optimation Advanced Measurements AB, Luleå, Sweden.
    Nordenskjold, Richard
    Optimation Advanced Measurements AB, Luleå, Sweden.
    Uchida, Seiichi
    Human Interface Laboratory, Kyushu University, Fukuoka, Japan.
    Liwicki, Marcus
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Embedded Internet Systems Lab.
    Depth Contrast: Self-supervised Pretraining on 3DPM Images for Mining Material Classification2022In: Computer Vision – ECCV 2022: 17th European Conference, Tel Aviv, Israel, October 23–27, 2022, Proceedings, Part VI / [ed] Avidan, S.; Brostow, B.; Cissé, M.; Farinella, G.M.; Hassner, H., Springer Nature, 2022, Vol. VI, p. 212-227Conference paper (Refereed)
  • 15.
    De, Kanjar
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Embedded Internet Systems Lab.
    Investigating pretrained self-supervised vision transformers for reference-based quality assessment2023Conference paper (Refereed)
  • 16.
    Dibs, Hayder
    et al.
    Water Resources Management Engineering Department, Faculty of Engineering, Al-Qasim Green University, Babylon, Iraq.
    Ali, Alaa Hussein
    Building and Construction Techniques Engineering Department, Al-Mustaqbal University College, 51001 Hillah, Babylon, Iraq.
    Al-Ansari, Nadhir
    Luleå University of Technology, Department of Civil, Environmental and Natural Resources Engineering, Mining and Geotechnical Engineering.
    Abed, Salwan Ali
    College of Science, University of Al-Qadisiyah, Diwaniyah, 58001, Iraq.
    Fusion Landsat-8 Thermal TIRS and OLI Datasets for Superior Monitoring and Change Detection using Remote Sensing2023In: Emerging Science Journal, E-ISSN 2610-9182, Vol. 7, no 2, p. 428-444Article in journal (Refereed)
    Abstract [en]

    Currently, updating the change detection (CD) of land use/land cover (LU/LC) geospatial information with high accuracy outcomes is important and very confusing with the different classification methods, datasets, satellite images, and ancillary dataset types available. However, using just the low spatial resolution visible bands of the remotely sensed images will not provide good information with high accuracy. Remotely sensed thermal data contains very valuable information to monitor and investigate the CD of the LU/LC. So, it needs to involve the thermal datasets for better outcomes. Fusion plays a big role to map the CD. Therefore, this study aims to find out a refining method for estimating the accurate CD method of the LU/LC patterns by investigating the integration of the effectiveness of the thermal satellite data with visible datasets by (a) adopting a noise removal model, (b) satellite images resampling, (c) image fusion, combining and integrating between the visible and thermal images using the Grim Schmidt spectral (GS) method, (d) applying image classification using Mahalanobis distances (MH), Maximum likelihood (ML) and artificial neural network (ANN) classifiers on datasets captured from the Landsat-8 TIRS and OLI satellite system, these images were captured from operational land imager (OLI) and the thermal infrared (TIRS) sensors of 2015 and 2020 to generate about of twelve LC maps. (e) The comparison was made among all the twelve classifiers' results. The results reveal that adopting the ANN technique on the integrated images of the combined TIRS and OLI datasets has the highest accuracy compared to the rest of the applied image classification approaches. The obtained overall accuracy was 96.31% and 98.40%, and the kappa coefficients were (0.94) and (0.97) for the years 2015 and 2020, respectively. However, the ML classifier obtains better results compared to the MH approach. The image fusion and integration of the thermal images improve the accuracy results by 5%–6% from the proposed method better than using low spatial-resolution visible datasets alone.

    Download full text (pdf)
    fulltext
  • 17.
    Dibs, Hayder
    et al.
    Water Resources Management Engineering Department, Al-Qasim Green University, Babylon, 51001, Iraq.
    Jaber, Hussein Sabah
    Department of Surveying, College of Engineering, University of Baghdad, Baghdad, Iraq.
    Al-Ansari, Nadhir
    Luleå University of Technology, Department of Civil, Environmental and Natural Resources Engineering, Mining and Geotechnical Engineering.
    Multi-Fusion algorithms for Detecting Land Surface Pattern Changes Using Multi-High Spatial Resolution Images and Remote Sensing Analysis2023In: Emerging Science Journal, E-ISSN 2610-9182, Vol. 7, no 4, p. 1215-1231Article in journal (Refereed)
    Abstract [en]

    Producing accurate Land-Use and Land-Cover (LU/LC) maps using low-spatial-resolution images is a difficult task. Pan-sharpening is crucial for estimating LU/LC patterns. This study aimed to identify the most precise procedure for estimating LU/LC by adopting two fusion approaches, namely Color Normalized Brovey (BM) and Gram-Schmidt Spectral Sharpening (GS), on high spatial-resolution Multi-sensor and Multi-spectral images, such as (1) the Unmanned Aerial Vehicle (UAV) system, (2) the WorldView-2 satellite system, and (3) low-spatial-resolution images like the Sentinel-2 satellite, to generate six levels of fused images with the three original multi-spectral images. The Maximum Likelihood method (ML) was used for classifying all nine images. A confusion matrix was used to evaluate the accuracy of each single classified image. The obtained results were statistically compared to determine the most reliable, accurate, and appropriate LU/LC map and procedure. It was found that applying GS to the fused image, which integrated WorldView 2 and Sentinel-2 satellite images and was classified by the ML method, produced the most accurate results. This procedure has an overall accuracy of 88.47% and a kappa coefficient of 0.85. However, the overall accuracies of the three classified multispectral images range between 86.84% to 76.49%. Furthermore, the accuracy assessment of the fused images by the Brovey method and the rest of the GS method and classified by the ML method ranges between 85.75% to 76.68%. This proposed procedure shows a lot of promise in the academic sphere for mapping LU/LC. Previous researchers have mostly used satellite images or datasets with similar spatial and spectral resolution, at least for tropical areas like the study area of this research, to detect land surface patterns. However, no one has previously investigated and examined the use and application of different datasets that have different spectral and spatial resolutions and their accuracy for mapping LU/LC. This study has successfully adopted different datasets provided by different sensors with varying spectral and spatial levels to investigate this.

    Download full text (pdf)
    fulltext
  • 18.
    Dibs, Hayder
    et al.
    Water Resources Engineering Department, Faculty of Engineering, Al-Qasim Green University, Al-Qasim, Iraq; Faculty of Engineering, Geospatial Information Science Research Centre, Department of Civil Engineering, University Putra Malaysia, Selangor, Malaysia.
    Mansor, Shattri
    Faculty of Engineering, Geospatial Information Science Research Centre, Department of Civil Engineering, University Putra Malaysia, Selangor, Malaysia.
    Ahmad, Noordin
    Faculty of Engineering, Geospatial Information Science Research Centre, Department of Civil Engineering, University Putra Malaysia, Selangor, Malaysia; National Space Agency Malaysia (ANGKASA), Kementerian Sains, Teknologi dan Inovasi, Pusat Angkasa Negara, Banting, Malaysia.
    Al-Ansari, Nadhir
    Luleå University of Technology, Department of Civil, Environmental and Natural Resources Engineering, Mining and Geotechnical Engineering.
    Robust Radiometric Normalization of the near Equatorial Satellite Images Using Feature Extraction and Remote Sensing Analysis2023In: Engineering, ISSN 1947-3931, E-ISSN 1947-394X, Vol. 15, no 2, p. 75-89Article in journal (Refereed)
    Abstract [en]

    Relative radiometric normalization (RRN) minimizes radiometric differences among images caused by inconsistencies of acquisition conditions rather than changes in surface. Scale invariant feature transform (SIFT) has the ability to automatically extract control points (CPs) and is commonly used for remote sensing images. However, its results are mostly inaccurate and sometimes contain incorrect matching caused by generating a small number of false CP pairs. These CP pairs have high false alarm matching. This paper presents a modified method to improve the performance of SIFT CPs matching by applying sum of absolute difference (SAD) in a different manner for the new optical satellite generation called near-equatorial orbit satellite and multi-sensor images. The proposed method, which has a significantly high rate of correct matches, improves CP matching. The data in this study were obtained from the RazakSAT satellite a new near equatorial satellite system. The proposed method involves six steps: 1) data reduction, 2) applying the SIFT to automatically extract CPs, 3) refining CPs matching by using SAD algorithm with empirical threshold, and 4) calculation of true CPs intensity values over all image’ bands, 5) preforming a linear regression model between the intensity values of CPs locate in reverence and sensed image’ bands, 6) Relative radiometric normalization conducting using regression transformation functions. Different thresholds have experimentally tested and used in conducting this study (50 and 70), by followed the proposed method, and it removed the false extracted SIFT CPs to be from 775, 1125, 883, 804, 883 and 681 false pairs to 342, 424, 547, 706, 547, and 469 corrected and matched pairs, respectively.

    Download full text (pdf)
    fulltext
  • 19.
    Divak, Martin
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering.
    Simulated SAR with GIS data and pose estimation using affine projection2017Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
    Abstract [en]

    Pilots or autonomous aircraft need to know where they are in relation to the environment. On board aircraft there are inertial sensors that are prone to drift which requires corrections by referencing against known items, places, or signals. One such method of referencing is with global navigation satellite systems, and others, that are highlighted in this work, are based on using visual sensors. In particular the use of Synthetic Aperture Radar is emerging as a viable alternative.

    To use radar images in qualitative or quantitative analysis they must be registered with geographical information. Position data on an aircraft or spacecraft is not sufficient to determine with certainty what or where it is one is looking at in a radar image without referencing other images over the same area. It is demonstrated in this thesis that a digital elevation model can be split up and classified into different types of radar scatterers. Different parts of the terrain yielding different types of echoes increases the amount of radar specific characteristics in simulated reference images.

    This work also presents an interpretation of the imaging geometry of SAR such that existing methods in Computer Vision may be used to estimate the position from which a radar image has been taken. This is a direct image matching without requiring registration that is necessary for other proposals of SAR-based navigation solutions. By determination of position continuously from radar images, aircraft could navigate independently of day light, weather, and satellite data.

    Download full text (pdf)
    fulltext
  • 20.
    Edstedt, Johan
    et al.
    Computer Vision Laboratory, Linköping University.
    Berg, Amanda
    Computer Vision Laboratory, Linköping University.
    Felsberg, Michael
    Computer Vision Laboratory, Linköping University.
    Karlsson, Johan
    Statens Medieråd, Stockholm.
    Benavente, Francisca
    Statens Medieråd, Stockholm.
    Novak, Anette
    Statens Medieråd, Stockholm.
    Grund Pihlgren, Gustav
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Embedded Internet Systems Lab.
    VidHarm: A Clip Based Dataset for Harmful Content Detection2022In: 2022 26th International Conference on Pattern Recognition (ICPR), Institute of Electrical and Electronics Engineers (IEEE), 2022, p. 1543-1549Conference paper (Refereed)
    Abstract [en]

    Automatically identifying harmful content in video is an important task with a wide range of applications. However, there is a lack of professionally labeled open datasets available. In this work VidHarm, an open dataset of 3589 video clips from film trailers annotated by professionals, is presented. An analysis of the dataset is performed, revealing among other things the relation between clip and trailer level annotations. Audiovisual models are trained on the dataset and an in-depth study of modeling choices conducted. The results show that performance is greatly improved by combining the visual and audio modality, pre-training on large-scale video recognition datasets, and class balanced sampling. Lastly, biases of the trained models are investigated using discrimination probing.VidHarm is openly available, and further details are available at the webpage https://vidharm.github.io/

  • 21.
    Estrela, Vania V.
    et al.
    Telecommunications Department, Universidade Federal Fluminense (UFF), Niteroi, RJ, Brazil.
    Hemanth, Jude
    Department of Electrical and Computer Engineering, Karunya Institute of Technology and Sciences, Coimbatore, India.
    Saotome, Osamu
    Instituto Tecnologico de Aeronautica (ITA), DCTA-ITA-IEEA, Sao Jose dos Campos, SP, Brazil.
    Nikolakopoulos, George
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Signals and Systems.
    Sabatini, Roberto
    RMIT University, School of Aerospace, Mechanical and Manufacturing Engineering, Melbourne, Australia.
    Conclusions2020In: Imaging and Sensing for Unmanned Aircraft Systems Volume 2: Deployment and applications, Institution of Engineering and Technology , 2020, p. 247-248Chapter in book (Other academic)
    Abstract [en]

    The current awareness in unmanned aerial vehicles (UAVs) has prompted not only military applications but also civilian uses. Aerial vehicles’ requirements aspire to guarantee a higher level of safety comparable to see-and-avoid conditions for piloted aeroplanes. The process of probing obstacles in the path of a vehicle and determining whether they pose a threat, alongside measures to avoid these issues, is known as see and avoid or sense and avoid. Other types of decision-making tasks can be accomplished using computer vision and sensor integration since they have a great potential to improve the performance of the UAVs. Macroscopically, UAVs are cyber-physical systems (CPSs) that can benefit from all types of sensing frameworks, despite severe design constraints, such as precision, reliable communication, distributed processing capabilities and data management. This book is paying attention to several issues that are still under discussions in the field of UAV-CPSs. Thus, several trends and needs are discussed to foster criticism from the readers and to provide further food for thought.

  • 22.
    Estrela, Vania V.
    et al.
    Telecommunications Department, Universidade Federal Fluminense, Niteroi, Brazil.
    Hemanth, Jude
    ECE Department, Karunya University, Coimbatore, India .
    Saotome, Osamu
    DCTA-ITA-IEEA, Sao Jose dos Campos, Brazil .
    Nikolakopoulos, George
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Signals and Systems.
    Sabatini, Roberto
    Department of Aerospace Engineering and Aviation, RMIT University, Melbourne, VIC, Australia .
    Conclusions2020In: Imaging and Sensing for Unmanned Aircraft Systems Volume 1: Control and Performance, Institution of Engineering and Technology , 2020, p. 333-335Chapter in book (Other academic)
    Abstract [en]

    The current awareness in UAVs has prompted not only military applications but also civilian uses. Aerial vehicles’ requirements aspire to guarantee a higher level of safety comparable to see-and-avoid conditions for piloted aeroplanes. The process of probing obstacles in the path of a vehicle, and to determine if they pose a threat, alongside measures to avoid problems, is known as see-and-avoid or sense and-avoid involves a great deal of decision-making. Other types of decisionmaking tasks can be accomplished using computer vision and sensor integration since they have great potential to improve the performance of UAVs. Macroscopically, Unmanned Aerial Systems (UASs) are cyber-physical systems (CPSs) that can benefit from all types of sensing frameworks, despite severe design constraints such as precision, reliable communication, distributed processing capabilities, and data management.

  • 23.
    Estrela, Vania V.
    et al.
    Telecommunications Department, Universidade Federal Fluminense, Niteroi, Brazil.
    Hemanth, JudeECE Department, Karunya University, Coimbatore, India.Saotome, OsamuInstituto Tecnologico de Aeronautica, Sao Jose dos Campos, Brazil .Nikolakopoulos, GeorgeLuleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Signals and Systems.Sabatini, RobertoAutonomous and Intelligent Systems Laboratory, RMIT University, Melbourne, VIC, Australia.
    Imaging and sensing for unmanned aircraft systems Volume 1: Control and performance2020Collection (editor) (Other academic)
    Abstract [en]

    This two volume book set explores how sensors and computer vision technologies are used for the navigation, control, stability, reliability, guidance, fault detection, self-maintenance, strategic re-planning and reconfiguration of unmanned aircraft systems (UAS). Volume 1 concentrates on UAS control and performance methodologies including Computer Vision and Data Storage, Integrated Optical Flow for Detection and Avoidance Systems, Navigation and Intelligence, Modeling and Simulation, Multisensor Data Fusion, Vision in Micro-Aerial Vehicles (MAVs), Computer Vision in UAV using ROS, Security Aspects of UAV and Robot Operating System, Vision in Indoor and Outdoor Drones, Sensors and Computer Vision, and Small UAVP for Persistent Surveillance. Volume 2 focuses on UAS deployment and applications including UAV-CPSs as a Testbed for New Technologies and a Primer to Industry 5.0, Human-Machine Interface Design, Open Source Software (OSS) and Hardware (OSH), Image Transmission in MIMO-OSTBC System, Image Database, Communications Requirements, Video Streaming, and Communications Links, Multispectral vs Hyperspectral Imaging, Aerial Imaging and Reconstruction of Infrastructures, Deep Learning as an Alternative to Super Resolution Imaging, and Quality of Experience (QoE) and Quality of Service (QoS).

  • 24.
    Estrela, Vania V.
    et al.
    Telecommunications Department, Federal Fluminense University, Niteroi, Brazil.
    Hemanth, JudeECE Department, Karunya University, Coimbatore, India.Saotome, OsamuInstituto Tecnológico de Aeronáutica, Sao Jose dos Camp, Brazil .Nikolakopoulos, GeorgeLuleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Signals and Systems.Sabatini, RobertoAerospace Engineering and Aviation, RMIT University, Melbourne, VIC, Australia.
    Imaging and sensing for unmanned aircraft systems Volume 2: Deployment and applications2020Collection (editor) (Other academic)
    Abstract [en]

    This two volume book set explores how sensors and computer vision technologies are used for the navigation, control, stability, reliability, guidance, fault detection, self-maintenance, strategic re-planning and reconfiguration of unmanned aircraft systems (UAS). Volume 1 concentrates on UAS control and performance methodologies including Computer Vision and Data Storage, Integrated Optical Flow for Detection and Avoidance Systems, Navigation and Intelligence, Modeling and Simulation, Multisensor Data Fusion, Vision in Micro-Aerial Vehicles (MAVs), Computer Vision in UAV using ROS, Security Aspects of UAV and Robot Operating System, Vision in Indoor and Outdoor Drones, Sensors and Computer Vision, and Small UAVP for Persistent Surveillance. Volume 2 focuses on UAS deployment and applications including UAV-CPSs as a Testbed for New Technologies and a Primer to Industry 5.0, Human-Machine Interface Design, Open Source Software (OSS) and Hardware (OSH), Image Transmission in MIMO-OSTBC System, Image Database, Communications Requirements, Video Streaming, and Communications Links, Multispectral vs Hyperspectral Imaging, Aerial Imaging and Reconstruction of Infrastructures, Deep Learning as an Alternative to Super Resolution Imaging, and Quality of Experience (QoE) and Quality of Service (QoS).

  • 25.
    Estrela, Vania V.
    et al.
    Telecommunications Department, Universidade Federal Fluminense, Niteroi, Brazil; UNICAMP, Brazil.
    Hemanth, Jude
    ECE Department, Karunya University (KU), India.
    Saotome, Osamu
    Instituto Tecnológico de Aeronáutica (ITA), Brazil.
    Nikolakopoulos, George
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Signals and Systems.
    Sabatini, Roberto
    School of Engineering, RMIT University, Australia.
    Preface2020In: Imaging and sensing for unmanned aircraft systems: Volume 1: Control and performance / [ed] Vania V. Estrela; Jude Hemanth; Osamu Saotome; George Nikolakopoulos; Roberto Sabatini, Institution of Engineering and Technology , 2020, p. xiii-xivChapter in book (Other academic)
  • 26.
    Estrela, Vania V.
    et al.
    Telecommunications Department, Universidade Federal Fluminense, Niteroi, Brazil; UNICAMP, Brazil.
    Hemanth, Jude
    ECE Department of Karunya University (KU), India.
    Saotome, Osamu
    Instituto Tecnológico de Aeronáutica (ITA), Brazil.
    Nikolakopoulos, George
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Signals and Systems.
    Sabatini, Roberto
    School of Engineering, RMIT University, Australia.
    Preface2020In: Imaging and Sensing for Unmanned Aircraft Systems: Volume 2: Deployment and Applications / [ed] Vania V. Estrela; Jude Hemanth; Osamu Saotome; George Nikolakopoulos; Roberto Sabatini, Institution of Engineering and Technology , 2020, p. xiii-xviiiChapter in book (Other academic)
  • 27.
    Fredriksson, Scott
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering.
    Design, Development and Control of a Quadruped Robot2021Independent thesis Advanced level (professional degree), 20 credits / 30 HE creditsStudent thesis
    Abstract [en]

    This thesis shows the development of a quadruped platform inspired by existing quadrupled robot designs. A robot by the name of Mjukost was designed, built, and tested. Mjukost uses 12 Dynamixel AX-12a smart servos and can extend its legs up to 19 cm with an operating height of 16 cm. All the custom parts in Mjukost are ether 3d printable or easy to manufacture, and the total estimated cost of Mjukost is around 900$. Mjukost has a simple control system that can position its body freely in 6 DOF using an inverse kinematic model and walk on flat ground using an open-loop walking algorithm. The performance experiments show that its slow control loopcauses difficulties for the robot to follow precise trajectories, but its still consistent in its motions.

    Download full text (pdf)
    fulltext
  • 28.
    Garcia, Laura
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Space Technology.
    UKF-SLAM Implementation for the Optical Navigation System of a Lunar Lander2017Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
    Download full text (pdf)
    fulltext
  • 29.
    Giacomini, Enrico
    et al.
    Luleå University of Technology, Department of Engineering Sciences and Mathematics, Fluid and Experimental Mechanics.
    Westerberg, Lars-Göran
    Luleå University of Technology, Department of Engineering Sciences and Mathematics, Fluid and Experimental Mechanics.
    Nikolakopoulos, George
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Signals and Systems.
    Drones for Planetary Exploration: Modeling Challenges2022In: Svenska Mekanikdagar 2022 / [ed] Pär Jonsén; Lars-Göran Westerberg; Simon Larsson; Erik Olsson, Luleå: Luleå tekniska universitet, 2022Conference paper (Other academic)
    Download full text (pdf)
    fulltext
  • 30.
    Gromova, Arina
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering.
    Weed Detection in UAV Images of Cereal Crops with Instance Segmentation2021Independent thesis Advanced level (degree of Master (One Year)), 10 credits / 15 HE creditsStudent thesis
    Abstract [en]

    Modern weeding is predominantly carried out by spraying whole fields with toxic pesticides, a process that accomplishes the main goal of eliminating weeds, but at a cost of the local environment. Weed management systems based on AI solutions enable more targeted actions, such as site-specific spraying, which is essential in reducing the need for chemicals. To introduce sustainable weeding in Swedish farmlands, we propose implementing a state-of-the-art Deep Learning (DL) algorithm capable of instance segmentation for remote sensing of weeds, before coupling an automated sprayer vehicle. Cereals have been chosen as the target crop in this study as they are among the most commonly cultivated plants in Northern Europe. We used Unmanned Aerial Vehicles (UAV) to capture images from several fields and trained a Mask R-CNN computer vision framework to accurately recognize and localize unique instances of weeds among plants. Moreover, we evaluated three different backbones (ResNet-50, ResNet101, ResNeXt-101) pre-trained on the MS COCO dataset and through transfer learning tuned the model towards our classification task. Some well-reported limitations in building an accurate model include occlusion among instances as well as the high similarity between weeds and crops. Our system handles these challenges fairly well. We achieved a precision of 0.82, recall of 0.61, and F1 score of 0.70. Still, improvements can be made in data preparation and pre-processing to further improve the recall rate. All and all, the main outcome of this study is the system pipeline which, together with post-processing using geographical field coordinates, could serve as a detector for half of the weeds in an end-to-end weed removal system.

    Download full text (pdf)
    fulltext
  • 31.
    Grund Pihlgren, Gustav
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Embedded Internet Systems Lab.
    Deep Perceptual Loss and Similarity2023Doctoral thesis, monograph (Other academic)
    Abstract [en]

    This thesis investigates deep perceptual loss and (deep perceptual) similarity; methods for computing loss and similarity for images as the distance between the deep features extracted from neural networks. The primary contributions of the thesis consist of (i) aggregating much of the existing research on deep perceptual loss and similarity, and (ii) presenting novel research into understanding and improving the methods. This novel research provides insight into how to implement the methods for a given task, their strengths and weaknesses, how to mitigate those weaknesses, and if these methods can handle the inherent ambiguity of similarity.

    Society increasingly relies on computer vision technology, from everyday smartphone applications to legacy industries like agriculture and mining. Much of that groundbreaking computer vision technology relies on machine learning methods for their success. In turn, the most successful machine learning methods rely on the ability to compute the similarity of instances.

    In computer vision, computation of image similarity often strives to mimic human perception, called perceptual similarity. Deep perceptual similarity has proven effective for this purpose and achieves state-of-the-art performance. Furthermore, this method has been used for loss calculation when training machine learning models with impressive results in various computer vision tasks. However, many open questions exist, including how to best utilize and improve the methods. Since similarity is ambiguous and context-dependent, it is also uncertain whether the methods can handle changing contexts.

    This thesis addresses these questions through (i) a systematic study of different implementations of deep perceptual loss and similarity, (ii) a qualitative analysis of the strengths and weaknesses of the methods, (iii) a proof-of-concept investigation of the method's ability to adapt to new contexts, and (iv) cross-referencing the findings with already published works.

    Several interesting findings are presented and discussed, including those below. Deep perceptual loss and similarity are shown not to follow existing transfer learning conventions. Flaws of the methods are discovered and mitigated. Deep perceptual similarity is demonstrated to be well-suited for applications in various contexts.

    There is much left to explore, and this thesis provides insight into what future research directions are promising. Many improvements to deep perceptual similarity remain to be applied to loss calculation. Studying how related fields have dealt with problems caused by ambiguity and contexts could lead to further improvements. Combining these improvements could lead to metrics that perform close to the optimum on existing datasets, which motivates the development of more challenging datasets.

    Download full text (pdf)
    fulltext
  • 32.
    Grund Pihlgren, Gustav
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Embedded Internet Systems Lab.
    Deep Perceptual Loss for Improved Downstream Prediction2021Licentiate thesis, comprehensive summary (Other academic)
    Download full text (pdf)
    fulltext
  • 33.
    Hagström, Björn
    et al.
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering.
    Wallström, Hampus
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering.
    Submillimeter 3D surface reconstruction of concrete floors2022Independent thesis Advanced level (professional degree), 20 credits / 30 HE creditsStudent thesis
    Abstract [en]

    During the creation of any concrete floor the concrete needs to be grinded down from it's very rough newly poured form to a more usable floor surface. Concrete floor grinding is very special in that the work area is often immensely large while the height difference on the surface is incredibly small, in-fact the the largest local difference of the surface from a peek to a valley during the grinding process is submillimeter and goes down to micrometer scale. Today's methods for measuring concrete surfaces are very few and all output one dimensional profiles of the surface in very time consuming processes which makes them unsuitable for real-time analysis of the surfaces during the grinding process. Because of this, the effectiveness of the work is dependent on the experience and intuition of the operator of the grinding machine as they have to make the decision of when to move on to the next step in the grinding process. Therefore it is desirable to create a better method for concrete surface measurement that can measure big areas in a short period of time. In this project a structured light method using sinusoidal phase shifting is implemented and evaluated with an easily movable setup that can measure the height of a concrete surface over an area. The method works by encoding the surface with a phase using a projector and analysing how the phase encoding warps when imaging it from an angle. By triangulation this can be made into a height map of the measured area. The end results show that the method is promising for this application and can detect the submillimeter differences. However, more suitable hardware and a more reliable calibration procedure are required to move this prototype towards a more practical measuring device.

    Download full text (pdf)
    Submillimeter_3D_surface_reconstruction_of_concrete_floors
  • 34.
    Hammarkvist, Tom
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering.
    Dump truck object detection with manual annotations2021Data set
    Abstract [en]

    Doing manual annotations can sometimes be resource heavy, depending on the amount of data. This dataset was designed to created to use in conjunction with a semi-automatic annotation method based on linear interpolation. The dataset contains 799 images, where 679 lies in the trainingset, and the rest lies in the validationset. The images are taken from 6 different video streams, where a remote controlled wheel loader approaches a miniature dump truck at different angles. 4 of the videos are used in the trainingset. The labels can contain up to 5 classes which are:

    0 - front wheel  1 - middle wheel 2 - back wheel 3 - tipping body 4 - cap

    This dataset was used to train a YOLOv3 model, hence the labels will be written in the YOLO labeling format.

  • 35.
    Hashmi, Khurram Azeem
    et al.
    Department of Computer Science, Technical University, 67663 Kaiserslautern, Germany; Mindgarage, Technical University of Kaiserslautern, 67663 Kaiserslautern, Germany; German Research Institute for Artificial Intelligence (DFKI), 67663 Kaiserslautern, Germany.
    Pagani, Alain
    German Research Institute for Artificial Intelligence (DFKI), 67663 Kaiserslautern, Germany.
    Liwicki, Marcus
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Embedded Internet Systems Lab.
    Stricker, Didier
    Department of Computer Science, Technical University, 67663 Kaiserslautern, Germany; German Research Institute for Artificial Intelligence (DFKI), 67663 Kaiserslautern, Germany.
    Afzal, Muhammad Zeshan
    Department of Computer Science, Technical University, 67663 Kaiserslautern, Germany; Mindgarage, Technical University of Kaiserslautern, 67663 Kaiserslautern, Germany; German Research Institute for Artificial Intelligence (DFKI), 67663 Kaiserslautern, Germany.
    Cascade Network with Deformable Composite Backbone for Formula Detection in Scanned Document Images2021In: Applied Sciences, E-ISSN 2076-3417, Vol. 11, no 16, article id 7610Article in journal (Refereed)
    Abstract [en]

    This paper presents a novel architecture for detecting mathematical formulas in document images, which is an important step for reliable information extraction in several domains. Recently, Cascade Mask R-CNN networks have been introduced to solve object detection in computer vision. In this paper, we suggest a couple of modifications to the existing Cascade Mask R-CNN architecture: First, the proposed network uses deformable convolutions instead of conventional convolutions in the backbone network to spot areas of interest better. Second, it uses a dual backbone of ResNeXt-101, having composite connections at the parallel stages. Finally, our proposed network is end-to-end trainable. We evaluate the proposed approach on the ICDAR-2017 POD and Marmot datasets. The proposed approach demonstrates state-of-the-art performance on ICDAR-2017 POD at a higher IoU threshold with an f1-score of 0.917, reducing the relative error by 7.8%. Moreover, we accomplished correct detection accuracy of 81.3% on embedded formulas on the Marmot dataset, which results in a relative error reduction of 30%.

    Download full text (pdf)
    fulltext
  • 36.
    Hashmi, Khurram Azeem
    et al.
    German Research Center for Artificial Intelligence, 67663 Kaiserslautern, Germany; Department of Computer Science, University of Kaiserslautern, 67663 Kaiserslautern, Germany; Mindgrage, University of Kaiserslautern, 67663 Kaiserslautern, Germany.
    Stricker, Didier
    search Center for Artificial Intelligence, 67663 Kaiserslautern, Germany; Department of Computer Science, University of Kaiserslautern, 67663 Kaiserslautern, Germany.
    Liwicki, Marcus
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Embedded Internet Systems Lab.
    Afzal, Muhammad Noman
    Bilojix Soft Technologies, Bahawalpur, Pakistan.
    Afzal, Muhammad Zeshan
    German Research Center for Artificial Intelligence, 67663 Kaiserslautern, Germany; Department of Computer Science, University of Kaiserslautern, 67663 Kaiserslautern, Germany; Mindgrage, University of Kaiserslautern, 67663 Kaiserslautern, Germany.
    Guided Table Structure Recognition through Anchor Optimization2021In: IEEE Access, E-ISSN 2169-3536, Vol. 9, p. 113521-113534Article in journal (Refereed)
    Abstract [en]

    This paper presents the novel approach towards table structure recognition by leveraging the guided anchors. The concept differs from current state-of-the-art systems for table structure recognition that naively apply object detection methods. In contrast to prior techniques, first, we estimate the viable anchors for table structure recognition. Subsequently, these anchors are exploited to locate the rows and columns in tabular images. Furthermore, the paper introduces a simple and effective method that improves the results using tabular layouts in realistic scenarios. The proposed method is exhaustively evaluated on the two publicly available datasets of table structure recognition: ICDAR-2013 and TabStructDB. Moreover, we empirically established the validity of our method by implementing it on the previous approaches. We accomplished state-of-the-art results on the ICDAR-2013 dataset with an average F-measure of 94.19% (92.06% for rows and 96.32% for columns). Thus, a relative error reduction of more than 25% is achieved. Furthermore, our proposed post-processing improves the average F-measure to 95.46% that results in a relative error reduction of more than 35%. Moreover, we surpassed the baseline results on the TabStructDB dataset with an average F-measure of 94.57% (94.08% for rows and 95.06% for columns).

    Download full text (pdf)
    fulltext
  • 37.
    Hedlund, Marcus
    et al.
    Luleå University of Technology, Department of Engineering Sciences and Mathematics.
    Holmström, Caroline
    Luleå University of Technology, Department of Engineering Sciences and Mathematics.
    Deak, Elliot Härenby
    Luleå University of Technology, Department of Engineering Sciences and Mathematics.
    Olsson, Robert
    Luleå University of Technology, Department of Engineering Sciences and Mathematics, Product and Production Development.
    Sjödahl, Mikael
    Luleå University of Technology, Department of Engineering Sciences and Mathematics, Fluid and Experimental Mechanics.
    Öhman, Johan
    Luleå University of Technology, Department of Engineering Sciences and Mathematics, Fluid and Experimental Mechanics.
    Convolutional Neural Networks Applied to Inline Particle Holography2020In: Imaging and Applied Optics Congress, Digital Holography and Three-Dimensional Imaging, Optical Society of America, 2020, article id JW2A.15Conference paper (Refereed)
    Abstract [en]

    Three-dimensional particle positioning from inline holograms is performed using convolutional neural networks. The faster R-CNN architecture is implemented for multi-particle identification and lateral positioning, and a second network estimates the depth position. Supervised learning is used to train the network using simulated holograms.

  • 38.
    Hessinger, Felix
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Space Technology.
    Automation of Operation and Testing for European Space Agency's OPS-SAT Mission2019Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
    Abstract [en]

    This thesis presents a solution for mission operation automation in European Space Agency’s (ESA) OPS-SAT mission. To achieve this, the ESA internal mission automation system (MATIS) in combination with the mission control software (SCOS) are used. They control the satellite and all ground peripherals and programmes to enable fully automated and unsupervised satellite passes. The goal of this work is the transition from the existing manual operation, with a human operator watching over and controlling all systems, to an automated system. This system supports the operation engineer and replaces the operator himself. A large section of this thesis consists of the setup, configuration, integration of all programmes and virtual machines and testing of the MATIS software, as well as the Service Management Framework (SMF) which connects MATIS to non-MATIS applications like SCOS. During testing, many problems could be identified, not only OPS-SAT specific ones, but also general problems applying to all missions that consider using MATIS for future operation automation. These findings and bugs discovered during testing are reported to the responsible authorities and presented in this work. Further features of this thesis are the elaborations of the mission operation automation concept and the satellite pass concept, providing an in-depth view of the automation and passes of OPS-SAT as well as the general concepts and thoughts, which can be used by other missions to accelerate integration. An additional key feature of this thesis is the newly developed standard for operation notation in Excel, which has been achieved in close cooperation with the operation engineer. Furthermore, to accelerate the process of switching from manual to automated procedures, several converters have been developed iteratively with the new standard. These converters allow fast transformation from Excel to the procedure programming language called PLUTO used by MATIS. Not only do the results and converters of this work accelerate the procedure integration by 80%, they also deliver a more stable mission automation system that can be used by other missions as well. Operation automation reduces the operational costs for satellites and space missions significantly, as well as reducing the human error to a minimum. Therefore, this thesis is the first step towards a future with complete automation in the area of satellite operations. Without this automation, future satellite cluster configurations, like Starlink from SpaceX, will not be possible to put into practice, due to their high complexity, exceeding the comprehensibility and reaction time of humans.

    Download full text (pdf)
    fulltext
  • 39.
    Huber, Johannes Albert Josef
    et al.
    Luleå University of Technology, Department of Engineering Sciences and Mathematics, Wood Science and Engineering.
    Forsberg, Fredrik
    Luleå University of Technology, Department of Engineering Sciences and Mathematics, Fluid and Experimental Mechanics.
    Local Stiffness of Softwood Based on Micro- and Macro-Scale Computed Tomography2023In: Proceedings ICM20: Experimental Mechanics in Engineering and Biomechanics: 20th International Conference on Experimental Mechanics / [ed] J.F. Silva Gomes, Institute of Science and Innovation in Mechanical and Industrial Engineering , 2023, p. 1001-1002, article id 20161Conference paper (Refereed)
    Abstract [en]

    Wood is a discontinuous cellular structure on a microscopic scale, but its mechanical behaviour resembles a continuum on a macroscopic scale. The structure of both domains can be studied by X-ray computed tomography (CT). A challenge for accurate CT-based models of wood is to set the values of the orthotropic stiffness tensor locally based on density. Micro-CT scans under in-situ loading may be used to estimate local stiffness in wood, based on strain fields derived from digital volume correlation. The goal of the present paper is to study how micro-CT scans of clearwood under in-situ loading can be used to predict stiffness locally as a function of the apparent macroscopic density, to improve the fidelity of FE models based on macro-CT scans.

    Download full text (pdf)
    fulltext
  • 40.
    Huber, Johannes Albert Josef
    et al.
    Luleå University of Technology, Department of Engineering Sciences and Mathematics, Wood Science and Engineering.
    Olofsson, Linus
    Luleå University of Technology, Department of Engineering Sciences and Mathematics, Wood Science and Engineering.
    Evaluation of Knots and Fibre Orientation by Gradient Analysis in X-ray Computed Tomography Images of Wood2023In: CompWood2023: Computational Methods in Wood Mechanics, Barcelona: International Center for Numerical Methods in Engineering (CIMNE), 2023, p. 143-144Conference paper (Other academic)
    Abstract [en]

    The mechanical properties of wood are governed by growth-dependent structures on themicro- and macroscopic level, which are subject to natural variation. Numerical modelsof wood on a scale of individual pieces of sawn timber may need to account e.g. for thegrowth ring orientation or the presence of knots and their effects on the local materialorientation, i.e. the variation of the fibre coordinate system (FCS). The FCS is composedof the mutually orthogonal longitudinal (l), radial (r) and tangential (t) directions. Agrowth surface represents a region of equal age, i.e. a former growth front of the tree,and at each point on a growth surface, the r direction consequently represents the surfacenormal. Growth surfaces surrounding a knot are usually approximated by analyticalsurfaces, onto which the l direction is modelled by hydrodynamic flow fields in lateral andby polynomials in the vertical direction, like in [1].

    A data-driven method used for detecting the location of knots and eliciting the in-planeprojection of the l direction on the surface of sawn timber is optical scanning combinedwith laser tracheid scanning [2]. A disadvantage of this method is that the internalstructure of the scanned timber remains unknown and needs to be extrapolated, againbased on assumptions of growth in wood. X-ray computed tomography (CT) of woodprovides images of the internal density distribution from which features like the pith, thegrowth rings, knots and defects can be extracted by image analysis.

    In a recent study [3], the local FCSs around knots were reconstructed by density gradientanalysis in CT images, on which finite element models were based for predicting thebending behaviour of sawn timber. Growth surfaces in wood represent regions of nearlyconstant density and can therefore be analysed in CT images by gradient-based methods.The goal of the present study was therefore to study how the gradient of the density fieldderived from CT images of wood can be used to determine growth surfaces, the region ofknots, the border between dead and live knot, and the locally varying FCSs.

    The material studied was comprised of small log sections of Scots pine (pinus sylvestris)(approximate diameters of 50 mm - 300 mm and length 250 mm) containing knot whorls,which were dried to fibre saturation. CT scans were acquired at a voxel size of 0.5 × 0.5 ×0.5 mm3 yielding a 3D image J(x, y, z). After scanning, the sections were cut throughthe centrelines of the knots and manual measurements were conducted of the dead knotborder, i.e. the position along the boundary of a knot after which a knot died off. At this border, the diameter of the knot stops increasing and the surrounding fibres grow aroundthe knot rather than merging with it. As the tree continues to grow, this will eventuallylead to bark being encased around the knot inside the tree.

    A purely data-driven analysis was performed based on the partial derivatives Jx , Jy , Jz ,from which the gradient structure tensor (GST) was constructed, see equation 1, where wσ is a Gaussian convolutional kernel. The eigenvalues and eigenvectors of the GSTwere extracted for each voxel and the resulting vector field was used in the subsequentanalyses. Equivalently, second order derivatives were studied to study curvature.

    The results indicate that gradient-based analyses on CT images of wood can be usedto approximate the locally varying FCSs around knots, see Figure 1 and that they mayfacilitate the determination of the region of dead and live knots.

    Download full text (pdf)
    fulltext
  • 41.
    Jernberg, John
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Computer Science.
    Identification of alkaline fens using convolutional neural networks and multispectral satellite imagery2021Independent thesis Advanced level (degree of Master (One Year)), 10 credits / 15 HE creditsStudent thesis
    Abstract [en]

    The alkaline fen is a particularly valuable type of wetland with unique characteristics.Due to anthropogenic risk factors and the sensitive nature of the fens, protection is highlyprioritized with identification and mapping of current locations being important parts ofthis process. To accomplish this in a cost effective manner for large areas, remote sensingmethods using satellite images might be very effective. Following the rapid developmentin computer vision, deep learning using convolutional neural networks (CNN) is thecurrent state of the art for satellite image classification. Accordingly, this study evaluatesthe combination of different CNN architectures and multispectral Sentinel 2 satelliteimages for identification of alkaline fens using semantic segmentation. The implementedmodels are different variations of the proven U-net network design. In addition, a RandomForest classifier was trained for baseline comparison. The best result was produced bya spatial attention U-net with a IoU-score of 0.31 for the alkaline fen class and a meanIoU-score of 0.61. These findings suggest that identification of alkaline fens is possiblewith the current method even with a small dataset. However, an optimal solution tothis task may require deeper research. The results also further establish deep learningto be the superior choice over traditional machine learning algorithms for satellite imageclassification.

    Download full text (pdf)
    fulltext
  • 42.
    Johansson, Olof
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering.
    Training of Object Detection Spiking Neural Networks for Event-Based Vision2021Independent thesis Advanced level (professional degree), 20 credits / 30 HE creditsStudent thesis
    Abstract [en]

    Event-based vision offers high dynamic range, time resolution and lower latency than conventional frame-based vision sensors. These attributes are useful in varying light condition and fast motion. However, there are no neural network models and training protocols optimized for object detection with event data, and conventional artificial neural networks for frame-based data are not directly suitable for that task. Spiking neural networks are natural candidates but further work is required to develop an efficient object detection architecture and end-to-end training protocol. For example, object detection in varying light conditions is identified as a challenging problem for the automation of construction equipment such as earth-moving machines, aiming to increase the safety of operators and make repetitive processes less tedious. This work focuses on the development and evaluation of a neural network for object detection with data from an event-based sensor. Furthermore, the strengths and weaknesses of an event-based vision solution are discussed in relation to the known challenges described in former works on automation of earth-moving machines. A solution for object detection with event data is implemented as a modified YOLOv3 network with spiking convolutional layers trained with a backpropagation algorithm adapted for spiking neural networks. The performance is evaluated on the N-Caltech101 dataset with classes for airplanes and motorbikes, resulting in a mAP of 95.8% for the combined network and 98.8% for the original YOLOv3 network with the same architecture. The solution is investigated as a proof of concept and suggestions for further work is described based on a recurrent spiking neural network.

    Download full text (pdf)
    fulltext
  • 43.
    Kanellakis, Christoforos
    et al.
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Signals and Systems.
    Karvelis, Petros
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Signals and Systems.
    Nikolakopoulos, George
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Signals and Systems.
    Image Enhancing in Poorly Illuminated Subterranean Environments for MAV Applications: A Comparison Study2019In: Computer Vision Systems: 12th International Conference, ICVS 2019, Thessaloniki, Greece, September 23–25, 2019, Proceedings / [ed] Dimitrios Tzovaras; Dimitrios Giakoumis; Markus Vincze; Antonis Argyros, Springer, 2019, p. 511-520Conference paper (Refereed)
    Abstract [en]

    This work focuses on a comprehensive study and evaluation of existing low-level vision techniques for low light image enhancement, targeting applications in subterranean environments. More specifically, an emerging effort is currently pursuing the deployment of Micro Aerial Vehicles in subterranean environments for search and rescue missions, infrastructure inspection and other tasks. A major part of the autonomy of these vehicles, as well as the feedback to the operator, has been based on the processing of the information provided from onboard visual sensors. Nevertheless, subterranean environments are characterized by a low natural illumination that directly affects the performance of the utilized visual algorithms. In this article, an novel extensive comparison study is presented among five State-of the-Art low light image enhancement algorithms for evaluating their performance and identifying further developments needed. The evaluation has been performed from datasets collected in real underground tunnel environments with challenging conditions from the onboard sensor of a MAV. 

  • 44.
    Kanellakis, Christoforos
    et al.
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Signals and Systems.
    Mansouri, Sina Sharif
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Signals and Systems.
    Fresk, Emil
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Signals and Systems.
    Kominiak, Dariusz
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Signals and Systems.
    Nikolakopoulos, George
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Signals and Systems.
    Aerial imaging and reconstruction of infrastructures by UAVs2020In: Imaging and Sensing for Unmanned Aircraft Systems Volume 2: Deployment and Applications, Institution of Engineering and Technology , 2020, p. 157-176Chapter in book (Other academic)
    Abstract [en]

    This chapter presents a compilation of experimental field trials aiming vision-based reconstruction of large-scale infrastructures using micro aerial vehicles (MAVs). The main focus of this study is on the sensor selection, the data-set generation and on the computer vision algorithms for generating three-dimensional (3D) models. In general, MAVs are distinguished for their ability to fly at various speeds, to stabilise their position and to perform manoeuvres close to large-scale infrastructures. The aforementioned merits constitute aerial robots a highly paced evolving robotic platform for infrastructure inspection and maintenance tasks. Different MAV solutions with task-oriented sensory modalities can be developed to address unique tasks, such as 3D modelling of infrastructures. In this chapter, aerial agents navigate around/along different environments, while collecting visual data for post-processing using structure from motion (SfM) and multi-view stereo (MVS) techniques to generate 3D models [1, 2]. The proposed framework has been successfully experimentally demonstrated in real indoor, outdoor and subterranean environments.

  • 45.
    Karlsson, Samuel
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering.
    Monocular vision-based obstacle avoidance for Micro Aerial Vehicles2020Independent thesis Advanced level (professional degree), 20 credits / 30 HE creditsStudent thesis
    Abstract [en]

    The Micro Aerial Vehicless (MAVs) are gaining attention in numerous applications asthese platforms are cheap and can do complex maneuvers. Moreover, most of the commer-cially available MAVs are equipped with a mono-camera. Currently, there is an increasinginterest to deploy autonomous mono-camera MAVs with obstacle avoidance capabilitiesin various complex application areas. Some of the application areas have moving obstaclesas well as stationary, which makes it more challenging for collision avoidance schemes.This master thesis set out to investigate the possibility to avoid moving and station-ary obstacles with a single camera as the only sensor gathering information from thesurrounding environment.One concept to perform autonomous obstacle avoidance is to predict the time near-collision based on a Convolution Neural Network (CNN) architecture that uses the videofeed from a mono-camera. In this way, the heading of the MAV is regulated to maximizethe time to a collision, resulting in the avoidance maneuver. Moreover, another interestingperspective is when due to multiple dynamic obstacles in the environment there aremultiple time predictions for different parts of the Field of View (FoV). The method ismaximizing time to a collision by choosing the part with the largest time to collision.However, this is a complicated task and this thesis provides an overview of it whilediscussing the challenges and possible future directions. One of the main reason was thatthe available data set was not reliable and was not provide enough information for theCNN to produce any acceptable predictions.Moreover, this thesis looks into another approach for avoiding collisions, using objectdetection method You Only Lock Once (YOLO) with the mono-camera video feed. YOLOis a state-of-the-art network that can detect objects and produce bounding boxes in real-time. Because of YOLOs high success rate and speed were it chosen to be used in thisthesis. When YOLO detects an obstacle it is telling where in the image the object is,the obstacle pixel coordinates. By utilizing the images FoV and trigonometry can pixelcoordinates be transformed to an angle, assuming the lens does not distort the image.This position information can then be used to avoid obstacles. The method is evaluated insimulation environment Gazebo and experimental verification with commercial availableMAV Parrot Bebop 2. While the obtained results show the efficiency of the method. To bemore specific, the proposed method is capable to avoid dynamic and stationary obstacles.Future works will be the evaluation of this method in more complex environments with multiple dynamic obstacles for autonomous navigation of a team of MAVs. A video ofthe experiments can be viewed at:https://youtu.be/g_zL6eVqgVM.

    Download full text (pdf)
    fulltext
  • 46.
    Karlsson, Samuel
    et al.
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Signals and Systems.
    Bai, Yifan
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Signals and Systems.
    Perception capabilities for ARWs: The art of perceiving the environment2023In: Aerial Robotic Workers: Design, Modeling, Control, Vision, and Their Applications / [ed] George Nikolakopoulos, Sina Sharif Mansouri, Christoforos Kanellakis, Elsevier, 2023, p. 67-78Chapter in book (Other academic)
    Abstract [en]

    This Chapter presents vision-based perception modules for an Aerial Robotic Worker, as a basic component for the autonomy of the platform. Generally, visual information could be involved in various levels of autonomy, from object tracking to workspace mapping and motion estimation. The major challenge today is the development of autonomously operating aerial agents capable of completing missions independently of human interaction. To this extent, onboard perception should be developed for the aerial platform to perceive its surroundings and estimate its motion, enhancing the overall navigation and guidance skills.

  • 47.
    Karlsson, Samuel
    et al.
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Signals and Systems.
    Koval, Anton
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Signals and Systems.
    Kanellakis, Christoforos
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Signals and Systems.
    Nikolakopoulos, George
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Signals and Systems.
    D+: A risk aware platform agnostic heterogeneous path planner2023In: Expert systems with applications, ISSN 0957-4174, E-ISSN 1873-6793, Vol. 215, article id 119408Article, review/survey (Refereed)
    Abstract [en]

    This article establishes the novel D+*, , a risk-aware and platform-agnostic heterogeneous global path planner for robotic navigation in complex environments. The proposed planner addresses a fundamental bottleneck of occupancy-based path planners related to their dependency on accurate and dense maps. More specifically, their performance is highly affected by poorly reconstructed or sparse areas (e.g. holes in the walls or ceilings) leading to faulty generated paths outside the physical boundaries of the 3-dimensional space. As it will be presented, D+* addresses this challenge with three novel contributions, integrated into one solution, namely: (a) the proximity risk, (b) the modeling of the unknown space, and (c) the map updates. By adding a risk layer to spaces that are closer to the occupied ones, some holes are filled, and thus the problematic short-cutting through them to the final goal is prevented. The novel established D+*  also provides safety marginals to the walls and other obstacles, a property that results in paths that do not cut the corners that could potentially disrupt the platform operation. D+*  has also the capability to model the unknown space as risk-free areas that should keep the paths inside, e.g in a tunnel environment, and thus heavily reducing the risk of larger shortcuts through openings in the walls. D+* is also introducing a dynamic map handling capability that continuously updates with the latest information acquired during the map building process, allowing the planner to use constant map growth and resolve cases of planning over outdated sparser map reconstructions. The proposed path planner is also capable to plan 2D and 3D paths by only changing the input map to a 2D or 3D map and it is independent of the dynamics of the robotic platform. The efficiency of the proposed scheme is experimentally evaluated in multiple real-life experiments where D+* is producing successfully proper planned paths, either in 2D in the use case of the Boston dynamics Spot robot or 3D paths in the case of an unmanned areal vehicle in varying and challenging scenarios.

    Download full text (pdf)
    fulltext
  • 48.
    Karvelis, Petros
    et al.
    Computer Technology Institute & Press “Diophantus” Patras, Greece.
    Petsios, Stefanos
    Computer Technology Institute & Press “Diophantus” Patras, Greece.
    Georgoulas, George G.
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Signals and Systems.
    Stylios, Chrysostomos
    Computer Technology Institute & Press “Diophantus” Patras, Greece.
    Short Time Wind Forecasting with Uncertainty2019In: The 10th International Conference on Information, Intelligence, Systems and Applications, 15-17 July 2019, Patras, Greece, IEEE, 2019, p. 511-518Conference paper (Refereed)
    Abstract [en]

    Forecasting the weather and especially the wind is important for a number of applications like wind farms or for maritime operations. Nowadays machine learning techniques are becoming more reliable and robust for forecasting due to the fact that a plethora of available datasets exist. However, forecasts for shorter time horizon less than two hour is not reliable due to the frequent wind fluctuations. Nevertheless, the need for algorithms that can have a small memory and cpu footprint is needed for hardware e.g. microcontrollers that are on board of vessels. In this manuscript a method for short time wind forecasting is proposed and scaled for a microcontroller. The method also computes prediction intervals with a certain probability. Our method was tested using real data recorded from a weather station on board of a ship conducting trips across the Aegean Sea (Greece).

  • 49.
    Khan, Muhammad Saif Ullah
    et al.
    Department of Computer Science, Technical University of Kaiserslautern, 67663 Kaiserslautern, Germany; German Research Institute for Artificial Intelligence (DFKI), 67663 Kaiserslautern, Germany.
    Pagani, Alain
    German Research Institute for Artificial Intelligence (DFKI), 67663 Kaiserslautern, Germany.
    Liwicki, Marcus
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Embedded Internet Systems Lab.
    Stricker, Didier
    Department of Computer Science, Technical University of Kaiserslautern, 67663 Kaiserslautern, Germany; German Research Institute for Artificial Intelligence (DFKI), 67663 Kaiserslautern, Germany.
    Afzal, Muhammad Zeshan
    Department of Computer Science, Technical University of Kaiserslautern, 67663 Kaiserslautern, Germany; German Research Institute for Artificial Intelligence (DFKI), 67663 Kaiserslautern, Germany; Mindgarage, Technical University of Kaiserslautern, 67663 Kaiserslautern, Germany.
    Three-Dimensional Reconstruction from a Single RGB Image Using Deep Learning: A Review2022In: Journal of Imaging, E-ISSN 2313-433X, Vol. 8, no 9, article id 225Article, review/survey (Refereed)
    Abstract [en]

    Performing 3D reconstruction from a single 2D input is a challenging problem that is trending in literature. Until recently, it was an ill-posed optimization problem, but with the advent of learning-based methods, the performance of 3D reconstruction has also significantly improved. Infinitely many different 3D objects can be projected onto the same 2D plane, which makes the reconstruction task very difficult. It is even more difficult for objects with complex deformations or no textures. This paper serves as a review of recent literature on 3D reconstruction from a single view, with a focus on deep learning methods from 2018 to 2021. Due to the lack of standard datasets or 3D shape representation methods, it is hard to compare all reviewed methods directly. However, this paper reviews different approaches for reconstructing 3D shapes as depth maps, surface normals, point clouds, and meshes; along with various loss functions and metrics used to train and evaluate these methods.

  • 50.
    Kim, Wong Yoke
    et al.
    Department of Mechatronics and Biomedical Engineering, Lee Kong Chian Faculty of Engineering and Science, Universiti Tunku Abdul Rahman, Selangor, Malaysia.
    Hum, Yan Chai
    Department of Mechatronics and Biomedical Engineering, Lee Kong Chian Faculty of Engineering and Science, Universiti Tunku Abdul Rahman, Selangor, Malaysia.
    Tee, Yee Kai
    Department of Mechatronics and Biomedical Engineering, Lee Kong Chian Faculty of Engineering and Science, Universiti Tunku Abdul Rahman, Selangor, Malaysia.
    Yap, Wun-She
    Department of Electrical and Electronic Engineering, Lee Kong Chian Faculty of Engineering and Science, Universiti Tunku Abdul Rahman, Selangor, Malaysia.
    Mokayed, Haman
    Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Embedded Internet Systems Lab.
    Lai, Khin Wee
    Department of Biomedical Engineering, Universiti Malaya, 50603, Kuala Lumpur, Malaysia.
    A modified single image dehazing method for autonomous driving vision system2023In: Multimedia tools and applications, ISSN 1380-7501, E-ISSN 1573-7721Article in journal (Refereed)
123 1 - 50 of 110
CiteExportLink to result list
Permanent link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf