APTHEA Tool: an analysis tool for infrared signatures on combat vehicles
2006 (English)Independent thesis Advanced level (professional degree), 20 credits / 30 HE credits
Student thesis
Abstract [en]
Infrared signatures can be simulated by software or measured with IR sensitive cameras. To enable comparisons between simulations and IR registrations, compare different designs, as well as compare results to specification requirements, this Matlab software has been developed. The APTHEA software calculates the apparent area of a vehicle in variable temperature spans, either from a simulated 3D model or from a real infrared image collected in the field. The program offers the possibility to choose any number of temperature intervals as well as the interval sizes and shows the rsulting area information in table form. For each temperature interval, statistical parameters such as mean temperature, median temperature, distribution function and standard deviation can be presented. The simulation part of the program interacts with 3D models created by a simulation program called RadThermIR, and the information, such as apparent area, is transferred from the models and processed in Matlab. In the infrared image analysis part of APTHEA, the aim is to automatically localize the vehicle and retrieve its perimeter. This is accomplished through a chain of image processing algorithms, which include: edge detection, contrast enhancement, morphological transforms, feature matching and object detection. This report focuses primarily on the latter part of the program, describing the automated vehicle detection algorithm. The performance of the implemented vehicle detection algorithm is rather satisfying, although many improvements are desirable and suggested as further developments.
Place, publisher, year, edition, pages
2006.
Keywords [en]
Technology, Signalbehandling, Bildbehandling, IR-signaturer
Keywords [sv]
Teknik
Identifiers
URN: urn:nbn:se:ltu:diva-45888ISRN: LTU-EX--06/157--SELocal ID: 389a1a76-7a41-43d6-8875-d06c319f5dfeOAI: oai:DiVA.org:ltu-45888DiVA, id: diva2:1019188
Subject / course
Student thesis, at least 30 credits
Educational program
Computer Science and Engineering, master's level
Examiners
Note
Validerat; 20101217 (root)
2016-10-042016-10-04Bibliographically approved