Image retrieval and compression are rigorous research field to solve the problem of storage and management of digital image. Several digital compression techniques are available for digital compression such as lossy and lossless compression methods. JPEG, an international organization issues an effective digital image compression standard. CBIR system have been proposed to reduce the storage of digital multimedia collections. Content Based Indexing and Retrieval method provides some advantages in compression of digital image. However, lossy compression technique reduce the visual appearance of images and also the values of real pixels have been altered. There is a filtering effect on pictorial qualities due to data loss. This purpose of this research is find an efficient retrieval and compression technique of image based on color and texture. This experiment investigates the compression effects on image retrieval using color and texture features and presents retrieval results of a Content Based Image Retrieval (CBIR) system. By using some performance metrics the system’s output can be evaluated. Each one is evaluated by different measurement. In this experiment only Two performance metric has been used namely F1 measure and Average Normalized Modified Retrieval Rank (ANMRR). Here four visual descriptors are used and measure performance individually. Two of the color feature and other two is texture features. The system is tested with two different image database color and gray images with visual feature vectors.
ISBN för värdpublikation: 978-3-031-19958-5; 978-3-031-19957-8