Abstract: This study presents a new approach for image retrieval in extracting and integrating the color, texture and shape features. The proposed descriptor converts the RGB color image into HSV color space. HSV color space is used in this approach make use of color, intensity and brightness of the color image. From the Hue (H) and Saturation (S) color features are extracted and from value color space texture features are extracted. To extract the texture features from the value component Local Maximum Edge Binary Patterns are applied (LMEBP). Apply Zernike moments on gray scale image to extract the shape features. To extract the feature vector all the histograms are concatenated three experiments have been carried out in demonstrating the worth of our approach. The presented method is tested on two databases, Corel-10k and MIT-Vistex. The retrieval performance has shown a significant improvement in terms of precision and recall as compared with Center-Symmetric Local Binary Pattern (CS-LBP) Local Edge Pattern for Segmentation (LEPSEG) and Local Edge Pattern for Image Retrieval (LEPINV) and other existing transform techniques in image retrieval system.
G. Sucharitha and Ranjan K. Senapati, 2018. Optimized Image Retrieval Using HSV Color Space, Local Edge Binary Patterns and Zernike Moments. Journal of Engineering and Applied Sciences, 13: 6777-6786.