Retrieve Content Images using Color Histogram, LBP and HOG

In this research work, Content-Based Image Retrieval is implemented using color histogram, Localized Binary Pattern, and Histogram calculated from oriented Gradients. The implementation consists of three steps preprocessing, feature extraction, classification. When recovering images based on content, extraction highlighting is an amazing test task. In this article, the highlights of a histogram, a local binary pattern to highlight surface components, and an ordered slope histogram to highlight inclusions in a shape. The machine classifier of the reinforcement vector is used for grouping. Research results show that a combination of each of the three key points is superior to a single element or a combination of two-component recovery methods.