Developing a novel approach for content based image retrieval using modified local binary patterns and morphological transform

Digital image retrieval is one of the major concepts in image processing. In this paper, a novel approach is proposed to retrieve digital images from huge databases which using texture analysis techniques to extract discriminant features together with color and shape features. The proposed approach consist three steps. In the first one, shape detection is done based on top-hat transform to detect and crop main object parts of the image, especially complex ones. Second step is included a texture feature representation algorithm which used Color Local Binary Patterns (CLBP) and local variance as discriminant operators. Finally, to retrieve mostly closing matching images to the query, log likelihood ratio is used. In order to, decrease the computational complexity, a novel algorithm is prepared disregarding not similar categories to the query image. It is done using log-likelihood ratio as non-similarity measure and threshold tuning technique. The performance of the proposed approach is evaluated applying on corel and simplicity image sets and it compared by some of other well-known approaches in terms of precision and recall which shows the superiority of the proposed approach. Low noise sensitivity, rotation invariant, shift invariant, gray scale invariant and low computational complexity are some of other advantages.