Effective image retrieval using polar cosine transform and local binary patterns

Abstract

In this paper, we present a novel, effective, and efficient approach to image retrieval. Basically, it is a fusion of both global and local features of images, which achieves significantly higher retrieval competency. Initially, the global features of images are determined using polar cosine transforms (PCTs). For local features, we use rotation invariant local binary patterns (RLBP) rather than using the existing ones, which help in improving the retrieval rate and are in alignment with the rotation invariant property of PCTs. The combination of both acquired global and local features is performed by mapping their features into a common domain. Finally, the proposed hybrid approach provides a robust feature set for image retrieval. Detailed experiments are performed on various sorts of image databases. The results of extensive set of experiments reveal the supremacy of the proposed approach over other approaches in terms of efficiency and retrieval results.