Content-Based Image Retrieval Performance Using Texture and Colour Features

Efficient searching becomes crucial for large image archive, with more and more digital images accessible on Internet. Consequently, content-based image retrieval (CBIR) has drawn pervasive research attentiveness in the last decade in the field of image processing, pattern recognition and computer vision. CBIR approach boils down to two core problems: feature extraction, followed by feature matching. CBIR is a technique which uses visual contents of image such as colour, shape and texture to search image from large-scale image database similar to the user's query image. Work is carried out on publicly available corel data set images. This paper presents comparative study of CBIR performance in terms of precision and recall measure using features provided by local histogram & global histogram, gray level co-occurrence matrix (GLCM) and colour GLCM (CGLCM). The feature matching procedure is based on three distance measures, namely Canberra distance, Euclidean distance and Manhattan distance. Experimental results show that the CGLCM-feature-based CBIR provides better performance when compared with some of the existing methods.