A Robust Image Matching Using FDCT and Local Textures for Image Retrieval Application

Image retrieval is the important technique in image processing. The paper presents the robust object recognition using texture and directional feature extraction. The system proposes texture descriptors such as fast discrete Curvelet transform based entropy feature and local directional pattern. The category recognition is to classify an object into one of several predefined categories. The Fast discrete Curvelet transform is used here to decompose the image into structural and textural details at different scale and orientation. It represents an object texture and Curvelet edge information from all orientation which is utilized to extract the entropy feature from each textural band. Entropy is a texture feature describes complexity pattern of an object. Second descriptor called Local directional pattern describes local primitives including different types of curves, corners and junctions. LDP computes the edge response values in all eight directions at each pixel position and generates a code from the relative strength magnitude. The performance measures such as precision and recall rate are measured to evaluate the system performance.