Content–Based Image Retrieval Algorithm Based On The Dual-Tree Complex Wavelet Transform: Efficiency Analysis

The following paper presents a new Content-Based Image Retrieval algorithm. In its realization , we applied the Dual-Tree Complex Wavelet Transform for image features extraction and Hausdorff distance for similarity distance computing between the feature vector of the query-image and the image feature vectors stored in the image database. To estimate its efficiency, we performed some experiments in terms of feature extraction time, distance measure time and number of generated wavelet coefficients. The final result shows that the algorithm produces results which satisfy the modern conditions for efficiency of the algorithm including optimum feature extraction rate in shortened data size and short distance measure time necessary for the highest speed of a CBIR system. Key-Words: The Dual-Tree Complex Wavelet Transform, CBIR, feature extraction, feature vector, efficiency, feature extraction time, distance measure time, wavelet coefficients