Image Retrieval using Fractional Coefficients of Orthogonal Wavelet Transformed Images with Seven Image Transforms
Hemant B. Kekre, Sudeep D. Thepade, Varun K. Banura, B. Tech, Aanchal Bhatia · 2011
25%, 12.5%, 6.25%, 3.125%, 1.5625% ,0.7813%, 0.39%, 0.195%, 0.097%, 0.048%, 0.024%, 0.012% and 0.006% of complete transformed image) are considered as feature vectors. Instead of using all coefficients of transformed images as the feature vector for image retrieval, these fourteen reduced coefficients sets are used, resulting into better performance and lower computations. The proposed CBIR techniques are implemented on a database having 1000 images spread across 11 categories. For each proposed CBIR technique 55 queries (randomly selected 5 images per category) are fired on the database and average precision and recall values are plotted to get precision-recall crossover point. The results have shown the performance improvement (higher precision-recall crossover point) with fractional coefficients compared to complete transform of image at reduced computations resulting in faster retrieval. The wavelet transform generated using Kekre transform for 0.048% reduced coefficient set gives the best performance among the proposed CBIR techniques.