Skip to Main Content
This paper proposed a novel method, which uses gray level co-occurrence matrix (GLCM) and singular value decomposition (SVD) to extract the features of texture images for image retrieval. In our algorithm, we take the top ten singular values which are based on SVD as the feature vectors, and then uses the Euclidean distance for similarity measure. Compared with traditional GLCM-based feature extraction methods (such as the mean and variance), our new method can improve the retrieval efficiency. Experiments show that the proposed approach is satisfactory and achieves better retrieval performance.
Date of Conference: 29-31 Oct. 2010