A Multi-Resolution Content-Based Retrieval Approach for Geographic Images

摘要: Current retrieval methods in geographic image databases use only pixel-by-pixel spectral information. Texture is an important property of geographical images that can improve effectiveness and efficiency. In this paper, we present a content-based approach utilizes the texture features images. Various are extracted using wavelet transforms. Based on features, design hierarchical to cluster for effective efficient retrieval, measuring distances between feature vectors space. Using wavelet-based multi-resolution decomposition, two different sets formulated clustering. For each set, distance measurement techniques designed experimented clustering database. The experimental results demonstrate efficiency when our used.