Block Truncation Coding Method to Achieve Image Information Identification
Lixin Wang · Xinxi wangluo anquan · 2012
Image clustering is the key technology for identification and searching of image information. The retrieval efficiency of image is critically important in pictorial information system. If we use low-level visual features for efficient image clustering in the content-based image identification, we will greatly improve the accuracy of image retrieval and identification. The Color Moment method and the Block Truncation Coding (BTC) method are used to extract color features respectively for clustering features using K-Means. The efficiencies of these two methods for the identification of image information are validated by experiment. The experimental result shows that the BTC (Block Truncated Coding) method to extract the image features for image clustering in order to identify image with higher efficiency.