基于多特征的纹理特征提取方法研究与应用.pdfVIP

GLCM LBP DWT Corel GLCM LBP DWT GLCM LBP DWT T P A doi issn j Research and Application of Texture Feature Extraction Based on Multifeatures MEI Langqi GUO Jianming LIU Qing School of A utom ation W uhan Univ e rs ity of Technolog y W uhan China Abstract T ex ture is a significant visual feature w hich is commonly used to identify and distinguish the image T his paper summarized and analyzed the current common method of tex ture feature ex traction including Gray Level Co occurrence M atrix GLCM Local Binary Pattern LBP and Discrete w avelet transform DWT With the w eight of configuration parameters a new tex ture ex traction method of multif eatures is proposed and implemented w hich com bines the three basic methods mentioned T he image tex ture description ability of different methods is compared through the tests on the image retrieval sy stem T he results show that the average precision of tex ture feature ex traction method based on multif eature combination increased comparing to GLCM algorithm increased comparing to LBP algo rithm increased comparing to DWT algorithm and increased comparing to Xu s tex ture f eature ex traction method w hen images w ere retrieved f rom Corel library T he new tex ture feature ex traction method proposed in this paper combined the advantages of each method and had good rotation invariance and scale invariance How ever it is necessary to ex tract the tex ture features of GLCM LBP and DWT at the same time the time required for the new tex ture feature ex tr