奇异值分解用于图像置乱程度评价研究 - 模式识别国家重点实验室.PDFVIP

160 ， （ ） 计算机工程与应用 2009 45 12 Computer Engineering and Applications ◎ 图形 图像 模式识别◎ 、 、 奇异值分解用于图像置乱程度评价研究 ， ， 1 2 1 2 2 吴成茂 ，田小平 ，谭铁牛 ， ， 1 2 1 2 2 ， ， WU Cheng-mao TIAN Xiao-ping TAN Tie-niu 1.西安邮电学院电子与信息工程系，西安 710121 2. 中国科学院自动化研究所模式识别国家重点实验室，北京 100080 ， ’ ， ’ ， 1.Department of Electronics and Information Engineering Xi an Institute of Post and Telecommunications Xi an 710121 China ， ， ， ， 2.National Key Laboratory of Pattern Recognition Institute of Automation Chinese Academy of Sciences Beijing 100080 China ， ， WU Cheng-mao TIAN Xiao-ping TAN Tie-niu.Studying on singular value decomposition applied in image scrambling ， ， （）： degree evaluation.Computer Engineering and Applications 2009 45 12 160-163. Abstract ：The new image scrambling degree evaluation method based on matrix singular value decomposition is proposed.This paper firstly constructs the absolute value matrix of difference matrix being composed of original matrix and scrambled one.Then the product of gray difference absolute value matrix and its transpose matrix is computed and is decomposed into left orthogonal ， matrix diagonal matrix and right orthogonal matrix.Last the discrete probability distribution is defined by means of diagonal ele－ ， ment of diagonal matrix and the information entropy of its probability distribution is computed and acted a