数字图像边缘检测方法的分析与研究55077.pdfVIP

20 10 V l. 20 N . 10 20 10 10 COMPU TER TECHNOLOGY AND DEVELOPM ENT Oct. 20 10 张太发, 高朝阳 ( 黑龙江科技学院数力系, 黑龙江哈尔滨 150027) : , , , , , , , , : ; ; ; ; : T P391. 4 :A : 1673- 629X( 2010) 10- 0141- 05 Analysis and Study of Edge Detection Approach in Digital Images ZHANG Taifa, GAO Cha yang ( Department f Mathematics and Mechanics, Heil ngjiang Inst itute f Science and T echn l gy, Harbin 150027) Abstract:An image is an inf rmati n system, and a great deal f inf rmati n is ffered by edges f the image. Edge is ne f the basic characters f an image, w hich ffers pe ple imp rtant param eters t describe and rec gnize bjects in image pr cessing. Edge detecti n is ne f the m st basic techn l gies f th e digital image pr cessing and analysis. The techn l gy f edge detecti n takes an im p rtant p si ti n in im age pr cessing. T he accuracy f the alg rithm influences the w h le capability f th e devel ped system directly. The basic princi ples f edge detecti n are intr duced, several k inds f edge detecti n meth ds are described, the advantages and disadvantages are ana lyzed, and the experimental results f different edge detecti n are analyzed and c mpared in detail. Finally, suggest that th e ptimal edge detecti n alg rithm can be ch sen largely depending n the nature f the pr blem in practice. Key words:edge detecti n; w avelet transf rm; neural netw rk ; genetic alg rithm ; mathematical m rph l gy 0 , , , [ 2~ 5] , [ 1] , ,