10 基于双向耦合扩散的保持特征的边缘锐化和图像增强.pdf

31 3 Vol. 31 No. 3 2008 3 CHINESE JOURNA L OF COMPU TERS M ar. 2008 1) , 2) 2) 3) 1) 付树军 阮秋琦 穆成坡 王文洽 1) ( 250100) 2) ( 100044) 3) ( 100081) , . . ( ) , ; . , , , . , . ; ; ; ; ; ; TP391 Feature Preserving Coupled Bidirectional Flow for Edge Sharpening and Image Enhancement 1) , 2) 2) 3) 1) FU Shu-Jun RU AN Q iu-Q i M U Cheng-Po WANG Wen-Qia 1) ( S ch ool of M at hemat ic s and S y stem S ci ence , S hand ong Unive rs i ty , J i nan 250100) 2) ( I nst it ut e of I nf ormat ion Sc ienc e, Be ij ing J iaotong Unive rs i ty , Beij ing 100044) 3) ( Sc hool of A er osp ace S ci ence and T echnol og y , Beij i ng I nst it ut e of T ech nol ogy , Beij i ng 100081) Abstract In the past decade there has been a growing amount of research concerning partial dif- ferential equations in image enhancement. In this paper, a feature preserving coupled bidirectional flow is presented, w here an inverse diffusion w ith a soft edge decision is performed to sharpen ed- ges along the normal directions to the isophote lines ( edges) , w hile a normal diffusion is done to remove noise and artifacts (jaggies) along the tangent directions on the contrary . T he tw o con- verse diffusion forces are split into a coup