梯度熵改进边缘检测的自适应阈值曲面分割算法-Journalof.PDFVIP

第32 卷第6期 东 北 大 学 学 报 ( 自 然 科 学 版 ) Vol132, No. 6 2 0 1 1 年 6 月 Journal of Northeastern U ni ersity( Natural Science) Jun. 2 0 1 1 魏 颖, 武 林, 贾 同, 林明秀 ( , 110819) : , # , Canny # , , , , # # : ; ; ; : TP 391. 4 1 : A : 1005-3026( 20 11) 06-0781-05 An Adaptive Threshold Surface Segmentation Algorithm Based on Improved Edge Detection Using Gradient Entropy WEI Ying, W U L in, JIA Tong, LIN M ing-x iu ( Key Laboratory of Medical Imaging Calculation of the Ministry of Education, Northeastern Uni ersity , Shenyang 110819, China. Corresponding author: WEI Ying , E-mail: w eiying @ ise. neu. edu. cn) Abstract: To sol e the segmentation problem of the non-uniform illumination and the w ide range distribution of grayscale gradation in target and background, considering the information on edge of an image is insensiti e to the change of light, quoting of gradient entropy to impro e Canny algorithm to detect appropriate edge. Using polynomial cur ed surface fitting of least square method to get threshold surface to segment targets in the une en background, so an adapti e threshold surface segmentation algorithm w ith impro ed edge detection has been proposed. Various non-uniform distribution of grayscale gradation in background images w ere erified, the experiment results indicated that the algorithm could get correct edge information, and get relati ely good segmentation results to the images with non-uniform distribution of grayscale gradation. Key ords: gradient entropy; edge detection; threshold surface;