The Application of Image Segmentation in Object Recognition of Mature Eggplant
Based on the eggplant images acquired in natural light, an auto threshold adaptive algorithm is used for image segmentation, which is conducted with the gray-scale, three orthogonal feature vectors based on R, G, Bs linear transformation and the hue component of HSV color model. The experiment result of Matlab shows that by the segmentation based on Otsu for gray-scale and the optimization for gray-scale histogram, the target and background have a good separation, but the impact of area light reflection exists on the region of eggplant surface; by the segmentation based on improved Otsu for the color feature vector (2G-R-B)/4 can eliminate the impact of light reflection partially; finally, the improved Otsu algorithm with hue for segmentation can effectively remove the impact of light reflection completely and obtain a satisfactory result. And the segmentation effect is dramatically improved by removing the noise of hue-segmented binary image through mathematical morphology.
Image segmentation Auto threshold Image recognition Otsu
Zhengming Li Sen Wang Jun Sun
School of Electrical & Information Engineering,Jiangsu University,Zhenjiang 212013,China
国际会议
北京
英文
1-7
2009-10-14（万方平台首次上网日期，不代表论文的发表时间）