سال انتشار: ۱۳۹۰
محل انتشار: هفتمین کنفرانس ماشین بینایی و پردازش تصویر
تعداد صفحات: ۵
Ghobad Moradi – Islamic Azad University, Ravansar Branch, Kermanshah, Iran
Mousa Shamsi – Faculty of Electrical Engineering Sahand University of Technology Tabriz, Iran
Mohammad Hossein Sedaaghi – Faculty of Electrical Engineering Sahand University of Technology Tabriz, Iran
Setareh Moradi – Member of Young Researcher Club, Branch of Kermanshah Azad University, Iran
Image segmentation is one of the important andcomplicated processes among image processing and computervision algorithm. Its purpose is to partition an input image intodisjoint parts. In this article an important application of imageprocessing in determination of apple quality is studied, and anautomatic algorithm is proposed in order to determine applesskin color defects. First, this image is converted from RGB tocolor space L*a*b*. Then fruit shape is extracted by ACMalgorithm. Finally, the image has segmented using SHFCMalgorithm. Experimental results on the acquired images showthat both FCM and SHFCM spend the same iterations toaccomplish the segmentation process and get the same results.However, the proposed SHFCM algorithm consumes less timethan the standard FCM algorithm. Accuracy of the proposedalgorithm on the acquired images is 91% and 96% for healthypixels and defected ones, respectively.