سال انتشار: ۱۳۹۰

محل انتشار: هفتمین کنفرانس ماشین بینایی و پردازش تصویر

تعداد صفحات: ۵

نویسنده(ها):

Seyed Jalaleddin Mousavi Rad – Department of Computer EngineeringUniversity of KurdistanSanandaj, Iran
Fardin Akhlaghian Tab – Department of Computer EngineeringUniversity of KurdistanSanandaj, Iran
Kaveh Mollazade – Dept. of Mech. Eng. of Agri. MachineryUniversity of KurdistanSanandaj, Iran

چکیده:

In this paper, an algorithm for classifying fivedifferent varieties of rice, using the color and texture features ispresented. The proposed algorithm consists of several steps:image acquisition, segmentation, feature extraction, featureselection, and classification. Sixty color and texture features wereextracted from rice kernels. The Set of features containedredundant, noisy or even irrelevant information so features wereexamined by four different algorithms. Finally twenty-two featureswere selected as the superior ones. A back propagation neuralnetwork-based classifier was developed to classify rice varieties.The overall classification accuracy was achieved as 96.67%.