Identification of Rice Varieties Using Co-occurrence Matrix Features of Bulk Samples and Support Vector Machine
کلمات کلیدی:
texture; co-occurrence matrix; rice classification;support vector machine
چکیده:
In this paper, an algorithm for classifying fivedifferent varieties of rice bulk samples is presented. Forty fourfeatures from co-occurrence matrix were extracted from bulksamples of rice. The Set of features contained redundant, noisy oreven irrelevant information. So, features were evaluated bystandard sequential forward algorithm. Finally, ten features wereselected as the superior ones. Support vector machine classifierwas developed to classify rice varieties. The overall classificationaccuracy was achieved as 93. 55%.