Apple Grading Method Based on Features Fusion of Size, Shape and Color

摘要

As grading results of apples based on the single feature such as size, shape or color are not accurate, this paper proposes a multi-feature information fusion method based on BP neural network and D-S evidential theory to improve the accuracy of apple grading. Firstly, size, shape and color features are extracted from the processed images of apples. Secondly, apples are classified with each kind of feature by BP network classifier and as independent evidences, the outputs of classifiers are combined to construct the basic probability assignment (BPA). Finally, using D-S fusion rules of evidences to make the decision and achieve the final grading result. The experimental results have shown that the decision information fusion method based on size, shape or color features has good performance on accuracy compared to the single feature-based method in apple grading.