부산대학교 도서관

Cherry color can reflect the sweetness and maturity of cherry to a certain extent, which is an important feature of cherry quality evaluation. At present, the grading of cherry color in China is almost manual entirely, which is slow and easy to get errors. In recent years, using machine vision technology to classify and sort agricultural products has gradually become a research hotspot, and achieved good results. In this paper, Convolutional Neural Network (CNN) was used to achieve real-time cherry color grading, and the validity of the method is verified by experiments. The experimental results show that the accuracy of color classification is over 99% and the recognition rate is 25 per second. Compared with the classical color threshold method and other research methods, this method has higher accuracy and applicability.