Research on Intelligent Grading System of Tremella Fuciformis Based on Machine Vision

Abstract

Highlights The experimental results showed that the grading accuracy of the intelligent grading system of Tremella fuciformis designed in the study was 97.07%. The grading speed designed in the study was five times that of manual grading. The grading system designed in the study can also be extended to classification fields of other edible fungi, fruits, and vegetables. It was proved by experiments that the system involved has high work efficiency and market value. Abstract. An intelligent grading system of dried Tremella fuciformis was designed to solve the problem of low intelligence in the course of production and processing of Tremella fuciformis in this study. The overall structure and working process of the grading system were described, and different grading standards were set according to the color, shape, size, and integrity of dried Tremella fuciformis. The image acquisition and image preprocessing of dried Tremella fuciformis were completed. The RGB model was used for the color feature extraction and recognition of dried Tremella fuciformis, and the diameter and integrity of dried Tremella fuciformis were judged by edge detection. A set of intelligent grading system of dried Tremella fuciformis was developed based on Microsoft Visual Studio 2017 platform. The experimental results showed that the grading accuracy of the grading system designed in the study was 97.07%, and the average grading speed was five times that of manual grading. It was better in reliability, speed, work efficiency, and robustness than the traditional manual grading. Keywords: Intelligent classification, Image recognition, Machine vision, Tremella fuciformis.