Automatic quality inspection: case study mango

Abstract

Post-harvest technology is very important with purpose of increasing the value of the products, reducing losses. Grading, sorting, and classifying of agricultural products is important steps to ensure a profitable and sustainable food industry. Intensive labors are replaced with better devices/machines that can be used in-line and generate sufficiently fast measurements for a high production volume. The image processing and computer vision systems have been widely used for identification, classification, grading and quality evaluation in the agriculture area. Defect identification and defect of mango fruits are challenging tasks, for the computer vision to achieve human levels of recognition. The proposed framework is useful in the supermarkets, processing industry… and can be utilized in computer vision for the automatic sorting of fruits from a set, consisting of different kind of fruits. The thesis proposed to develop an automated tool: an integrated machine vision system that can classify mangos using features including weight, size, and external defects based on digital image analysis. Currently, the thesis has found the equation for calculating the weight of mango. Researchers apply Otsu’s method only for two-dimensional with results 0.78-0.92 indicates a reasonable dependency between 2D and weight. Moreover, to detect the sap burn defect and mechanical damage defect of mango, the image processing technique has been used. However, the program worked at a slow/average rate so it requires for further improvement. Keywords: Cat Chu, Cat Hoa Loc, mango, weighing system, quality classification.