A NON-DESTRUCTIVE TESTING METHOD FOR AGRICULTURAL PRODUCT QUALITY INDICATORS

摘要

#$%^&*AU2020102405A420201105.pdf#####ABSTRACT The present invention discloses a method for non-destructive detection of quality indicators of agricultural products, which comprises the following steps: Step 1: Build a machine vision 5 detection platform, and then collect RGB images of an agricultural product in a vision sensor in all directions; Step 2: Use the images collected in Step 1 to image segmentation, and then divide the region of interest for the agricultural product, and extract the contours of the agricultural product; Step 3: Use the contours of the agricultural products extracted in Step 2, and use the shape invariant feature extraction operator to extract the shape feature vectors to overcome the 0 problems of different sizes, positions and orientations of agricultural products. Step 4: Use the sample images of the agricultural product with different shapes collected in step 1, and use the feature vectors extracted in step 2 and step 3 to train a support vector machine to determine the optimal parameters and detection accuracy of the support vector machine. 7DRAWINGS inspection platormD - acq is t~fiion -- 2-- c lmdel -- Csegmentation Training support veto Shape invariantl feature'! Convex hull Contour machine extraction extraction Forecast the quality of agricultural products FIG.1 40 50 20 FG FIG.2