Research on detection method of sprouted potato based on SVM and weighted Euclidean distance

Abstract

[Objective] According to the shortcomings of existing detection techniques, this paper proposes a new method based on SVM (Support Vector Machine) and weighted Euclidean distance detection to improve the detection accuracy of sprouted potatoes. [Method] Firstly, the original image of potato was obtained based on industrial camera, gray image and median filtering were used to ensure the image quality of the image, then B and H component training SVM classifiers were extracted in RGB color space and HSV color space respectively. After this, the well-trained SVM classifier was used to segment the potato image and background. Finally, the weighted Euclidean distance and morphology method were adopted to detect and mark the potato germination site. [Results] Under the help of the MatlabR2014a software platform, the weighted Euclidean distance and the traditional Euclidean distance method were employed to test the I and II potato samples. The experimental results reveal that the average recognition rate of the weighted Euclidean distance method is 90.6%, compared with 88.4% of the traditional Euclidean distance which indicates that the recognition rate of the method in this paper is higher, and the detection effect on the germinated potato is better.