Geometry-based Mass Grading of Betelnut Using Image Processing

This paper presents a system for betelnut shapes grading into four classes according to standards stipulated in China. The system effectively obtains the semi-major axis, semi-minor axis, perimeter, area and other characteristic parameters of betelnut through edge detection, and obtains the grade of each parameter of betelnut based on the subordinate function classification method, and then comprehensively evaluates the level of betelnut according to the weighted algorithm. The results show that the betelnut size detection error is not more than 5%,and the shape classification accuracy is above 90%. The average time to grade one betelnut is below 450ms.