A Study on Betel Nut Kernel Center Detection Method Based on Computer Vision Technology
In the automated processing of betel nuts,the processes of dehusking and dipping are crucial.To enhance production efficiency,it is imperative to accurately detect the position of the betel nut kernel center,enabling efficient dehusking and dipping.Addressing the issues of low efficiency in contour detection and poor applicability of detection results in the current automated production process of betel nuts,a computer vision-based method for automated detection of betel nut kernel centers is proposed.Betel nut images collected are preprocessed,and the foreground information of the betel nut is obtained through adaptive threshold segmentation.The edge information of the betel nut is then detected,based on which the coordinates of the betel nut kernel center are calculated using algebraic descriptions of the betel nut's elliptical shape.Results indicate that the method can accurately detect the kernel center positions of betel nuts with various shapes under different backgrounds.In the betel nut images with varying orientations,the detection accuracy of the recall factor is 100％,with an average detection time of 0.39 ms,demonstrating excellent detection performance.
computer visionareca nut core center detectioncanny edge detectionmorphological operation
NETL
NSTL
万方数据
