基于计算机视觉技术的槟榔核中心检测方法研究  

A Study on Betel Nut Kernel Center Detection Method Based on Computer Vision Technology

作　　者：李杰 张乐 LI Jie;ZHANG Le(College of Electronic Information and Electrical Engineering,Shangluo University,Shangluo726000,Shaanxi)

机构地区：[1]商洛学院电子信息与电气工程学院,陕西商洛726000

出　　处：《商洛学院学报》2024年第4期43-49,共7页Journal of Shangluo University

基　　金：商洛学院科研基金项目(22SKY002);陕西省教育厅专项科研计划项目(23JK0417)。

摘　　要：在槟榔的自动化加工工艺中去核与点卤是非常重要的环节,为提高生产效率,需要准确检测出槟榔核中心的位置,才能高效地实现槟榔的去核与点卤。针对目前槟榔自动化生产过程中槟榔内轮廓检测效率不高、检测结果适用性差等问题,提出了一种基于计算机视觉技术的槟榔核中心自动化检测方法。采集到的槟榔图片经过预处理后,采用自适应阈值分割的方法获取槟榔的前景信息并检测槟榔的边缘信息,在此基础上根据槟榔椭圆形状的代数描述计算出槟榔核中心位置坐标。结果表明,该方法可以精确检测出不同背景下多种形貌的槟榔核中心位置,在摆盘的槟榔图片中,槟榔核中心位置的检出率为100%,平均检测时间为0.39 ms,检测效果良好。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.

关 键 词：计算机视觉 槟榔核中心检测 CANNY边缘检测 形态学运算 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]