基于机器视觉的脱绒棉种“破碎”特征检测技术

The Detected Technology of Destroyed Characteristic of Cotton seeds Based on Machine Vision

针对新疆地区存在的破碎棉种问题,基于机器视觉技术搭建了破碎棉种检测系统。通过对脱绒棉种图像进行分析和预处理后,提取出脱绒棉种的特征参数—面积、周长、圆形度和长短轴等,并对特征参数的分析研究,确定选取圆形度作为检测破碎棉种的特征。随机选取了100粒新陆早-24#棉种进行试验,检测精度达到了87.5%。

The paper is to achieve the detection to crashed cottonseeds of Xinjiang,built up detecting system of crashed cottonseeds based on the machine vision technology,Distilled the key character parameters of cottonseeds：proportion,perimeter,circularity,major and minor axis after took analysis and pretreatment to the image of cottonseeds.Chose circularity as the detecting character of crashed cottonseeds based on the analysis of these character parameters.Selected 100 cottonseeds of Xinluzao-24 at random for the experiment,the detecting precision had achieved 81%.