摘要
针对目前中国苹果分选大部分还是经由人工筛选的现状,提出一种基于机器视觉的苹果缺陷快速检测方法。采用亮度自动校正技术消除苹果表面亮度不均匀分布,根据缺陷候选区域的数量,完成对苹果的初步判断,并使用加权相关向量机进一步对有缺陷的苹果进行判断。通过试验对文中方法的有效性和准确性进行验证。试验结果表明,该检测方法对1 000个测试样本的识别准确率为99.1%,对各种缺陷的检测精度较高。
In view of the current situation that most of apple sorting in China is still manually screened,a rapid detection method is proposed for apple defects based on machine vision.According to the number of defect candidate areas,the preliminary judgment of apples is completed,and the weighted correlation vector machine is used to further judge the defective apples.The effectiveness and accuracy of the proposed method are verified by experiments.The experimental results show that the recognition accuracy of this method for 1000 test samples is 99.1%,and the detection accuracy of various defects is high.
作者
高辉
马国峰
刘伟杰
GAO Hui;MA Guo-feng;LIU Wei-jie(College of Artificial Intelligence,Zhengzhou Railway Vocational&Technical College,Zhengzhou,Henan 451460,China;College of Mechanical and Electrical Engineering,Henan University of Technology,Zhengzhou,Henan 450001,China)
出处
《食品与机械》
北大核心
2020年第10期125-129,148,共6页
Food and Machinery
基金
河南省高等学校重点科研项目计划(编号:20A460008)。
关键词
机器视觉
缺陷快速检测
加权向量机
苹果缺陷
亮度校正
mechanical vision
rapid defect detection
identification and classification
weighted vector machine
apple defects
