摘要
以花生粒为研究对象,通过实验研究探索视觉硬件采集装置,最终提出了一种双目视觉与颜色R分量为特征值的霉变花生粒识别方法。由双目摄像机采集目标图像后输入LabVIEW的软件处理系统,对图像进行预处理、灰度直方图分析、颜色特征值的提取与分析、构建分类器,以及提取待测花生粒图像颜色特征值与分类器匹配,完成识别任务,最终设计出霉变花生粒识别系统。测试结果表明,该研究对霉变花生粒识别的准确率达到了95%,基本达到对霉变花生粒识别检测的标准。
Taking peanut kernels as the research object,this study explored visual hardware acquisition devices through experimental research,and ultimately proposed a method for identifying moldy peanut kernels based on binocular vision and the R component of color as the characteristic value.After the target image was captured by a binocular camera and input into the LabVIEW software processing system,the image undergoes preprocessing,grayscale histogram analysis,extraction and analysis of color characteristic values,classifier construction,and matching the extracted color characteristic values of the peanut kernel image with the classifier to complete the identification task.Finally,a moldy peanut kernel identification system was designed.Test results show that the accuracy rate of this method for identifying moldy peanut kernels reaches 95%,basically meeting the standard for moldy peanut kernel detection.
作者
金庆睿
刘文超
麻荣欣
王雪晴
JIN Qingrui;LIU Wenchao;MA Rongxin;WANG Xueqing(School of Electric and Mechanical Engineering,Pingdingshan University,Pingdingshan 467000,China)
出处
《吉林化工学院学报》
2025年第7期89-94,共6页
Journal of Jilin Institute of Chemical Technology
基金
2023年河南省科技厅项目(23210224008)
平顶山学院青年基金项目(PXY-QNJJ-2018001)。
