摘要
大米垩白是检测大米质量等级的重要指标,在利用机器视觉技术进行垩白粒率的检测中,采用能够利用阈值法获取垩白米粒的DH-HV 1303UC摄像头采集的大米灰度图像,进行垩白检测算法的研究,选用双阈值法对大米图像进行分割,分别得到大米籽粒图像和只含垩白米粒的大米图像.在检测中,针对大米籽粒间存在连接的情况,提出了一种基于形态学中腐蚀、膨胀算法的垩白米粒检测方法.试验结果表明,在垩白米粒含量不同籽粒间连接程度不同的情况下,此检测算法均可达到较高的检测精度.
Chalky rice is an important criterion in the detection of rice quality.In the rice chalkiness detecting algorithm,firstly,the DH-HV1303UC camera is used to obtain the gray-scale image of rice,which can extract chalky rice by the method of double-threshold.Touching rice grain is an important problem in chalky rice detecting based on machine vision.A novel method for chalky rice detection with touching rice grain based on erosion and dilation process is proposed.The experimental results show that in the state of different rate of chalky rice and the connection between the rice grain,the novel method can achieve higher detection accuracy.
出处
《广西工学院学报》
CAS
2010年第2期36-40,共5页
Journal of Guangxi University of Technology
基金
广西教育厅科研项目经费(200808X326)资助
关键词
大米
垩白粒率
灰度图像
检测
籽粒连接
rice
chalky rice rate
gray-scale image
detection
touching rice grain
