摘要
用计算机视觉装置和Matlab软件获取鸡蛋颜色参数(H、I、S),通过试验获得鸡蛋的新鲜度指标(哈夫值),用其作为样本数据建立BP神经网络模型,得到鸡蛋新鲜度与其图像颜色参数之间的最优关系。系统先自动判别鸡蛋壳色再分类检测鸡蛋新鲜度,经检验,建立的BP神经网络具有较好的泛化功能和鲁棒性,对褐壳蛋和白壳蛋新鲜度的正确识别率分别为87.258%、89.029%。
The egg color parameter （H, I, S） was obtained by using the computer vision device and Matlab software, the egg fresh degree（Haft） can be gotten through the test. Using them as the sample data to set up BP neural network, so the optimal relation may be obtained. The system automatically distinguished the egg shell color first, then the egg fresh degree can be dectected. Test results showed that the BP neural network has wonderful performance and robust property, the correct discerning rate to detect egg fresh degree of brown shell egg and white shell egg is 87. 258%, 89. 029% respectively.
出处
《农业机械学报》
EI
CAS
CSCD
北大核心
2006年第1期104-106,共3页
Transactions of the Chinese Society for Agricultural Machinery
基金
国家星火计划资助项目(项目编号:2001EA760026)
湖北省重点科技攻关项目(项目编号:20002P0603)
关键词
鸡蛋
新鲜度
无损检测
计算机视觉
BP神经网络
Egg, Fresh degree, Non-destructive detection, Computer vision, BP neural network
