摘要
选取郑单958、辽单565、京科25玉米品种作为研究对象,从玉米籽粒的数字图像中提取与玉米籽粒的颜色、形状、尺寸等有关的6个形态结构参数,利用支持向量机(SVM)算法进行训练识别,同时与BP人工神经网络(NN)方法进行比较,结果表明,SVM算法识别效率较高,达到92.3%。
Corn varieties Zhengdan958,Liaodan565 and Jingke25 were selected as the research objects.Six general characteristics about color,shape and size were picked up from the digital image of corn seed.Support vector machine（SVM） was used to recognize,and the results were compared with neural network（NN） algorithm.The results indicate that the identification rate of SVM strategy is 92.3%.
出处
《农业机械学报》
EI
CAS
CSCD
北大核心
2009年第3期180-183,共4页
Transactions of the Chinese Society for Agricultural Machinery
基金
河北省自然科学基金资助项目(E2005000291)
关键词
玉米
品种
识别
图像处理
神经网络
支持向量机
Corn
Breed
Recognition
Image prosessing
Neural network
Support vector machine
