当前位置: 首页 > 文章 > 基于支持向量机的玉米田间杂草识别方法 农业机械学报 2009,40 (1) 162-166
Position: Home > Articles > Weed Identification Method Based on SVM in the Corn Field Transactions of the Chinese Society for Agricultural Machinery 2009,40 (1) 162-166

基于支持向量机的玉米田间杂草识别方法

作  者:
吴兰兰;刘剑英;文友先;邓晓炎
单  位:
华中农业大学工程技术学院;华中农业大学理学院
关键词:
玉米;杂草识别;支持向量机;预处理;核函数
摘  要:
提出了一种基于图像处理和支持向量机(SVM)技术的玉米和杂草识别方法。首先根据玉米与杂草、土壤彩色图像的特征提出一类图像灰度化方法,并通过对灰度图像的除噪处理有效地分离目标对象。然后从处理好的图像中提取出目标对象的形状特征参数作为输入特征向量,进而提出玉米田间杂草识别的支持向量机方法。试验结果表明了方法的有效性,通过适当选取核函数识别率可达到98.3%。
译  名:
Weed Identification Method Based on SVM in the Corn Field
作  者:
Wu Lanlan1 Liu Jianying1 Wen Youxian1 Deng Xiaoyan2(1.College of Engineering & Technology,Huazhong Agricultural University,Wuhan 430070,China2.College of Basic Sciences,Huazhong Agricultural University,Wuhan 430070,China)
关键词:
Corn,Weed identification,Support vector machine,Preprocessing,Kernel function
摘  要:
This paper proposed a method for corn-weed recognition by using the combination technique of image processing and support vector machine(SVM).A gray processing algorithm was proposed based on the features of corn-weed color images.The object could be separated effectively by denoising the gray image.The shape features of the object were extracted and taken as feature vectors,which could be used to propose the SVM method for the recognition of corn-weed.Comparing the SVM method with the neural-network one,the former is better than the latter one seeing from the experimental results.Experimental results also show that the presented method is effective,and this method gives a recognition rate 98.3% with the properly selected kernel function.

相似文章

计量
文章访问数: 16
HTML全文浏览量: 0
PDF下载量: 0

所属期刊

推荐期刊