| 基于HSI颜色模型的杂草与 土壤背景分割方法研究 |
| |
| 引用本文： | 杨转.基于HSI颜色模型的杂草与 土壤背景分割方法研究[J].河北农业大学学报,2011,34(4):124-127. |
| |
| 作者姓名： | 杨转 |
| |
| 作者单位： | 廊坊市广播电视大学,河北廊坊,065000 |
| |
| 摘 要： | 针对杂草识别中如何将杂草与土壤背景分离问题,提出了利用HSI颜色模型中的H分量分割杂草和土壤背景的方法:该方法首先把RGB彩色空间转换到HSI彩色空间,然后根据色度分量(H)确定阈值将灰度图像转化为二值图像,并在此基础上进行适当腐蚀、膨胀,实现了杂草和土壤背景的准确分割.实验结果表明,该方法的准确率达到90%,验证了该...
|
| 关 键 词： | 杂草识别 HSI颜色空间 图像分割 |
Research on the segmentation method for identifying weed from soil background based on HSI color-space |
| |
| YANG Zhuan.Research on the segmentation method for identifying weed from soil background based on HSI color-space[J].Journal of Agricultural University of Hebei,2011,34(4):124-127. |
| |
| Authors: | YANG Zhuan |
| |
| Institution: | YANG Zhuan(Langfang Radio and Television University,Langfang 065000,China) |
| |
| Abstract: | This paper focuses on how to identify weed from soil background in weed identification.A background segmentation method is proposed,which takes HSI as color-space and H as characteristic variant.In this approach,the RGB color-space is firstly transformed into HSI color space.Then according to the H characteristic variant,the threshold is determined for changing gray image into binary image.In the end the weed is identified from soil background by dealing with the binary image through adaptively eroding and dilating.The experiment results show the method can achieve the accuracy rate of 90%,and thereby proving the validity of the method. |
| |
| Keywords: | weed identification HSI color space image segmentation |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
