| 基于流域算法的谷物籽粒图像分割技术 |
| |
| 引用本文： | 凌云,王一鸣,孙明,张小超. 基于流域算法的谷物籽粒图像分割技术[J]. 农业机械学报, 2005, 36(3): 95-98 |
| |
| 作者姓名： | 凌云 王一鸣 孙明 张小超 |
| |
| 作者单位： | 中国农业大学信息与电气工程学院,博士生,100083,北京市;中国农业大学信息与电气工程学院,教授,博士生导师;中国农业大学信息与电气工程学院,副教授,博士;中国农业机械化科学研究院,研究员,100083,北京市 |
| |
| 基金项目： | “九五”国家科技攻关计划资助项目 (项目编号 :990 10 0 112 ) |
| |
| 摘 要： | 为解决谷物外观品质检测过程中获取图像的籽粒粘连问题，提出了一种基于先验知识的流域分割算法。首先，设定合理的区域面积阈值Ath，标记并去除图像中的单个籽粒区域，结果记为M；然后，设定有效腐蚀次数Eth，对粘连籽粒区域进行流域分割，结果记为Mc；最后，取Ms与Mc的并集，作为分割结果。试验结果表明，该算法分割效果较好，分割精度达到95．4％。
|
| 关 键 词： | 谷粒 图像分割 流域算法 机器视觉 |
| 修稿时间： | 2003-06-18 |
Application of Watershed Algorithm to Paddy Image Segmentation |
| |
| Ling Yun,Wang Yiming,Sun Ming,Zhang Xiaochao. Application of Watershed Algorithm to Paddy Image Segmentation[J]. Transactions of the Chinese Society for Agricultural Machinery, 2005, 36(3): 95-98 |
| |
| Authors: | Ling Yun Wang Yiming Sun Ming Zhang Xiaochao |
| |
| Affiliation: | 1.China Agricultural University 2.Chinese Academy of Agricultural Me chanization Sciences |
| |
| Abstract: | A new watershed algorithm based on priori knowledge is presented in this paper, which was adapted to paddy imag e segmentation. Firstly, single seed regions are marked and removed from the clu stering paddy image according to single area threshold ( A th ), and the result image is marked M s . Then, the clustering seed regions are segmented by watershed algorithm according to effective erosion degree threshold ( E th ), and the segmented image is marked M c . Finally, the union of M s and M c is taken as the result image. The experimental results showed t hat the method was valid and efficient, and its precision reached 95.4%. |
| |
| Keywords: | Paddy Image segmentation Watershed algorit hm Machine vision |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
