摘要
植物叶片叶脉提取有助于对植物叶片进行分析研究。模糊聚类能够自动将叶脉和叶肉区域区分开,但是受光照不均影响,全局模糊聚类方法容易错分叶脉和叶肉区域。本文提出一种基于局部模糊聚类的植物叶脉提取方法。首先利用最大类间方差法提取叶片区域;然后根据Canny边缘检测算法检测到的叶脉边缘提取包含叶脉的初始叶脉区域;最后利用模糊聚类对初始叶脉区域进一步分类,得到最终的叶脉区域。实验表明本文方法能够较准确的提取叶脉区域,避免光照不均对聚类结果的影响。
The vein extraction from plant leaves contributes to the analysis and study of plant leaves. Although fuzzy clustering can automatically distinguish the leaf vein from the mesophyll area, the global fuzzy clustering method has the tendency to misclassiiy the leaf vein and the mesophyll region under the influence of uneven illumination. In this paper,a method of plant vein extraction based on the local fuzzy clustering is proposed. First, the maximum inter- class variance method is used to extract the leaf region. Then, the initial vein region that contains the vein is extrac- ted according to the edge detection algorithm of canny. Finally, the initial leaf vein region is further classified by the fuzzy clustering to obtain the final leaf vein area. The experimental results show that the proposed method can extract the leaf vein more accurately and avoid the influence of uneven illumination on the clustering results.
作者
杨泽静
张征
郑伯川
YANG Zejing;ZHANG Zheng;ZHENG Bochuan(College of Mathematics and Information,China West Normal University,Nanehong Siehuan 637009,China)
出处
《西华师范大学学报（自然科学版）》
2018年第3期325-330,共6页
Journal of China West Normal University(Natural Sciences)
基金
南充市研发资金项目(17YFZJ0018)
西华师范大学英才基金项目(17YC396)
四川省苗子工程项目(2017017)
关键词
边缘检测
植物叶片
模糊聚类
叶脉提取
edge detection
plant leaves
fuzzy clustering
leaf vein extraction
