摘要
森林冠层图像分割是采用数字图像处理方法获取森林冠层参数的关键步骤,针对林木冠层鱼眼图像背景复杂、分辨率高的特点,提出一种PSO优化三维Otsu法的森林冠层图像分割算法,将传统三维OTSU进行分解,以减少运算时间和存储空间,并利用PSO算法优化二维Otsu求取阈值过程。实验结果表明:提出的算法能够快速并准确地分割森林冠层鱼眼图像,尤其在树梢细节区域分割效果更好。所提方法可为林木冠层参数测量提供技术支持。
Forest canopy image segmentation is a key step to forest canopy parameters using digital image processing methods.According to the characteristics of forest canopy fisheye images with complex background and high resolution,this paper proposed a PSO-optimized forest canopy image decomposition Otsu method 3D segmentation algorithm.In this method,the 3D OTSU was decomposed,which reduced the computing time and storage space,and the PSO algorithm was adopted to optimize the two-dimensional Otsu to obtain the threshold process.Experimental results showed that the algorithm proposed in this paper can quickly and accurately segment the fisheye images of forest canopy,especially in the details of treetops.The proposed method can provide a technical support for forest canopy parameters measurement.
作者
朱良宽
邵思协
景维鹏
刘亮
ZHU Liang-kuan;SHAO Si-xie;JING Wei-peng;LIU Liang(School of Mechanical and Electrical Engineering,Northeast Forestry University,Harbin 150040,China)
出处
《哈尔滨理工大学学报》
CAS
北大核心
2019年第5期128-133,共6页
Journal of Harbin University of Science and Technology
基金
国家自然科学基金(31370710)
黑龙江省博士后安家启动基金(LBH-Q13007)
关键词
森林冠层图像
图像处理
自动阈值分割
三维Otsu法
粒子群优化
forest canopy image
image processing
automatic threshold segmentation
three dimensional Otsu method
particle swarm optimization