结合色彩特征和空域特征的成捆原木轮廓识别①-计算机系统应用.PDF

计 算 机 系 统 应 用 2013 年 第 22 卷 第 7 期 结合色彩特征和空域特征的成捆原木轮廓识别① 景 林, 林耀海, 温永仙, 黄世国, 林宜宽 (福建农林大学 计算机与信息学院, 福州 350002) 摘 要: 成捆原木检尺自动化不仅能够提高生产效率, 而且是林木资源管理的需要. 通过图像处理得原木检尺直 径和统计根数, 这是实现检尺自动化中的重要环节. 由于原木生产作业现场获取的成捆原木图像背景多样, 拍照 时自然光照条件不一, 成捆原木轮廓识别需要较多的人工交互. 通过研究成捆原木端面彩色图像中原木的色彩 特征, 利用它滤除大量的不相关背景, 留下原木色彩近似的像素点; 并结合拉普拉斯滤波器, 获得原木边缘, 实 现成捆原木轮廓的分离, 从而获得原木轮廓. 关键词: 原木端面图像; 色彩特征; 检尺自动化; 拉普拉斯滤波器 Method for Outline Identification of Bundled Logs Based Upon Color and Spatial Features JING Lin, LIN Yao-Hai, WEN Yong-Xian, HUANG Shi-Guo, LIN Yi-Kuan (School of Computer Science and Technology, Fujian Agriculture and Forest University, Fuzhou 350002, China) Abstract: Scaling automation for bundled logs not only improves production efficiency, but also is a requirement for forest resource management. After image processing, diameter of each log and the number of logs is obtained, and this process is key point of scaling automation for bundled logs. Because the imaging environment of bundled logs is complex, much man-machine interaction is needed while using the existing methods for outline identification of bundled logs. In this paper, the color feature of logs in the image of cross section of bundled logs has been discussed, which is used to remove the irrelevant pixels in the image and obtain the pixels of logs. In the next step, Laplace filter is used to work out the edge in the image, implementing the separation of log outlines. Finally, we achieve outline identification of bundled logs. Key words: image of cross section of bundled logs; color feature; scaling automation; Laplace filter 原木捆轮廓识别是指利用数码相机获得原木捆端 对于成捆原木检尺自动化, 一般认为, 其核心是 面彩色图像, 通过图像处理获得木捆中每根原木检尺 从生产环境采集到的图像中获得较好的目标轮廓. 国 直径并统计根数, 这是原木检尺自动化的主要环节.