﻿ 基于分水岭算法的木材缺陷边缘检测
 林业科学  2006, Vol. 42 Issue (3): 111-115 PDF
0

#### 文章信息

Wang Guozhu, Zhou Haibin, Li Wenbin, Sa Chao.

Wood Defect Edge Detection Based on Watershed Algorithm

Scientia Silvae Sinicae, 2006, 42(3): 111-115.

### 作者相关文章

1. 北京林业大学  北京 100083;
2. 中国林业科学研究院木材工业研究所  北京 100091

Wood Defect Edge Detection Based on Watershed Algorithm
Wang Guozhu1, Zhou Haibin2, Li Wenbin1, Sa Chao1
1. Beijing Forestry University  Beijing 100083;
2. Research Institute of Wood Industry, CAF  Beijing 100091
Abstract: The paper first describes the watershed algorithm and solves the problem of the over-segmentation from the watershed algorithm by using the mark watershed transform; then wood defect image is treated with algorithm; finally the comparison is made between the original image and the edge image detected. The result showed that the wood defect image could be segmented with the mark-controlled watershed this algorithm and the defect edge image be exactly detected. Moreover, the treatment also provided the convenience for the following treatment such as pattern recognition.
Key words: watershed algorithm    wood defects    edge detection    over_segmentation

1 分水岭算法

 图 1 地形示意图 Fig. 1 The terrain profile

 图 2 闭开滤波后的图像形态学梯度 Fig. 2 The gradient after closing-opening
 图 3 分水岭算法处理结果图 Fig. 3 The result with watershed arithmetic filter

 (1)

 (2)

 图 4 标志集和流域 Fig. 4 The marker set and basins
2 二值数学形态学的基本运算

2.1 二值腐蚀

 (3)

2.2 二值膨胀

 (4)

2.3 二值闭
 (5)

 图 5 闭运算 Fig. 5 The closing operation
2.4 二值开
 (6)

 (7)

 图 6 开运算 Fig. 6 The opening operation
3 缺陷边缘检测

 图 7 板材反相图像 Fig. 7 The reverse image of the lumber
 图 8 对板材反相图像二值化处理 Fig. 8 The binary image
 图 9 对图 6闭开滤波处理 Fig. 9 The closing-opening filtering of the fig. 6
 图 10 内标记 Fig. 10 The inner marker
 图 11 外标记 Fig. 11 The external marker
 图 12 标记点集 Fig. 12 The set of the markers

 图 13 缺陷边缘图像 Fig. 13 The edge image of the defect
 图 14 原图像与边缘图像的合并图像 Fig. 14 The orginal and edge images

 图 15 其他缺陷原图像与边缘图像 Fig. 15 The original and edge image of the living knot and the decay a活节Live node; b腐朽Molder.
4 结论

 陈小梅, 倪国强, 刘明奇. 2001. 基于分水岭算法的红外图像分割方法. 光电子·激光, 12(10): 1072-1075. 崔屹. 2000. 图像处理与分析——数学形态学方法及应用. 北京: 科学出版社. 龚天旭, 彭嘉雄. 2003. 基于分水岭变换的彩色图像分割. 华中科技大学学报, 31(9): 74-76. 马丽红, 张宇, 邓健平. 2003. 基于形态开闭滤波二值标记和纹理特征合并的分水岭算法. 中国图形图像学报, 8(1): 77-83. 王伟凝.2001.基于数学形态学的视网膜血管图像增强.华南理工大学硕士学位论文 Beucher S, Meyer F. 1993. "The morphological approach to segmentation: the watershed transformation" mathematical morphology in image processing. E R Dougherty Ed. New York: Marcel Dekker481. John M. 1999. Image segmentation and analysis via multiscale gradient watershed hierarchies. IEEE Transactions on Image Processing, 8(1): 69-79.