基于模糊聚类分析的木材缺陷CT图像分割

为了提高木材的利用率,在木材加工之前对木材缺陷CT图像进行分割,将节子和空洞等缺陷分割出来,通过观察缺陷的位置便于工人师傅下锯。利用计算机断层扫描（CT）技术获取木材缺陷图像,将数字图像处理技术与模糊聚类算法相结合,在标准的模糊C均值算法的基础上改进,采用半模糊聚类的分析方法对木材缺陷图像进行分割检测。实验结果表明：基于半模糊聚类的图像检测方法在木材图像检测上取得了较好的效果,缺陷边缘处很平滑,细节保留完整,更多的保留了边缘上的信息。从而证明了半模糊聚类分析法在木材缺陷CT图像处理方面具有可行性。

Wood Defect CT Image Segmentation Based on Fuzzy Cluster Analysis

In view of wood defect detection problem,a semi-fuzzy clustering analytical method for detecting wood defects image segmentation was proposed in the paper based on the analysis of previous studies on image segmentation and fuzzy clustering algorithm. Firstly,the original image of wood was obtained by computer tomography（ CT）. Then,fuzzy clustering analysis was applied to CT image segmentation to obtain the relationship parameters based on edge information and quickly find the center of the initial class. A precise initial membership degree matrix can then be obtained. Simulation results showed that the image detection method based on semifuzzy clustering wood image detection has achieved good results,which proves the feasibility of semi-fuzzy clustering method of image processing in timber and provides meaningful attempt in timber intelligent detection algorithms.