基于分形维数特征的原木漏节图像的研究

采用X射线法透射木材,根据检测透射物体后射线的强度差异,建立木材X射线的分形模型。采用计算盒子维数的方法对原木漏节X射线图像进行有效的边缘检测,用分形维数D的大小来定量描述原木漏节图像的不规则程度,以确定缺陷所在。研究表明背景部分与边缘部分的分形维数存在一定程度的差别,正常部分分维数值大约在2.007 3左右,漏节边缘分维数值大约在1.400 0~1.900 0之间。

Study of Log with Rotten Knot Image Based on Fractal Dimension Feature

X-ray testing method was used to detect log inner defects nondestructively in this paper.Fractal model of log x-ray image was built according to the difference of intensity after x-ray passing through log.Box-counting method was adopted to detect edges of log images effectively.And the irregular degree of images was described in quantity by using the values of fractal dimension to confirm the defects position.The results showed that there were differences in fractal dimension between the background regions and the edges.The values of fractal dimension in background regions were about 2.007 3,while that of the edges were between 1.400 0 and 1.900 0.