一种基于分形的木材细胞图像特征提取方法*

木材纹理图像的分形维数可以代表木材很多的纹理特征，它是木材树种的一项重要数字特征参数。为了提取木材纹理特征，提出采用改进差分盒子法计算木材纹理细胞图像的分形维数值。理论分析和实验结果表明：该方法能够提取木材纹理特征，并且能在一定程度上克服图像比例大小对于特征提取所造成的影响，是木材纹理特征参数提取的一种重要方法。

Feature extraction algorithm of wood cell image based on fractal dimension

Fractal dimension of wood texture image can represent a lot of texture features, and it is an important characteristic parameter of timber species. Therefore, this paper used improved box-counting approach to compute the fractal dimension of wood texture cell image. In terms of the theory analysis and experiments, proved that the method can extract the characteristics of wood texture, and can overcome the influence of the proportion size. It is an important method of the wood texture feature extraction.