基于图像表面积的分形布朗运动分数维算法研究

基金项目: 国家973计划资助项目(2002CCA0470)

详细信息

Arithmetic Research of Fractal Dimension with Image Face Based on Fractional Brownian Motion

¹ School of Remote Sensing and Information Engineering, Wuhan University, 129 Luoyu Road, Wuhan 430079, China,² Chinese Academy of Surveying and Mapping, 16 Beitaiping Road, Beijing 100039, China

Funds: 国家973计划资助项目(2002CCA0470)

摘要: 针对数字图像的特点,提出了一种计算图像表面积的简单方法。在此基础上,根据分形布朗运动理论,提出基于图像表面积的分数维计算方法。用遥感图像上几种典型地物的图像窗对该方法进行了实验,并与其他几种常用的分数维计算方法进行了比较。实验表明,本文提出的方法与理论预期相符合的程度高于其他分数维计算方法。
- 分形布朗运动 /
- 纹理特征 /
- 遥感分类
Abstract: If every pixel in an image is regarded as column and the image can be a columniation, then the image face can be calculated by a simple, accurate method. On the basis of this method, a new arithmetic for calculating fractal dimension is proposed with the theory of FBM. Experiment results show that the arithmetic is better than several current arithmetics of fractal dimension.
- fractional Brownian motion /
- texture feature /
- classification of remote sensing