| 基于小波和分形提取磨粒图像特征参数的研究 |
| |
| 引用本文： | 李国宾,孟歆,关德林,魏海军. 基于小波和分形提取磨粒图像特征参数的研究[J]. 内燃机学报, 2006, 24(5): 476-479 |
| |
| 作者姓名： | 李国宾 孟歆 关德林 魏海军 |
| |
| 作者单位： | 大连海事大学,轮机工程学院,辽宁,大连,116026;大连海事大学,轮机工程学院,辽宁,大连,116026;大连海事大学,轮机工程学院,辽宁,大连,116026;大连海事大学,轮机工程学院,辽宁,大连,116026 |
| |
| 基金项目： | 教育部高等学校博士学科点专项科研基金 |
| |
| 摘 要： | 根据小波变换和分形理论在多尺度分析与自相似本质上的一致性，提出了一种提取磨粒图像特征参数的新方法。给出了磨粒图像的小波变换方法和小波变换近似系数的分形维数计算方法，然后对磨粒图像进行了分析，实现了磨粒图像的特征提取。结果表明：小波变换近似系数的分形维数反映了磨粒图像上磨粒尺度及数量，可以作为磨粒图像的综合特征参数。
|
| 关 键 词： | 磨粒图像 小波变换 分形理论 小波变换系数 分形维数 |
| 文章编号： | 1000-0909（2006）05-0476-04 |
| 收稿时间： | 2006-02-17 |
| 修稿时间： | 2006-05-08 |
Extracting Feature Parameter of Wear Particle Image Based on Wavelet and Fractal Theory |
| |
| LI Guo-bin,MENG Xin,GUAN De-lin,WEI Hai-jun. Extracting Feature Parameter of Wear Particle Image Based on Wavelet and Fractal Theory[J]. Transactions of Csice, 2006, 24(5): 476-479 |
| |
| Authors: | LI Guo-bin MENG Xin GUAN De-lin WEI Hai-jun |
| |
| Affiliation: | College of Marine Engineering, Dalian Maritime University, Dalian 116026, China |
| |
| Abstract: | A new method was developed to extract the feature parameters of the wear particle image according to the consistency in multi-scale analysis and self-similarity in both wavelet transform and fractal theory. The wavelet transforming approach of wear particle image and fractal dimension algorithm of wavelet transform coefficient are given, and the feature of wear particle image is obtained by means of wavelet decomposition and fractal analysis. The study shows that the fractal dimension of wavelet transform approximate coefficient can reflect the scale and the numbers of wear particles in wear particle image analysis, and can be used as the comprehensive feature parameter in wear particle image analysis. |
| |
| Keywords: | Wear particle image Wavelet transform Fractal theory Wavelet transform coefficient Fractal dimension |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
