【24h】

Texture Classification Via Stationary-Wavelet Based Contourlet Transform

机译:通过基于平稳小波的Contourlet变换进行纹理分类

获取原文
获取原文并翻译 | 示例

摘要

A directional multiresolution approach was proposed for texture analysis and classification based on a modified contourlet transform named the stationary wavelet-based contourlet transform (SWBCT). In the phase for extracting features after the decomposition, energy measures, Hu moments and co-occurrence matrices were calculated respectively. The progressive texture classification algorithm had better performance compared with several other methods using wavelet, stationary wavelet, brushlet, contourlet and Gabor filters. Moreover, in the case that there are only small scale samples for training, our method can also obtain a satisfactory result.
机译:提出了一种基于方向性多分辨率的纹理分析和分类方法,该方法基于改进的轮廓波变换,称为平稳小波基轮廓波变换(SWBCT)。在分解后的特征提取阶段,分别计算能量度量,Hu矩和共现矩阵。与使用小波,固定小波,电刷,轮廓波和Gabor滤波器的其他几种方法相比,渐进纹理分类算法具有更好的性能。此外,在仅用于训练的小规模样本的情况下,我们的方法也可以获得令人满意的结果。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号