Statistical model and local binary pattern based texture feature extraction in dual-tree complex wavelet transform domain

Statistical model and local binary pattern based texture feature extraction in dual-tree complex wavelet transform domain
复制标题

双树复小波变换域中基于统计模型和局部二值模式的纹理特征提取

DOI:
10.1007/s11045-017-0474-z
复制
发表时间:
2017-02
影响因子:
2.5
通讯作者:
Guowei Yang
Guowei Yang
中科院分区:
工程技术4区
文献类型:
--
作者:
Peng Yang;Guowei Yang

文献摘要

参考文献

相似文献

提出了一种新的双树复小波变换域特征提取方法。给定一个输入图像,我们通过DTCWT分解得到所有的高通方向子图像和一组不同分辨率的金字塔低通子图像。然后分别利用广义伽玛密度模型和局部二值模式对高通子图像和低通子图像进行特征表征。将这两种特征结合起来进行纹理分类,在Brodatz、Outex和UMD数据集上的实验结果表明,本文提出的方法比现有方法具有更高的分类精度。
This paper presents a new feature extraction method in dual-tree complex wavelet transform domain. Given an input image, we obtain all highpass directional subimages and a set of pyramid lowpass subimages with different resolutions by applying DTCWT decomposition. After that, generalized Gamma densitymodels and local binary pattern are utilized respectively to characterize features of both highpass and lowpass subimages. The two kinds of features are combined for texture classification, and the experimental results on datasets Brodatz, Outex and UMD demonstrate that our proposed method can achieve superior classification accuracy than other state-of-the-art methods.
DOI: 10.1016/j.compeleceng.2011.06.008
发表时间: 2011-09-01
影响因子: 4.3
作者:
Celik, Turgay;Tjahjadi, Tardi
通讯作者: Tjahjadi, Tardi
DOI: 10.1109/tip.2009.2032313
发表时间: 2010
影响因子: 10.6
作者:
R. Kwitt;A. Uhl
通讯作者: R. Kwitt;A. Uhl
DOI: 10.1109/tip.2013.2262285
发表时间: 2013-10-01
影响因子: 10.6
作者:
Krylov, Vladimir A.;Moser, Gabriele;Zerubia, Josiane
通讯作者: Zerubia, Josiane
DOI: 10.1016/j.patrec.2014.04.009
发表时间: 2014-08
期刊: Pattern Recognit. Lett.
影响因子: --
作者:
Mina Koleini;M. Ahmadzadeh;S. Sadri
通讯作者: Mina Koleini;M. Ahmadzadeh;S. Sadri
DOI: 10.1007/s00371-013-0887-0
发表时间: 2014-11
期刊: The Visual Computer
影响因子: --
作者:
Nishant Shrivastava;V. Tyagi
通讯作者: Nishant Shrivastava;V. Tyagi