基于多分辨率的厄米高斯矩的LBP纹理分类

doi:10.3969/j.issn.1006-2475.2016.09.015

计算机与现代化

基于多分辨率的厄米高斯矩的LBP纹理分类

（华南师范大学物理与电信工程学院,广东广州510006)

收稿日期:2016-03-03 出版日期:2016-09-12 发布日期:2016-09-13
作者简介:张晶晶(1988-)，女，河南信阳人，华南师范大学物理与电信工程学院硕士研究生，研究方向:数字图像处理技术。

LBP Texture Classification Based on Multi-resolution of Gaussian-Hermite Moments

(School of hysics & Telecommunication Engineering, South China Normal University, Guangzhou 510006, China)

Received:2016-03-03 Online:2016-09-12 Published:2016-09-13

摘要/Abstract

摘要： 针对LBP（局部二值模式)纹理描述子局限于在单一分辨率下捕获纹理图像的纹理信息的问题，提出一种基于多分辨率的厄米高斯矩的LBP纹理分类方法。首先结合图像纹理的多分辨率特性，采用厄米高斯矩对图像进行多分辨率重构，然后利用LBP纹理描述子对重构图像进行特征提取，最后采用K近邻特征空间距离的分类方法进行纹理分类。选取KTH-TIPS纹理数据库的纹理图像进行测试实验，实验结果表明，与传统LBP纹理分类方法相比，使用多分辨率的厄米高斯矩的LBP纹理分类方法进行纹理分类，可以更加全面地描述图像的纹理信息，使纹理分类准确率更高。

关键词: LBP, 多分辨率, 厄米高斯矩, 图像重构, 纹理分类

Abstract: LBP（Local Binary Pattern)as a texture descriptor to capture images of local texture information in a single resolution, a method based on multi-resolution of the Gaussian-Hermite moments and LBP texture classification is proposed. Firstly, with the multi-resolution characteristic of image texture, the multi-resolution of the Gaussian-Hermite moments is used to reconstruct image. Then LBP is used for the feature extraction of reconstruction image. Finally, the K-nearest neighbor feature space distance method is used to classify image texture. KTH-TIPS texture library images are selected for texture classification test, the experimental results show that compared with the traditional LBP texture classification method, the method based on multi-resolution of the Gaussian-Hermite moments and local binary pattern texture classification is used to texture classification, which make more comprehensive to describe the image texture information to get higher classification accuracy.

Key words: LBP, multi-resolution, Gaussian-Hermite moments, image reconstruction, texture classification

中图分类号:

TN911.73

张晶晶. 基于多分辨率的厄米高斯矩的LBP纹理分类[J]. 计算机与现代化, doi: 10.3969/j.issn.1006-2475.2016.09.015.

ZHANG Jing-jing. LBP Texture Classification Based on Multi-resolution of Gaussian-Hermite Moments[J]. Computer and Modernization, doi: 10.3969/j.issn.1006-2475.2016.09.015.

参考文献

［1］ Haralick R M, Shanmugam K, Dinstein I H. Textural features for image classification［J］. IEEE Transactions on Systems, Man and Cybernetics, 1973,SMC3(6):610-621.
［2］ Randen T, Husoy J H. Filtering for texture classification:A comparative study ［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1999,21(4):291-310.
［3］王惠明,史萍. 图像纹理特征的提取方法［J］. 中国传媒大学学报(自然科学版), 2006,13(1):49-52.
［4］ Reed T R, Dubuf J M H. A review of recent texture segmentation and feature extraction techniques［J］. CVGIP:Image Understanding, 1993,57(3):359-372.
［5］ Ojala T, Pietikinen M, Harwood D. Performance evaluation of texture measures with classification based on Kullback discrimination of distributions［C］// Proceedings of the 12th IAPR International Conference on Pattern Recognition. 1994,1:582-585.
［6］ Ojala T, Pietikinen M, Harwood D. A comparative study of texture measures with classification based on feature distributions［J］. Pattern Recognition, 1996,29(1):51-59.
［7］ Qian Xueming, Hua Xian-sheng, Cheng Ping, et al. PLBP: An effective local binary patterns texture descriptor with pyramid representation［J］. Pattern Recognition, 2011,44(10-11):2502-2515.
［8］ Liao S, Law M W K, Chung A C S. Dominant local binary patterns for texture classification［J］. IEEE Transactions on Image Processing, 2009,18(5):1107-1118.
［9］ Papakostas G A, Koulouriotis D E, Karakasis E G, et al. Moment-based local binary patterns: A novel descriptor for invariant pattern recognition applications［J］. Neurocomputing， 2013,99(1):358-371.
［10］Gerhard O E. Line-of-sight velocity proles in spherical galaxies: Breaking the degeneracy between anisotropy and mass［J］. Monthly Notices of the Royal Astronomical Society, 1993,265(1):213-230.
［11］Yang Bo, Dai Mo. Image analysis by Gaussian-Hermite moments［J］. Signal Process， 2011,91(10):2290-2303.
［12］Ojala T, Pietikinen M, Harwood D. A comparative study of texture measures with classification based on feature distributions［J］. Pattern Recognition, 1996,29(1):51-59.
［13］Ojala T, Pietikinen M, Menp T. Gray scale and rotation invariant texture classification with local binary patterns［C］// Proceedings of the 6th European Conference on Computer Vision. 2000,1842:404-420.
［14］Ojala T, Pietikainen M, Maenpaa T. Multiresolution gray-scale and rotation invariant texture classification with local binary patterns［J］. IEEE Transaction on Pattern Analysis and Machine Intelligence, 2002,24(7):971-987.
［15］Hayman E, Caputo B, Fritz M, et al. On the signicance of real-world conditions for material classication［C］// ECCV. 2004,3024:253-266.

[1]	杨孙哲, 孙爱珍. 基于HSV颜色与LBP纹理特征的水稻氮素营养诊断[J]. 计算机与现代化, 2023, 0(07): 86-92.
[2]	李海燕, 吴大愚, 柳强. 战略战役兵棋多级多分辨率网格模型建模方法[J]. 计算机与现代化, 2023, 0(06): 27-32.
[3]	徐康健, 龚泓宁, 李灿, 佟德宇. 基于SVR的嵌入强度自适应的图像水印算法[J]. 计算机与现代化, 2023, 0(03): 38-42.
[4]	任朝东, 张得礼. 基于RPN网络和改进LBP特征的充电口检测算法[J]. 计算机与现代化, 2021, 0(08): 64-69.
[5]	李睿, 商家赫. 改进的CamShift无人机目标跟踪算法[J]. 计算机与现代化, 2020, 0(11): 83-88.
[6]	牛道鸿1，马晓东1，吴雪冰1，王芳1,2 . 基于改进的Adaboost和LBP危险物品检测算法研究[J]. 计算机与现代化, 2020, 0(02): 122-.
[7]	王刚. 基于角度径向变换的旋转不变纹理分类[J]. 计算机与现代化, 2015, 0(6): 46-50.
[8]	武静. 基于四元数傅里叶梅林变换的旋转不变彩色纹理分类[J]. 计算机与现代化, 2014, 0(5): 104-107,112.
[9]	兰佩，方超. 基于全局与局部特征融合的人脸识别方法[J]. 计算机与现代化, 2014, 0(3): 109-113.
[10]	赵李坤;蒋新华;杨海燕;. 基于MLBPH-FF和SVM的驾驶员疲劳检测[J]. 计算机与现代化, 2013, 1(9): 82-85,9.
[11]	郑碧波;陈伟清;田配云. 基于Radon变换和EMD的旋转不变纹理分类 [J]. 计算机与现代化, 2013, 1(6): 67-70.
[12]	田配云;郑碧波;陈伟清. 基于彩色局部角度二值模式的纹理分类研究[J]. 计算机与现代化, 2013, 1(4): 86-89,9.
[13]	陈海;李科. 基于小波变换的数字水印技术[J]. 计算机与现代化, 2010, 1(9): 107-110.

基于多分辨率的厄米高斯矩的LBP纹理分类

LBP Texture Classification Based on Multi-resolution of Gaussian-Hermite Moments

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 13

编辑推荐

Metrics

本文评价