Journal of Software:2014.25(S2):278-289

(武汉大学 计算机学院, 湖北 武汉 430072)
Efficient Multi-Scale Texture Recognition Algorithm
SUN Jun,HE Fa-Zhi,CHEN Xiao,YUAN Zhi-Yong
(School of Computer Science, Whuhan University, Whuhan 430072, China)
Chart / table
Similar Articles
Article :Browse 1302   Download 1163
Received:May 09, 2014    Revised:August 19, 2014
> 中文摘要: 局部二值模式(local binary patterns,简称LBP)是一种有效的纹理描述算子,具有算法复杂度低、消耗内存、原理清晰的优点.Damper-Shafter的证据理论满足比贝叶斯概率论更弱的条件,具有直接表达"不确定"和"不知道"的能力.提出了一种结合二者优势的纹理识别方法.该方法首先计算图像金字塔,并利用多尺度LBP去测量两个纹理图像之间的相似度;然后,通过融合每个测试样本的信息来组合测试纹理与每个训练样本相似性度量;最后,识别结果由不同纹理之间的最大证据类决定.实验结果表明,该方法对给定的图像数据集1和数据集2分别取得了96.43%和91.67%的正确率,优于最初基于LBP的纹理识别方法.
Abstract:As an effective texture description operator, local binary patterns (LBP) has the advantages of low computation complexity, low memory consumption and clear principle. Damper-Shafter evidence theory satisfies the conditions weaker than Bayesian probability theory and can directly express states of "uncertain" and "don't know". To exploit the advantages of above two concepts, a new texture recognition method is proposed. Firstly, the approach computes image pyramid and uses the distributions of multi-scale LBP to measure the similarity between two texture images. Secondly, the method combines the similarity measurement between the test texture and each training sample to combine the information given by each training sample. Finally, the recognition result is determined by the maximum evidence among different texture classes. Experimental results show that the proposed method achieves a correction rate of 96.43%, and 91.67%, for data set 1 and data set 2, respectively, outperforming the original LBP based texture recognition algorithm.
文章编号:     中图分类号:    文献标志码:
基金项目:国家自然科学基金(61472289);国家重点基础研究发展计划(973)(2011CB707904) 国家自然科学基金(61472289);国家重点基础研究发展计划(973)(2011CB707904)
Foundation items:
Reference text:


SUN Jun,HE Fa-Zhi,CHEN Xiao,YUAN Zhi-Yong.Efficient Multi-Scale Texture Recognition Algorithm.Journal of Software,2014,25(S2):278-289