增强旋转不变LBP算法及其在图像检索中的应用

doi:10.11896/j.issn.1002-137X.2019.07.040

摘要/Abstract

摘要： 基于内容的图像检索是目前图像检索领域的研究重点。LBP纹理特征是基于内容的图像检索领域常用的特征。传统的LBP算法应用于图像检索系统时检索效率低,且不具有旋转不变性。旋转不变LBP(rotation invariant LBP,LBPri)算法虽然具备旋转不变性,但检索精度不高。为了提高基于内容的图像检索的精度和效率,在传统LBP算法的基础上提出一种增强旋转不变LBP描述符(Enhanced rotation invariant LBP,ELBPri)。ELBPri描述符首先对原始图像提取Harris角点,以角点为中心采样;其次用旋转不变LBP算法的编码方式对采样后的图像编码;然后统计各图像的LBP直方图;最后计算各图像LBP直方图之间的欧氏距离,并根据相似性排序。实验结果表明,相比LBPri描述符,CBIR系统应用ELBPri描述符检索一般纹理图像集时的平均查准率提高了5.64%,平均检索用时缩短了0.4ms;检索旋转纹理图像集时的平均查准率提高了5.94%,平均检索用时缩短了0.12ms。

关键词: ELBPri描述符, Harris算法, LBP伪灰度图, 基于内容的图像检索, 欧氏距离

Abstract: CBIR (Content-based image retrieval) is a hot topic in image retrieval.LBP texture features are commonly used in CBIR.When the classic LBP algorithm is applied to the image retrieval system,the retrieval efficiency is low,and it does not have the characteristics of rotation invariance.Although the rotation invariant LBP (LBPri) algorithm has the characteristics of rotation invariance,its retrieval efficiency is low.In order to improve the precision and efficiency of CBIR,based on the classical LBP algorithm,this paper proposed an enhanced rotation invariant LBP descriptor (ELBPri).Firstly,the ELBPri descriptor extracts the Harris corners from the original grayscale,and then samples the original grayscale in the center of the Harris corners.Secondly,ELBPri descriptor encodes the sampled image in rotation invariant LBP.Thirdly,the LBP histograms of each image are counted.Finally,ELBPri descriptor calculates the Eucli-dean distance between the LBP histograms of the images and sort them according to similarity.Experimental results show that compared with LBPri descriptor,the average precision of the ELBPri descriptor used in the retrieval of gene-ral texture image sets by the CBIR system is increased by 5.64%,and the average query time is shortened by 0.4ms.The average precision is increased by 5.94% when retrieving rotation texture image sets,and the average query time is shortened by 0.12ms.

Key words: Content-based image retrieval, ELBPri descriptor, Euclidean distance, Harris algorithm, LBP pseudo grayscale

中图分类号:

TP391

孙伟, 赵玉普. 增强旋转不变LBP算法及其在图像检索中的应用[J]. 计算机科学, 2019, 46(7): 263-267. https://doi.org/10.11896/j.issn.1002-137X.2019.07.040

SUN Wei, ZHAO Yu-pu. Enhanced Rotation Invariant LBP Algorithm and Its Application in Image Retrieval[J]. Computer Science, 2019, 46(7): 263-267. https://doi.org/10.11896/j.issn.1002-137X.2019.07.040

参考文献

[1]XIA Z,XIONG N N,VASILAKOS A V,et al.EPCBIR:An efficient and privacy-preserving content-based image retrieval scheme in cloud computing[J].Information Sciences,2017,387:195-204.
[2]ZHANG Y K,LI Y F,SUN J G.Image Retrieval Algorithm Based on Improved Color Coherence Vectors and Contribution to Clustering[J].Computer Science,2015,42(2):243-245.(in Chinese)
张永库,李云峰,孙劲光.基于改进颜色聚合向量与贡献度聚类的图像检索算法[J].计算机科学,2015,42(2):311-315.
[3]KUSHWAHA A K,SRIVASTAVA S,SRIVASTAVA R. Multi-view human activity recognition based on silhouette and uniform rotation invariant local binary patterns[M].New York:Springer-Verlag,2017.
[4]WU H,WU W,ZHANG J,et al.Research on image retrieval algorithm based on LBP and LSH[C]∥Green Energy and Sustainable Development I:Proceedings of the International Confe-rence on Green Energy and Sustainable Development.2017:020038.
[5]LIU L,LAO S,FIEGUTH P,et al.Median Robust Extended Local Binary Pattern for Texture Classification[J].IEEE Tran-sactions on Image Processing,2016,25(3):1368-1381.
[6]XU X,LI S,LIU L.Face recognition based on multi-level histogram sequence center-symmetric local binary pattern and fisherface[C]∥Advanced Information Technology,Electronic and Automation Control Conference.IEEE,2017:448-451.
[7]QIN C,CHEN X,YE D,et al.A novel image hashing scheme with perceptual robustness using block truncation coding[J].Information Sciences,2016,361-362:84-99.
[8]SUN J D,WU X S.Content-Based Image Retrieval Based on Texture Spectrum Descriptors [J].Journal of Computer-Aided Design and Computer Graphics,2010,22(3):516-520.(in Chinese)
孙君顶,毋小省.纹理谱描述符及其在图像检索中的应用[J].计算机辅助设计与图形学学报,2010,22(3):516-520.
[9]AHONEN T,MATAS J,HE C,et al.Rotation Invariant Image Description with Local Binary Pattern Histogram Fourier Features[C]∥Image Analysis,Scandinavian Conference,Scia 2009.Oslo,Norway,2009:61-70.
[10]KUMAR T G S,NAGARAJAN V.Combining LBP and con- tourlet features for image retrieval[C]∥International Confe-rence on Communication and Signal Processing.IEEE,2016:1193-1196.
[11]JAIN S,ZAVERI T,PATEL S.Weighted fusion of LBP and CCH features for effective content based image retrieval[C]∥International Conference on Signal Processing and Communications.IEEE,2016:1-5.
[12]WU Y H,XU G,JIANG J J,et al.Research on Workpiece Ima ge Feature Recognition Based on LBP and SVM[J].Journal of Chongqing University of Technology (Natural Science),2016,30(1):77-84.(in Chinese)
吴益红,许钢,江娟娟,等.基于LBP和SVM的工件图像特征识别研究[J].重庆理工大学学报(自然科学),2016,30(1):77-84.
[13]SUN L,WANG S,XING J,et al.Self-adaption harris corner de- tection algorithm based on image contrast area[C]∥Control and Decision Conference.IEEE,2015:2287-2290.
[14]JIAN C,KAN L I,GAO C X,et al.Application of Local Features in Aerial Image Mosaic[J].Journal of University of Electronic Science & Technology of China,2013,42(1):125-129.
[15]FAKHFAKH S,TMAR M,MAHDI W.Image Retrieval Based on Using Hamming Distance[J].Procedia Computer Science,2015,73:320-327.
[16]XU S P,LIU X P,LI C Q,et al.Texture Descriptor Based on Spatial Statistical Features of Local Binary Pattern Code Pair[J].Pattern Recognition and Artificial Intelligence,2013,26(8):769-776.(in Chinese)
徐少平,刘小平,李春泉,等.基于LBP值对空间统计特征的纹理描述符[J].模式识别与人工智能,2013,26(8):769-776.
[17]YU H Y,HUANG Y X.An Image Retrieval Algorithm Based on SURF for Embedded System[C]∥International Conference on Intelligent Computation Technology and Automation.2017:86-88.

相关文章 15

[1]	杨文坤, 原晓佩, 陈小锋, 郭睿. 三维激光雷达点云空间多特征分割 Spatial Multi-feature Segmentation of 3D Lidar Point Cloud 计算机科学, 2022, 49(8): 143-149. https://doi.org/10.11896/jsjkx.210300275
[2]	王春静, 刘丽, 谭艳艳, 张化祥. 基于模糊颜色特征和模糊相似度的图像检索方法 Image Retrieval Method Based on Fuzzy Color Features and Fuzzy Smiliarity 计算机科学, 2021, 48(8): 191-199. https://doi.org/10.11896/jsjkx.200800202
[3]	张蔷, 黄樟灿, 谈庆, 李华峰, 湛航. 基于动态近邻套索算子的金字塔演化策略 Pyramid Evolution Strategy Based on Dynamic Neighbor Lasso 计算机科学, 2021, 48(6): 215-221. https://doi.org/10.11896/jsjkx.200400115
[4]	王淑云, 干宗良, 刘峰. 基于层级聚类回归模型的人脸超分辨率重建算法 Face Hallucination Reconstruction Algorithm Based on Hierarchical Clustering Regression Model 计算机科学, 2019, 46(8): 298-302. https://doi.org/10.11896/j.issn.1002-137X.2019.08.049
[5]	吴丹丹, 吕鑫. 分布式结构下基于用户协作的匿名区域构建算法 Location Anonymous Algorithm Based on User Collaboration under Distributed Structure 计算机科学, 2019, 46(4): 158-163. https://doi.org/10.11896/j.issn.1002-137X.2019.04.025
[6]	陈静杰, 车洁. 基于IK-medoids算法的飞机油耗聚类方法 IK-medoids Based Aircraft Fuel Consumption Clustering Algorithm 计算机科学, 2018, 45(8): 306-309. https://doi.org/10.11896/j.issn.1002-137X.2018.08.055
[7]	陈静杰,车洁. 基于标准欧氏距离的燃油流量缺失数据填补算法 Fuel Flow Missing-value Imputation Method Based on Standardized Euclidean Distance 计算机科学, 2017, 44(Z6): 109-111. https://doi.org/10.11896/j.issn.1002-137X.2017.6A.023
[8]	沈鹍霄,兰义华,卢玉领,尚耐丽,马晓普. 乳腺辅助诊断系统中可疑肿块分割方法研究 Research on Segmentation Methods in Breast Computer-aided Detection 计算机科学, 2015, 42(Z11): 195-198.
[9]	周剑云,王丽珍,杨增芳. 基于加权欧氏距离的空间Co-location模式挖掘算法研究 Algorithm of Mining Spatial Co-location Patterns Based on Weighted Euclidean Distance 计算机科学, 2014, 41(Z6): 425-428.
[10]	陈刚,岳晓冬,陈宇飞. 基于稀疏低秩描述的图像检索方法 Image Retrieval Method Based on Sparse Low-rank Representation 计算机科学, 2014, 41(3): 302-305.
[11]	邹月娴,霍佳森,刘吉,李奕,邓文军. 基于特征融合的WCE图像检索 WCE Image Retrieval by Using Fused Feature 计算机科学, 2013, 40(Z11): 320-324.
[12]	杨族桥,陈跃鹏,张青. 综合视觉注意模型的显著性局部特征提取算法研究 Salient Local Feature Extraction Algorithm Based on Integrated Visual Attention Model 计算机科学, 2013, 40(8): 289-292.
[13]	郑丽萍,李俊青,于承敏,张民. 基于像素层标识点的双边滤波快速实现 Fast Implementation of Bilateral Filtering with Identification Point in Pixel Layer 计算机科学, 2013, 40(7): 262-265.
[14]	石跃祥,文华,龚平,莫浩澜,金银国. 基于模糊神经网络的语义映射方法及其在自然图像检索中的应用 Projection of Semantics and Retrieval in Natural Scenery Images Based on Fuzzy Nerve Network 计算机科学, 2013, 40(12): 122-126.
[15]	郭士会，杨明，王晓芳，赵东方，邓世涛. 基于FSRM的相关反馈图像检索算法 Relevance Feedback Algorithm Based on Fuzzy Semantic Relevance Matrix in Image Retrieval 计算机科学, 2012, 39(Z6): 540-542.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed