高斯曲率耦合相关性制约规则的图像匹配算法
|
摘要
目的针对当前较多图像匹配算法在匹配过程中因忽略了特征点之间的相关性而导致算法存在匹配正确度和鲁棒性不佳等不足,设计一种高斯曲率模型耦合相关性制约规则的图像匹配算法。方法首先,利用高斯滤波后图像的一阶矩阵和Hessian矩阵来构造高斯曲率模型,对Hessian算子进行改进,以充分检测图像的特征点。然后,通过求取扇形区域内的Haar小波响应获取特征点的主方向,并根据特征点邻域中像素点的灰度平均值计算特征向量,从而形成特征描述子,完成对特征点的描述。利用特征点集的均值与协方差矩阵来构造相关性模型,对特征点的相关度完成度量,从而定义相关性制约规则,对特征点的相似度进行判断,完成特征点的匹配。最后,利用RANSAC算法提纯匹配特征点,完成图像的匹配。结果仿真实验表明,与当前图像匹配算法相比较,文中算法不仅匹配正确度较高,且具有较强的鲁棒性,在旋转角度为50°时,其正确匹配精度仍可达到87%以上。结论所提算法在多种几何攻击下仍具有较高的匹配精度,在图像处理、信息安全等领域具有良好的参考价值。
The work aims to design an image matching algorithm based on Gauss curvature model and correlation constraint rule with respect to such deficiencies as low matching accuracy and robustness induced by ignoring the correlation between feature points in the matching process of many current image matching algorithms. First, the first-order matrix of the image after the Gauss filter and the Hessian matrix were used to construct the Gauss curvature model, and the Hessian operator was improved to fully detect the feature points of the image. Then, the main direction of the feature points was obtained by obtaining the Haar wavelet response in the sector area, and the eigenvectors were calculated according to the average gray-value of the pixel points in the neighborhood of the feature points, and thus the feature descriptors were formed to complete the description of feature points. The correlation model was constructed based on the mean and covariance matrix of the feature point set and the relevancy of feature points was measured. The similarity ofthe feature points was judged, and the matching of the feature points was completed. RANSAC algorithm was used to purify the matching feature points and complete the matching of images. The simulation results showed that, compared with the current image matching algorithm, the proposed algorithm not only had higher matching accuracy, but also had stronger robustness, whose correct matching accuracy could still reach over 87% when the rotation angle was 50°. The proposed algorithm still has higher matching accuracy under various geometric attacks, which has good reference value in image processing, information security and other fields.
The work aims to design an image matching algorithm based on Gauss curvature model and correlation constraint rule with respect to such deficiencies as low matching accuracy and robustness induced by ignoring the correlation between feature points in the matching process of many current image matching algorithms. First, the first-order matrix of the image after the Gauss filter and the Hessian matrix were used to construct the Gauss curvature model, and the Hessian operator was improved to fully detect the feature points of the image. Then, the main direction of the feature points was obtained by obtaining the Haar wavelet response in the sector area, and the eigenvectors were calculated according to the average gray-value of the pixel points in the neighborhood of the feature points, and thus the feature descriptors were formed to complete the description of feature points. The correlation model was constructed based on the mean and covariance matrix of the feature point set and the relevancy of feature points was measured. The similarity ofthe feature points was judged, and the matching of the feature points was completed. RANSAC algorithm was used to purify the matching feature points and complete the matching of images. The simulation results showed that, compared with the current image matching algorithm, the proposed algorithm not only had higher matching accuracy, but also had stronger robustness, whose correct matching accuracy could still reach over 87% when the rotation angle was 50°. The proposed algorithm still has higher matching accuracy under various geometric attacks, which has good reference value in image processing, information security and other fields.
引文
[1]谭云兰,汤鹏杰,夏洁武.基于自适应引导滤波的全景图像增强算法研究[J].井冈山大学学报(自然科学版),2018,39(4):34-43.TAN Yun-lan,TANG Peng-jie,XIA Jie-wu.Research on Panoramic Image Enhancement Algorithms Based on Adaptive Guided Filtering[J].Journal of Jinggangshan University(Natural Science Edition),2018,39(4):34-43.
[2]黄友文,黄雅兰.一种基于Harris与SIFT算子结合的商标搜索方法[J].电视技术,2014,38(3):35-39.HUANG You-wen,HUANG Ya-lan.A Trademark Search Method Based on Combination of Harris and SIFT Operator[J].Television Technology,2014,38(3):35-39.
[3]金萧.特征区域内抗几何攻击水印算法研究[D].山东:中国石油大学(华东),2015.JIN Xiao.Research on Anti-Geometric Attack Watermarking Algorithm in Feature Area[D].Shandong:China University of Petroleum,2015.
[4]CHEN Yong,SHANG Lei.Improved SIFT Image Registration Algorithm on Characteristic Statistical Distributions and Consistency Constraint[J].Optik-International Journal for Light and Electron Optics,2016,127(2):900-911.
[5]DOU Jian-fang,QIN Qin,TU Zi-mei.Robust Image Matching with Cascaded Outliers Removal[J].Pattern Recognition and Image Analysis,2017,27(3):480-493.
[6]赵双,杨慕升.结合小波变换的Shi-Tomasi算法遮挡图像匹配研究[J].机械设计与制造,2017,11(1):118-121.ZHAO Shuang,YANG Mu-sheng.Research on Occlusion Matching Algorithm Based on Shi-Tomasi Algorithm Combined with Wavelet Transform[J].Mechanical Design and Manufacturing,2017,11(1):118-121.
[7]ZHU Shun-zhi,LIU Li-zhao,CHEN S.Image Feature Detection Algorithm Based on the Spread of Hessian Source[J].Multimedia Systems,2017,23(1):105-117.
[8]CAZOULAT G,OWEN D,MATUSZAK M M.Biomechanical Deformable Image Registration of Longitudinal Lung CT Images Using Vessel Information[J].Phys Med Biol,2016,61(13):4826-4839.
[9]MA Yuan-yuan,HU Xiao-hua,HE Ting-ting.Hessian Regularization Based Symmetric Nonnegative Matrix Factorization for Clustering Gene Expression and Microbiome Data[J].Methods,2016,111(1):80-84.
[10]YANG Jin-zhu,MA Shu-ang,SUN Qi.Improved Hessian Multiscale Enhancement Filter[J].Bio-Medical Materials and Engineering,2014,24(6):3267-3275.
[11]IVAKI M N.Deforming A Hypersurface By Gauss Curvature and Support Function[J].Journal of Functional Analysis,2016,271(8):2133-2165.
[12]FERRAZ L,XAVIER B.A Scale Invariant Interest Point Detector for Discriminative Blob Detection[C]//Proc of the 4th Iberian Conference on Pattern Recognition and Image Analysis,Berlin:Springer,2009:233-240.
[13]QI Bing-jie,LIU Jin-guo,ZHANG Bo-yan.Research on Matching Performance of SIFT and SURF Algorithms for High Resolution Remote Sensing Image[J].ChinesE Optics,2017,10(3):331-339.
[14]王卫兵,白小玲,徐倩.SURF和RANSAC的特征图像匹配[J].哈尔滨理工大学学报,2018,23(1):117-121.WANG Wei-bing,BAI Xiao-ling,XU Qian.Feature Image Matching of SURF and RANSAC[J].Journal of Harbin University of Science and Technology,2018,23(1):117-121.
[15]李兰,谢勤岚.一种改进Harris-SIFT算子的图像匹配算法[J].舰船电子工程,2017,37(4):32-34.LI Lan,XIE Qin-lan.An Image Matching Algorithm Based on Improved Harris-SIFT Operator[J].Ship Electronic Engineering,2017,37(4):32-34.
[16]CLEMENCE D,ANDREA N,ANTJE T.SAR-SIFT for Matching Multiple SAR Images and Radargrammetry[J].PFG-Journal of Photogrammetry[J].Remote Sensing and Geoinformation Science,2017,85(3):149-158.
[17]ZAHRA H N,MEHDI N.An Adaptive Image Registration Method Based on SIFT Features and RANSACTransform[J].Computers&Electrical Engineering,2017,62(1):524-537.
[18]李丽丽,王宝亮,常鹏.SURF算法和改进的RANSAC算法用于移动应用测试界面检测的研究[J].南开大学学报(自然科学版),2017,50(5):8-13.LI Li-li,WANG Bao-liang,CHANG Peng.Research on SURF Algorithm and Improved RANSAC Algorithm for Mobile Application Test Interface Detection[J].Journal of Nankai University(Natural Science Edtion),2017,50(5):8-13.
[19]赵天阳,王见.一种用于光照变化图像匹配的改进KAZE算法[J].计算机测量与控制,2017,25(6):185-188.ZHAO Tian-yang,WANG Jian.An Improved KAZEAlgorithm for Image Matching of Illumination Change[J].Computer Measurement and Control,2017,25(6):185-188.
[20]LI Ao-mei,JIANG Wan-li,YUAN Wei-hua.An Improved FAST Plus SURF Fast Matching Algorithm[J].Advances in Information and Communication Technology,2017,107(1):306-312.
[21]TSAI C H,LIN Y C.An Accelerated Image Matching Technique for UAV Orthoimage Registration[J].ISPRSJournal of Photogrammetry and Remote Sensing,2017,128(1):130-145.
[22]ZHAO Shuang-ming,YU Guo-rong.A New Image Registration Algorithm Using SDTR[J].Neurocomputing,2017,234(c):174-184.
[23]龚卫国,张旋,李正浩.基于改进局部敏感散列算法的图像配准[J].光学精密工程,2011,19(6):1375-1383.GONG Wei-guo,ZHANG Xuan,LI Zheng-hao.Image Registration Based on Improved Local Sensitive Hashing Algorithm[J].Optical Precision Engineering,2011,19(6):1375-1383.
[24]杨新锋,滕书华,夏东.基于空间变换迭代的SIFT特征图像匹配算法[J].红外与激光工程,2013,42(12):3496-3501.YANG Xin-feng,TENG Shu-hua,XIA Dong.SIFTMatching Algorithm with Geometry Constraint[J].Infrared and Laser Engineering,2013,42(12):3496-3501.
[2]黄友文,黄雅兰.一种基于Harris与SIFT算子结合的商标搜索方法[J].电视技术,2014,38(3):35-39.HUANG You-wen,HUANG Ya-lan.A Trademark Search Method Based on Combination of Harris and SIFT Operator[J].Television Technology,2014,38(3):35-39.
[3]金萧.特征区域内抗几何攻击水印算法研究[D].山东:中国石油大学(华东),2015.JIN Xiao.Research on Anti-Geometric Attack Watermarking Algorithm in Feature Area[D].Shandong:China University of Petroleum,2015.
[4]CHEN Yong,SHANG Lei.Improved SIFT Image Registration Algorithm on Characteristic Statistical Distributions and Consistency Constraint[J].Optik-International Journal for Light and Electron Optics,2016,127(2):900-911.
[5]DOU Jian-fang,QIN Qin,TU Zi-mei.Robust Image Matching with Cascaded Outliers Removal[J].Pattern Recognition and Image Analysis,2017,27(3):480-493.
[6]赵双,杨慕升.结合小波变换的Shi-Tomasi算法遮挡图像匹配研究[J].机械设计与制造,2017,11(1):118-121.ZHAO Shuang,YANG Mu-sheng.Research on Occlusion Matching Algorithm Based on Shi-Tomasi Algorithm Combined with Wavelet Transform[J].Mechanical Design and Manufacturing,2017,11(1):118-121.
[7]ZHU Shun-zhi,LIU Li-zhao,CHEN S.Image Feature Detection Algorithm Based on the Spread of Hessian Source[J].Multimedia Systems,2017,23(1):105-117.
[8]CAZOULAT G,OWEN D,MATUSZAK M M.Biomechanical Deformable Image Registration of Longitudinal Lung CT Images Using Vessel Information[J].Phys Med Biol,2016,61(13):4826-4839.
[9]MA Yuan-yuan,HU Xiao-hua,HE Ting-ting.Hessian Regularization Based Symmetric Nonnegative Matrix Factorization for Clustering Gene Expression and Microbiome Data[J].Methods,2016,111(1):80-84.
[10]YANG Jin-zhu,MA Shu-ang,SUN Qi.Improved Hessian Multiscale Enhancement Filter[J].Bio-Medical Materials and Engineering,2014,24(6):3267-3275.
[11]IVAKI M N.Deforming A Hypersurface By Gauss Curvature and Support Function[J].Journal of Functional Analysis,2016,271(8):2133-2165.
[12]FERRAZ L,XAVIER B.A Scale Invariant Interest Point Detector for Discriminative Blob Detection[C]//Proc of the 4th Iberian Conference on Pattern Recognition and Image Analysis,Berlin:Springer,2009:233-240.
[13]QI Bing-jie,LIU Jin-guo,ZHANG Bo-yan.Research on Matching Performance of SIFT and SURF Algorithms for High Resolution Remote Sensing Image[J].ChinesE Optics,2017,10(3):331-339.
[14]王卫兵,白小玲,徐倩.SURF和RANSAC的特征图像匹配[J].哈尔滨理工大学学报,2018,23(1):117-121.WANG Wei-bing,BAI Xiao-ling,XU Qian.Feature Image Matching of SURF and RANSAC[J].Journal of Harbin University of Science and Technology,2018,23(1):117-121.
[15]李兰,谢勤岚.一种改进Harris-SIFT算子的图像匹配算法[J].舰船电子工程,2017,37(4):32-34.LI Lan,XIE Qin-lan.An Image Matching Algorithm Based on Improved Harris-SIFT Operator[J].Ship Electronic Engineering,2017,37(4):32-34.
[16]CLEMENCE D,ANDREA N,ANTJE T.SAR-SIFT for Matching Multiple SAR Images and Radargrammetry[J].PFG-Journal of Photogrammetry[J].Remote Sensing and Geoinformation Science,2017,85(3):149-158.
[17]ZAHRA H N,MEHDI N.An Adaptive Image Registration Method Based on SIFT Features and RANSACTransform[J].Computers&Electrical Engineering,2017,62(1):524-537.
[18]李丽丽,王宝亮,常鹏.SURF算法和改进的RANSAC算法用于移动应用测试界面检测的研究[J].南开大学学报(自然科学版),2017,50(5):8-13.LI Li-li,WANG Bao-liang,CHANG Peng.Research on SURF Algorithm and Improved RANSAC Algorithm for Mobile Application Test Interface Detection[J].Journal of Nankai University(Natural Science Edtion),2017,50(5):8-13.
[19]赵天阳,王见.一种用于光照变化图像匹配的改进KAZE算法[J].计算机测量与控制,2017,25(6):185-188.ZHAO Tian-yang,WANG Jian.An Improved KAZEAlgorithm for Image Matching of Illumination Change[J].Computer Measurement and Control,2017,25(6):185-188.
[20]LI Ao-mei,JIANG Wan-li,YUAN Wei-hua.An Improved FAST Plus SURF Fast Matching Algorithm[J].Advances in Information and Communication Technology,2017,107(1):306-312.
[21]TSAI C H,LIN Y C.An Accelerated Image Matching Technique for UAV Orthoimage Registration[J].ISPRSJournal of Photogrammetry and Remote Sensing,2017,128(1):130-145.
[22]ZHAO Shuang-ming,YU Guo-rong.A New Image Registration Algorithm Using SDTR[J].Neurocomputing,2017,234(c):174-184.
[23]龚卫国,张旋,李正浩.基于改进局部敏感散列算法的图像配准[J].光学精密工程,2011,19(6):1375-1383.GONG Wei-guo,ZHANG Xuan,LI Zheng-hao.Image Registration Based on Improved Local Sensitive Hashing Algorithm[J].Optical Precision Engineering,2011,19(6):1375-1383.
[24]杨新锋,滕书华,夏东.基于空间变换迭代的SIFT特征图像匹配算法[J].红外与激光工程,2013,42(12):3496-3501.YANG Xin-feng,TENG Shu-hua,XIA Dong.SIFTMatching Algorithm with Geometry Constraint[J].Infrared and Laser Engineering,2013,42(12):3496-3501.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |