具有光照鲁棒的图像匹配方法
|
摘要
针对现有的基于局部特征的图像匹配算法对光照变化敏感、匹配正确率低等问题,提出一种具有光照鲁棒性的图像匹配算法。首先使用实时对比保留去色(RTCP)算法灰度化图像,然后利用对比拉伸函数模拟不同光照变换对图像的影响从而提取抗光照变换特征点,最后采用局部强度顺序模式建立特征点描述符,根据待匹配图像局部特征点描述符的欧氏距离判断是否为成对匹配点。在公开数据集上,所提算法与尺度不变特征变换(SIFT)算法、加速鲁棒特征(SURF)算法、"风"(KAZE)算法和ORB算法在匹配速度和匹配正确率上进行了对比实验。实验结果表明:随着图像亮度差异的增加,SIFT算法、SURF算法、"风"(KAZE)算法和ORB算法匹配正确率下降迅速,所提算法下降缓慢并且正确率均高于80%;所提算法特征点检测较慢和描述符维数较高,平均耗时为23. 47 s,匹配速度不及另外四种算法,但匹配质量却远超过它们。对实时性要求不高的系统中,所提算法可以克服光照变化对图像匹配造成的影响。
Focusing on the problem that current image matching algorithm based on local feature has low correct rate of illumination change sensitive matching, an image matching algorithm with illumination robustness was proposed. Firstly, a Real-Time Contrast Preserving decolorization( RTCP) algorithm was used for grayscale image, and then a contrast stretching function was used to simulate the influence of different illumination transformation on image to extract feature points of antiillumination transformation. Finally, a feature point descriptor was established by using local intensity order pattern.According to the Euclidean distance of local feature point descriptor of image to be matched, the Euclidean distance was determined to be a pair matching point. In open dataset, the proposed algorithm was compared with Scale Invariant Feature Transform( SIFT) algorithm, Speeded Up Robust Feature( SURF) algorithm, the wind ( KAZE) algorithm and ORB( Oriented FAST and Rotated, BRIEF) algorithm in matching speed and accuracy. The experimental results show that with the increase of image brightness difference, SIFT algorithm, SURF algorithm, the wind algorithm and ORB algorithm reduce matching accuracy rapidly, and the proposed algorithm decreases matching accuracy slowly and the accuracy is higher than80%. The proposed algorithm is slower to detect feature points and has a higher descriptor dimension, with an average time of23. 47 s. The matching speed is not as fast as the other four algorithms, but the matching quality is much better than them.The proposed algorithm can overcome the influence of illumination change on image matching.
Focusing on the problem that current image matching algorithm based on local feature has low correct rate of illumination change sensitive matching, an image matching algorithm with illumination robustness was proposed. Firstly, a Real-Time Contrast Preserving decolorization( RTCP) algorithm was used for grayscale image, and then a contrast stretching function was used to simulate the influence of different illumination transformation on image to extract feature points of antiillumination transformation. Finally, a feature point descriptor was established by using local intensity order pattern.According to the Euclidean distance of local feature point descriptor of image to be matched, the Euclidean distance was determined to be a pair matching point. In open dataset, the proposed algorithm was compared with Scale Invariant Feature Transform( SIFT) algorithm, Speeded Up Robust Feature( SURF) algorithm, the wind ( KAZE) algorithm and ORB( Oriented FAST and Rotated, BRIEF) algorithm in matching speed and accuracy. The experimental results show that with the increase of image brightness difference, SIFT algorithm, SURF algorithm, the wind algorithm and ORB algorithm reduce matching accuracy rapidly, and the proposed algorithm decreases matching accuracy slowly and the accuracy is higher than80%. The proposed algorithm is slower to detect feature points and has a higher descriptor dimension, with an average time of23. 47 s. The matching speed is not as fast as the other four algorithms, but the matching quality is much better than them.The proposed algorithm can overcome the influence of illumination change on image matching.
引文
[1]LOWE D G. Distinctive image features from scale-invariant keypoints[J]. International Journal of Computer Vision, 2004, 60(2):91-110.
[2]BAY H, ESS A, TUYTELAARS T, et al. SURF:speeded up robust features[J]. Computer Vision and Image Understanding,2008, 110(3):346-359.
[3]MOREL J M, YU G. ASIFT:a new framework for fully affine invariant image comparison[J]. SIAM Journal on Imaging Sciences,2009, 2(2):438-469.
[4]ANCUTI C O, ANCUTI C, BEKAERT P. Decolorizing images for robust matching[C]//Proceedings of the 2010 IEEE International Conference on Image Processing. Piscataway, NJ:IEEE, 2010:149-152.
[5]LEUTENEGGER S, CHLI M, SIEGWART R. BRISK:binary robust invariant scalable keypoints[C]//ICCV 2011:Proceedings of the 2011 IEEE International Conference on Computer Vision. Piscataway, NJ:IEEE, 2011:2548-2555.
[6]RUBLRR E, RABAUD V, KONOLIGE K, et al. ORB:an efficient alternative to SIFT or SURF[C]//ICCV 2011:Proceedings of the2011 International Conference on Computer Vision. Piscataway, NJ:IEEE, 2011:2564-2571.
[7]ALAHI A, ORTIZ R, VANDERGHEYNST P. FREAK:fast retina keypoint[C]//Proceedings of the2012 IEEE Conference on Computer Vision and Pattern Recognition. Washington, DC:IEEE Computer Society, 2011:510-517.
[8]GEVREKCI M, GUNTURK B K. Illumination robust interest point detection[J]. Computer Vision and Image Understanding, 2009,113(4):565-571.
[9]刘天亮,戴修斌,陈昌红,等.对光照变化鲁棒的快速关键点提取与匹配[J].东南大学学报(自然科学版),2012,42(3):413-418.(LIU T L, DAI X B, CHEN C H, et al. Fast keypoint extraction and matching robust to illumination changes[J]. Journal of Southeast University(Natural Science Edition), 2012, 42(3):413-418.)
[10]PENG O, SHOUYI Y, LEIBO L, et al. A FAST extreme illumination robust feature in affine space[C]//Proceedings of the 2014International Conference on Pattern Recognition. Piscataway, NJ:IEEE, 2014:2365-2370.
[11]刘欢,郝矿荣,丁永生,等.光照鲁棒的抗模糊新组合不变矩图像匹配方法[J].传感技术学报,2013,26(9):1258-1264.(LIU H, HAO K R, DING Y S, et al. An image matching method of new illumination-robust and anti-blur combined moment invariants[J]. Chinese Journal of Sensors and Actuators, 2013, 26(9):1258-1264.)
[12]吕文先,彭启民,吕玉增.光照鲁棒的非线性相关图像匹配方法[J].计算机辅助设计与图像学报,2009,21(6):825-830.(LYU W X, PENG Q M, LYU Y Z. Illumination-robust image matching based on non-linear correlation[J]. Journal of Computer Aided Design&Computer Graphics, 2009, 21(6):825-830.)
[13]LU C W, XU L, JIA J Y. Contrast preserving decolorization[C]//Proceedings of the 2012 IEEE International Conference on Computational Photography. Piscataway, NJ:IEEE, 2012:1-7.
[14]WANG Z H, FAN B, WANG G, et al. Exploring local and overall ordinal information for robust feature description[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2015, 12(30):2513-2529.
[15]ALCANTARILLA P F, BARTOLI A, DAVISON A J. KAZE feature[C]//Proceedings of the 12th European Conference on Computer Vision. Berlin:Springer, 2012:214-227.
[16]LU C W, XU L, JIA J Y. Real-time contrast preserving decolorization[C]//Proceedings of the SIGGRAPH Asia 2012 Technical Briefs. New York:ACM, 2012:Article No. 34.
[17]卢红阳,刘且根,熊娇娇,等.基于最大加权投影求解的彩色图像灰度化对比度保留算法[J].自动化学报,2017,43(5):843-854.(LU H Y, LIU Q G, XIONG J J, et al. Maximum weighted projection solver for contrast preserving decolorization[J]. Acta Automatica Sinica, 2017, 43(5):843-854.)
[18]MAIR E, HAGER G, BURSCHKA D, et al. Adaptive and generic corner detection based on the accelerated segment test[C]//Proceedings of the 2010 European Conference on Computer Vision,LNCS 6312. Berlin:Springer, 2010:183-196.
[19]CALONDER M, LEPETIT V, STRECHA C, et al. BRIEF:computing a local binary descriptor very fast[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2012, 34(7):1281-1298.
[20]王辉,陈小雕,王毅刚.改进的非线性全局映射灰度化方法[J].计算机辅助设计与图形学学报,2013,25(10):1476-1479.(WANG H, CHEN X D, WANG Y G. The improved color-to-gray via nonlinear global mapping[J]. Journal of Computer-Aided Design and Computer Graphics, 2013, 25(10):1476-1479.)
[21]TANG F, LIM S H, CHAN N L, et al. A novel feature descriptor invariant to complex brightness changes[C]//Proceedings of the 2009IEEE Conference on Computer Vision and Pattern Recognition. Washington, DC:IEEE Computer Society, 2009:2631-2638.
[2]BAY H, ESS A, TUYTELAARS T, et al. SURF:speeded up robust features[J]. Computer Vision and Image Understanding,2008, 110(3):346-359.
[3]MOREL J M, YU G. ASIFT:a new framework for fully affine invariant image comparison[J]. SIAM Journal on Imaging Sciences,2009, 2(2):438-469.
[4]ANCUTI C O, ANCUTI C, BEKAERT P. Decolorizing images for robust matching[C]//Proceedings of the 2010 IEEE International Conference on Image Processing. Piscataway, NJ:IEEE, 2010:149-152.
[5]LEUTENEGGER S, CHLI M, SIEGWART R. BRISK:binary robust invariant scalable keypoints[C]//ICCV 2011:Proceedings of the 2011 IEEE International Conference on Computer Vision. Piscataway, NJ:IEEE, 2011:2548-2555.
[6]RUBLRR E, RABAUD V, KONOLIGE K, et al. ORB:an efficient alternative to SIFT or SURF[C]//ICCV 2011:Proceedings of the2011 International Conference on Computer Vision. Piscataway, NJ:IEEE, 2011:2564-2571.
[7]ALAHI A, ORTIZ R, VANDERGHEYNST P. FREAK:fast retina keypoint[C]//Proceedings of the2012 IEEE Conference on Computer Vision and Pattern Recognition. Washington, DC:IEEE Computer Society, 2011:510-517.
[8]GEVREKCI M, GUNTURK B K. Illumination robust interest point detection[J]. Computer Vision and Image Understanding, 2009,113(4):565-571.
[9]刘天亮,戴修斌,陈昌红,等.对光照变化鲁棒的快速关键点提取与匹配[J].东南大学学报(自然科学版),2012,42(3):413-418.(LIU T L, DAI X B, CHEN C H, et al. Fast keypoint extraction and matching robust to illumination changes[J]. Journal of Southeast University(Natural Science Edition), 2012, 42(3):413-418.)
[10]PENG O, SHOUYI Y, LEIBO L, et al. A FAST extreme illumination robust feature in affine space[C]//Proceedings of the 2014International Conference on Pattern Recognition. Piscataway, NJ:IEEE, 2014:2365-2370.
[11]刘欢,郝矿荣,丁永生,等.光照鲁棒的抗模糊新组合不变矩图像匹配方法[J].传感技术学报,2013,26(9):1258-1264.(LIU H, HAO K R, DING Y S, et al. An image matching method of new illumination-robust and anti-blur combined moment invariants[J]. Chinese Journal of Sensors and Actuators, 2013, 26(9):1258-1264.)
[12]吕文先,彭启民,吕玉增.光照鲁棒的非线性相关图像匹配方法[J].计算机辅助设计与图像学报,2009,21(6):825-830.(LYU W X, PENG Q M, LYU Y Z. Illumination-robust image matching based on non-linear correlation[J]. Journal of Computer Aided Design&Computer Graphics, 2009, 21(6):825-830.)
[13]LU C W, XU L, JIA J Y. Contrast preserving decolorization[C]//Proceedings of the 2012 IEEE International Conference on Computational Photography. Piscataway, NJ:IEEE, 2012:1-7.
[14]WANG Z H, FAN B, WANG G, et al. Exploring local and overall ordinal information for robust feature description[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2015, 12(30):2513-2529.
[15]ALCANTARILLA P F, BARTOLI A, DAVISON A J. KAZE feature[C]//Proceedings of the 12th European Conference on Computer Vision. Berlin:Springer, 2012:214-227.
[16]LU C W, XU L, JIA J Y. Real-time contrast preserving decolorization[C]//Proceedings of the SIGGRAPH Asia 2012 Technical Briefs. New York:ACM, 2012:Article No. 34.
[17]卢红阳,刘且根,熊娇娇,等.基于最大加权投影求解的彩色图像灰度化对比度保留算法[J].自动化学报,2017,43(5):843-854.(LU H Y, LIU Q G, XIONG J J, et al. Maximum weighted projection solver for contrast preserving decolorization[J]. Acta Automatica Sinica, 2017, 43(5):843-854.)
[18]MAIR E, HAGER G, BURSCHKA D, et al. Adaptive and generic corner detection based on the accelerated segment test[C]//Proceedings of the 2010 European Conference on Computer Vision,LNCS 6312. Berlin:Springer, 2010:183-196.
[19]CALONDER M, LEPETIT V, STRECHA C, et al. BRIEF:computing a local binary descriptor very fast[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2012, 34(7):1281-1298.
[20]王辉,陈小雕,王毅刚.改进的非线性全局映射灰度化方法[J].计算机辅助设计与图形学学报,2013,25(10):1476-1479.(WANG H, CHEN X D, WANG Y G. The improved color-to-gray via nonlinear global mapping[J]. Journal of Computer-Aided Design and Computer Graphics, 2013, 25(10):1476-1479.)
[21]TANG F, LIM S H, CHAN N L, et al. A novel feature descriptor invariant to complex brightness changes[C]//Proceedings of the 2009IEEE Conference on Computer Vision and Pattern Recognition. Washington, DC:IEEE Computer Society, 2009:2631-2638.