一种基于面阵摆扫式航空影像的特征匹配方法

doi:10.12082/dqxxkx.2022.210394

摘要/Abstract

摘要：

特征匹配是面阵摆扫式航空影像处理的关键步骤,针对传统特征匹配方法在面阵摆扫式航空影像匹配时存在匹配点数量少,分布不匀均的问题,本文提出一种基于自适应亮度空间的特征匹配方法。首先根据影像POS（Postion Oriental System）信息求解待匹配影像间变换关系进行影像校正,在校正后的影像上构建自适应亮度空间,使用ORB算子和BEBLID算法在亮度空间上获取特征点和二进制特征描述符,然后基于汉明距离获取初始匹配点,使用RANSAC算法剔除粗差,最后将匹配点变换到原始影像上得到最终匹配结果。本文选取6组具有视角差异及亮度变化的面阵摆扫式航空影像进行实验,将本文算法与SIFT、SURF、ORB、ORB+BEBLID、ASIFT等匹配方法进行比较,结果表明：本文算法通过建立影像间变换关系,构建自适应亮度空间,使得算法提取的特征点数量增加1.5倍,获取匹配点数量是其他算法的3倍以上,且匹配点分布更加均匀,匹配效率高于其他算法,验证了本文算法在具有亮度变化及视角差异的面阵摆扫式航空影像上匹配的有效性。

关键词: 影像匹配, 自适应亮度空间, BEBLID, RANSAC, 单应变换, 面阵摆扫式航空影像, POS

Abstract:

Feature matching is a key step in the area array swing-scan aerial image processing. Traditional feature matching method has some limitations in the area array swing-scan aerial image matching, e.g., the number of matching points is small, and the distribution is uneven. This paper proposes a method based on adaptive feature matching method in brightness space. First, according to the image Position Oriental System (POS) information, we solve the transformation relationship between the images to be matched for image correction, build an adaptive brightness space on the corrected image, and use ORB operator and BEBLID algorithm to obtain feature points and binary feature descriptors in the brightness space. Then, based on the Hamming distance, we obtain the initial matching points, use the RANSAC algorithm to eliminate gross error. Finally, the matching points are transformed to the original image to obtain the final matching result. This paper selects six groups of swept aerial images with different viewing angles and brightness changes for experiments and compares the algorithm proposed in this paper with SIFT, SURF, ORB, ORB+BEBLID, and ASIFT matching methods. The results show that the algorithm in this paper builds an adaptive brightness space by establishing the transformation relationship between images, so that the number of feature points extracted by the algorithm is increased by 1.5 times, and the number of matching points obtained is more than 3 times that of other algorithms. The distribution of matching points is more uniform, and the matching efficiency is higher than other algorithms. The algorithm verifies the effectiveness of the algorithm proposed in this paper to match the area array swing-scan aerial images with brightness changes and viewing angle differences.

Key words: image matching, adaptive brightness space, BEBLID, RANSAC, homography, area array swing scanning aerial image, POS

张昆, 王涛, 张艳, 郑迎辉, 赵祥, 李芳芳. 一种基于面阵摆扫式航空影像的特征匹配方法[J]. 地球信息科学学报, 2022, 24(3): 522-532.DOI:10.12082/dqxxkx.2022.210394

ZHANG Kun, WANG Tao, ZHANG Yan, ZHENG Yinghui, ZHAO Xiang, LI Fangfang. A Feature Matching Method based on Area Array Swing-Scan Aerial Image[J]. Journal of Geo-information Science, 2022, 24(3): 522-532.DOI:10.12082/dqxxkx.2022.210394

图/表 12

图1

图2

图3

图4

图5

图6

表1

图7

图8

图9

图10

表2

参考文献 32

[1]	王义坤. Research of wide swath imaging technology based on area-array detector adopting whiskbroom scanning mode[D]. 上海:中国科学院研究生院(上海技术物理研究所), 2015.
	[ Wang Y K. Research of wide swath imaging technology based on area-array detector adopting whiskbroom scanning mode[D]. Shanghai: Graduate School of Chinese Academy of Sciences(Shanghai Institute of Technical Physics), 2015. ]
[2]	李赛, 胡勇, 巩彩兰, 等. 面阵摆扫热红外航空影像分步几何校正方法[J]. 红外与毫米波学报, 2020, 39(2):242-250.
	[ Li S, Hu Y, Gong C L, et al. A step-by-step geometric correction and error analysis of swing-swept array thermal infrared aerial image[J]. Journal of Infrared and Millimeter Waves, 2020, 39(2):242-250. ] DOI: 10.11972/j.issn.1001-9014.2020.02.012
[3]	杨国鹏, 周欣, 韦红波, 等. 面阵摆扫航空相机序列图像的大区域无缝拼接[J]. 测绘科学, 2020, 45(3):46-52,60.
	[ Yang G P, Zhou X, Wei H B, et al. Large-area seamless mosaic for the sequence images from a frame-sweep aerial cameras with plane array[J]. Science of Surveying and Mapping, 2020, 45(3):46-52,60. ] DOI: 10.16251/j.cnki.1009-2307.2020.03.008
[4]	李海星, 惠守文, 丁亚林. 国外航空光学测绘装备发展及关键技术[J]. 电子测量与仪器学报, 2014, 28(5):469-477.
	[ Li H X, Hui S W, Ding Y L. Development and key techniques of optical mapping equipment in foreign airborne[J]. Journal of Electronic Measurement and Instrumentation, 2014, 28(5):469-477. ] DOI: 10.13382/j.jemi.2014.05.002
[5]	Lowe D G. Distinctive image features from scale-invariant keypoints[J]. International Journal of Computer Vision, 2004, 60(2):91-110. DOI: 10.1023/B:VISI.0000029664.99615.94 doi: 10.1023/B:VISI.0000029664.99615.94
[6]	Ke Y, Sukthankar R. PCA-SIFT: a more distinctive representation for local image descriptors[C]// Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2004. CVPR 2004. IEEE, 2004: II, DOI: 10.1109/CVPR.2004.1315206
[7]	刘立, 彭复员, 赵坤, 等. 采用简化SIFT算法实现快速图像匹配[J]. 红外与激光工程, 2008, 37(1):181-184.
	[ Liu L, Peng F Y, Zhao K, et al. Simplified SIFT algorithm for fast image matching[J]. Infrared and Laser Engineering, 2008, 37(1):181-184. ]
[8]	Bay H, Tuytelaars T, Van Gool L. SURF: speeded up robust features[M]// Computer Vision - ECCV 2006. Berlin, Heidelberg: Springer Berlin Heidelberg, 2006:404-417. DOI: 10.1007/11744023_32
[9]	钱学飞, 沈映政, 王友昆, 等. 一种改进SURF的视觉影像匹配算法[J]. 测绘通报, 2021(1):99-102,107.
	[ Qian X F, Shen Y Z, Wang Y K, et al. An improved SURF method on the visual image area[J]. Bulletin of Surveying and Mapping, 2021(1):99-102,107. ] DOI: 10.13474/j.cnki.11-2246.2021.0018
[10]	Rublee E, Rabaud V, Konolige K, et al. ORB: An efficient alternative to SIFT or SURF[C]// 2011 International Conference on Computer Vision. IEEE, 2011:2564-2571. DOI: 10.1109/ICCV.2011.6126544
[11]	Calonder M, Lepetit V, Strecha C, et al. Binary robust independent elementary features[C]. European Conference on Computer Vision, 2010:778-792. DOI: 10.1007/978-3-642-15561-1_56
[12]	廖泓真, 王亮, 孙宏伟, 等. 一种改进的ORB特征匹配算法[J]. 北京航空航天大学学报, 2021, 5(19):1-7.
	[ Liao H Z, Wang LS H, et al. An improved ORB feature matching algorithm[J]. Journal of Beijing University of Aeronautics and Astronautics, 2021, 5(19):1-7. ] DOI: 10.13700/j.bh.1001-5965.2020.0359
[13]	Morel J M, Yu G S. ASIFT: A new framework for fully affine invariant image comparison[J]. SIAM Journal on Imaging Sciences, 2009, 2(2):438-469. DOI: 10.1137/080732730 doi: 10.1137/080732730
[14]	佟国峰, 李勇, 刘楠, 等. 大仿射场景的混合特征提取与匹配[J]. 光学学报, 2017, 37(11):215-222.
	[ Tong G F, Li Y, Liu N, et al. Mixed feature extraction and matching for large affine scene[J]. Acta Optica Sinica, 2017, 37(11):215-222. ] DOI: 10.3788/AOS201737.1115003
[15]	赵鹏图, 达飞鹏. 基于局部特征的大视角图像匹配[J]. 光学学报, 2019, 39(5):133-140.
	[ Zhao P T, Da F P. Image matching with large viewing angle based on local features[J]. Acta Optica Sinica, 2019, 39(5):133-140. ] DOI: 10.3788/AOS201939.0510002
[16]	张明祥, 王泽根, 白茹月, 等. 一种由粗到精的光学与SAR遥感图像配准算法[J]. 地球信息科学学报, 2020, 22(11):2238-2246. doi: 10.12082/dqxxkx.2020.190742
	[ Zhang M X, Wang Z G, Bai R Y, et al. A coarse-to-fine optical and SAR remote sensing image registration algorithm[J]. Journal of Geo-Information Science, 2020, 22(11):2238-2246. ] DOI: 10.12082/dqxxkx.2020.190742
[17]	何腕营, 王竞雪. 基于直线间几何属性特征相似性约束的组直线匹配算法[J]. 地球信息科学学报, 2020, 22(9):1766-1778. doi: 10.12082/dqxxkx.2020.190334
	[ He W Y, Wang J X. Pair-wise line matching algorithm based on feature similarity constraints of geometric attributes between the lines[J]. Journal of Geo-Information Science, 2020, 22(9):1766-1778. ] DOI: 10.120 82/dqxxkx.2020.190334
[18]	Heidelberg S V B. Erratum[M]// Relict Species. Berlin, Heidelberg: Springer Berlin Heidelberg, 2009:450-450. DOI: 10.1007/978-3-642-33783-3_16
[19]	Mukherjee P, Lall B. Saliency and KAZE features assisted object segmentation[J]. Image and Vision Computing, 2017, 61:82-97. DOI: 10.1016/j.imavis.2017.02.008 doi: 10.1016/j.imavis.2017.02.008
[20]	Wan X E, Liu J G, Yan H S, et al. Illumination-invariant image matching for autonomous UAV localisation based on optical sensing[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2016, 119:198-213. DOI: 10.1016/j.isprsjprs.2016.05.016 doi: 10.1016/j.isprsjprs.2016.05.016
[21]	Wu B, Zeng H, Hu H. Illumination invariant feature point matching for high-resolution planetary remote sensing images[J]. Planetary and Space Science, 2018, 152:45-54. DOI: 10.1016/j.pss.2018.01.007 doi: 10.1016/j.pss.2018.01.007
[22]	Lai Y, Tariq M. Group enhancement for matching of multi-view image with overlap fuzzy feature[J]. Multimedia Tools and Applications, 2020, 79(3/4):2069-2084. DOI: 10.1007/s11042-019-08173-0 doi: 10.1007/s11042-019-08173-0
[23]	杨佳宾, 姜永涛, 杨幸彬, 等. 基于Dense SIFT特征的无人机影像快速拼接方法[J]. 地球信息科学学报, 2019, 21(4):588-599. doi: 10.12082/dqxxkx.2019.190001
	[ Yang J B, Jiang Y T, Yang X B, et al. A fast mosaic algorithm of UAV images based on dense SIFT feature matching[J]. Journal of Geo-Information Science, 2019, 21(4):588-599. ] DOI: 10.12082/dqxxkx.2019.190001
[24]	龚健雅, 季顺平. 摄影测量与深度学习[J]. 测绘学报, 2018, 47(6):693-704.
	[ Gong J Y, Ji S P. Photogrammetry and deep learning[J]. Acta Geodaetica et Cartographica Sinica, 2018, 47(6):693-704. ] DOI: 10.11947/j.AGCS.2018.20170640
[25]	Suárez I, Sfeir G, Buenaposada J M, et al. BEBLID: Boosted efficient binary local image descriptor[J]. Pattern Recognition Letters, 2020, 133:366-372. DOI: 10.1016/j.patrec.2020.04.005 doi: 10.1016/j.patrec.2020.04.005
[26]	于广瑞, 孙国新, 时春霖, 等. 一种基于差分技术的无人机影像高效拼接方法[J]. 测绘科学, 2020, 45(10):110-118.
	[ Yu G R, Sun G X, Shi C L, et al. An efficient UAV images mosaic method based on difference technology[J]. Science of Surveying and Mapping, 2020, 45(10):110-118. ] DOI: 10.16251/j.cnki.1009-2307.2020.10.016
[27]	陈迎春. 基于色调映射的快速低照度图像增强[J]. 计算机工程与应用, 2020, 56(9):234-239.
	[ Chen Y C. Fast low-light image enhancement based on tone mapping[J]. Computer Engineering and Applications, 2020, 56(9):234-239. ] DOI: 10.3778/j.issn.1002-8331.1903-0076
[28]	郝志成, 吴川, 杨航, 等. 基于双边纹理滤波的图像细节增强方法[J]. 中国光学, 2016, 9(4):423-431.
	[ Hao Z C, Wu C, Yang H, et al. Image detail enhancement method based on multi-scale bilateral texture filter[J]. Chinese Optics, 2016, 9(4):423-431. ] DOI: 10.3788/CO.20160904.0423
[29]	Ahn H, Keum B, Kim D, et al. Adaptive local tone mapping based on retinex for high dynamic range images[C]// 2013 IEEE International Conference on Consumer Electronics (ICCE). IEEE, 2013:153-156. DOI: 10.1109/ICCE.2013.6486837
[30]	Drago F, Myszkowski K, Annen T, et al. Adaptive logarithmic mapping for displaying high contrast scenes[J]. Computer Graphics Forum, 2003, 22(3):419-426. DOI: 10.1111/1467-8659.00689 doi: 10.1111/cgf.2003.22.issue-3
[31]	Reinhard E, Stark M, Shirley P, et al. Photographic tone reproduction for digital images[J]. ACM Transactions on Graphics, 2002, 21(3):267-276. DOI: 10.1145/566654.566575 doi: 10.1145/566654.566575
[32]	Rosten E, Drummond T. Machine learning for high-speed corner detection[C]. European conference on computer vision. 2006:430-443. DOI: 10.1007/11744023_34

实验影像	影像分辨率/像元	焦距/mm	GSD/cm	倾斜角/。	航高/m
a组影像	4864×3232	300	5.6/6.4	16.9/-32.6	2190/2190
b组影像	4864×3232	300	5.4/5.8	3.8/19.6	2200/2210
c组影像	4864×3232	300	6.7/6.8	40.2/27.8	2197/2207
d组影像	4864×3232	300	5.3/5.5	3.1/13.8	2197/2207
e组影像	4864×3232	300	5.3/6.2	-4.8/24.8	2107/2210
f组影像	4864×3232	300	5.3/6.3	-3.6/29.8	2210/2207

分组	算法	左/右影像特征点/个	最终匹配点对/个	匹配效率/%
e	SIFT	1062/3159	38	3.5
	SURF	5990/12 260	67	1.1
	ORB	2825/15 362	90	3.1
	ORB+BEBLID	2825/15 362	51	1.8
	ASIFT	13 624/123 688	175	0.5
	本文算法	37 800/40 631	1363	3.6
f	SIFT	3132/1649	46	2.7
	SURF	14 166/9627	51	0.5
	ORB	16 128/8512	72	0.8
	ORB+BEBLID	16 238/8512	96	1.1
	ASIFT	80 640/10 588	151	1.4
	本文算法	23 261/12 186	863	7.0