Wide baseline image matching for 3D objects

doi:10.3724/SP.J.1087.2013.00635

Journal of Computer Applications ›› 2013, Vol. 33 ›› Issue (03): 635-639.DOI: 10.3724/SP.J.1087.2013.00635

• Network and distributed techno • Previous Articles Next Articles

Wide baseline image matching for 3D objects

LI Wei^1,2,3*, SHI Zelin^1,2, YIN Jian⁴

1.Shenyang Institute of Automation, Chinese Academy of Sciences, Shenyang Liaoning 110016, China;
2.Key Laboratory of Opto-Electronic Information Processing, Chinese Academy of Sciences, Shenyang Liaoning 110016, China;
3.School of Computer and Control Engineering, University of Chinese Academy of Sciences, Beijing 100049, China;
4.Research Institute on General Development and Argumentation of Equipment of Air Force, Beijing 100076, China

Received:2012-09-11 Revised:2012-10-09 Online:2013-03-01 Published:2013-03-01
Contact: LI WEI

立体目标的宽基线图像匹配

李威^1,2,3*,史泽林^1,2,尹健⁴

1.中国科学院沈阳自动化研究所,沈阳 110016;
2.中国科学院光电信息处理重点实验室,沈阳 110016; 3.中国科学院大学计算机与控制学院,北京 100049;
4.空军装备研究院总体所,北京 100076

通讯作者: 李威
作者简介:李威(1982-),男,河南唐河人,博士研究生,主要研究方向:自动目标识别与跟踪; 史泽林(1965-),男,江苏宜兴人,研究员,博士生导师,主要研究方向:光电成像、光电跟踪、图像处理、目标识别。
基金资助:
国家973计划项目; 中国科学院国防科技创新基金资助项目(CXJJ-11)。

Abstract

Abstract: An affine invariant local feature detector has been put forward for 3D object image matching. In order to cope with view angle and scale changes, this algorithm changed the image structure to fit the circular Gaussian filter according to the principle that Gaussian filter and image structure should be compatible. Local image structures were measured by covariance matrixes of Maximally Stable Extremal Regions (MSER) having been detected in the image. Anisotropic image structures must be rotated and squeezed into isotropic image structures to guarantee the correctness of image transformation. Finally, Scale Invariant Feature Transform (SIFT) features were extracted on isotropic image structures. Coordinates of SIFT features should be changed into the original image coordinates after being extracted. The experimental results indicate that the local features extracted by this algorithm are fully affine invariant. They are suitable to be used in wide baseline image matching for 3D objects.

Key words: 3D object image matching, wide baseline, affine invariant, local feature extraction, anisotropic

摘要： 为了匹配立体目标的图像特征,提出一种仿射不变的局部特征提取算法。根据高斯滤波器的形状和大小要与图像结构相适应的原理,该算法利用图像中的最大稳定极值区域(MSER)的协方差矩阵衡量局部图像结构,并将局部图像结构变换到圆形高斯滤波器适用的形式下,以解决视角和尺度变化问题。为了保证图像变换的正确性,采用旋转压缩的方式将各向异性的图像结构变换为各向同性的图像结构。最后在各向同性的图像结构上提取尺度不变特征变换(SIFT)特征点,并将SIFT特征点的坐标变回原图像坐标。实验结果表明该算法提取的局部特征是完全仿射不变的,在立体目标的宽基线图像匹配中表现出良好的效果。

关键词: 立体目标图像匹配, 宽基线, 仿射不变, 局部特征提取, 各向异性

CLC Number:

LI Wei SHI Zelin YIN Jian. Wide baseline image matching for 3D objects[J]. Journal of Computer Applications, 2013, 33(03): 635-639.

李威史泽林尹健. 立体目标的宽基线图像匹配[J]. 计算机应用, 2013, 33(03): 635-639.

References

[1]TUYTELAARS T, MIKOLAJCZYK K. Local invariant feature detectors: a survey [J]. Foundations and Trends in Computer Graphics and Vision, 2008, 3(3):177-280.
[2]MIKOLAJCZYK K, TUYTELAARS T, SCHMID C, et al. A comparison of affine region detectors [J]. International Journal of Computer Vision, 2005, 65(1): 43-72.
[3]MIKOLAJCZYK K, SCHMID C. A performance evaluation of local descriptors [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2005, 27(10): 1615-1630.
[4]LOWE D G. Distinctive image features from scale-invariant keypoints [J]. International Journal of Computer Vision, 2004, 60(2): 91-110.
[5]LINDEBERG T. Feature detection with automatic scale selection [J]. International Journal of Computer Vision, 1998, 30(2): 79-116.
[6]MOREL J M, YU G S. Is SIFT scale invariant?[J]. Inverse Problems and Imaging, 2011,5(1): 115-136.
[7]LINDEBERG T, GARDING J. Shape-adapted smoothing in estimation of 3-D shape cues from affine deformations of local 2-D brightness structure [J]. Image and Vision Computing, 1997, 15(6): 415-434.
[8]MIKOLAJCZYK K, SCHMID C. Scale & affine invariant interest point detectors [J]. International Journal of Computer Vision, 2004, 60(1): 63-86.
[9]RUAN L, SRIDHA S, CLINTON F. Hessian-based affine adaptation of salient local image features [J]. Journal of Mathematical Imaging and Vision, 2012, 44(2):150-167.
[10]YU G S, MOREL J M. ASIFT: a new framework for fully affine invariant image comparison [J]. SIAM Journal on Imaging Sciences, 2009, 2(2):438-469.
[11]MATAS J, CHUM O, URBAN M, et al. Robust wide-baseline stereo from maximally stable extremal regions [J]. Image and Vision Computing, 2004, 22(10): 761-767.
[12]GRDING J, LINDEBERG T. Direct computation of shape cues using scale-adapted spatial derivative operators [J]. International Journal of Computer Vision, 1996, 17(2): 163-191.
[13]FISCHLER M A, BOLLES R C. Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography [J]. Communications of the ACM, 1981, 24(6): 381-395.

[1]	QIU Yunfei, LIU Xing. Local feature point matching algorithm with anti-affine property [J]. Journal of Computer Applications, 2020, 40(4): 1133-1137.
[2]	YANG Wenxia, ZHANG Liang. Logarithmic function based non-local total variation image inpainting model [J]. Journal of Computer Applications, 2018, 38(6): 1784-1789.
[3]	HAI Tao, ZHANG Lei, LIU Xuyan, ZHANG Xingang. Image enlargement based on improved complex diffusion adaptivly coupled nonlocal transform domain model [J]. Journal of Computer Applications, 2018, 38(4): 1151-1156.
[4]	HUANG Zhibiao, YAO Yu. Ultrasound image segmentation based on pixel clustering [J]. Journal of Computer Applications, 2017, 37(2): 569-573.
[5]	HAI Tao, XI Zhihong. Image enlargement based on anisotropic forth-order partial differential equation coupled to second-order partial differential equation [J]. Journal of Computer Applications, 2015, 35(4): 1084-1088.
[6]	ZHANG Fang CUI Xueying ZHANG Quan DONG Chanchan SUN Weiya BAI Yunjiao GUI Zhiguo. MLEM low-dose CT reconstruction algorithm based on variable exponent anisotropic diffusion and non-locality [J]. Journal of Computer Applications, 2014, 34(12): 3605-3608.
[7]	CHEN Jingyuan LI Jianhua GUO Weibin. Improved parallel simulation of silicon anisotropic etching based on GPU [J]. Journal of Computer Applications, 2013, 33(12): 3317-3320.
[8]	WU Shibin CHEN Bo DONG Wanli GAO Xiaoming. Ultrasound image anisotropic diffusion de-speckling method based on Mallat-Zhong discrete wavelet transform wavelet [J]. Journal of Computer Applications, 2013, 33(11): 3201-3203.
[9]	LIU Wanzhen FU Zhongliang. Local variance based anisotropic diffusion denoising method for ultrasonic image [J]. Journal of Computer Applications, 2013, 33(09): 2599-2602.
[10]	SHANG Guanhong LIU Yi ZHANG Quan GUI Zhiguo. Fuzzy diffusion PET reconstruction algorithm based on anatomical non-local means prior [J]. Journal of Computer Applications, 2013, 33(09): 2627-2630.
[11]	MA AAmin YANG Rongqian NING Hai BAI Hongmin LI Lihua WU Xiaoming. 3D scalp extraction and reconstruction of MRI brain images [J]. Journal of Computer Applications, 2013, 33(05): 1439-1442.
[12]	LI Xiao-hong ZHANG Quan LIU Yi GUI Zhi-guo. High quality median prior image reconstruction algorithm based on wavelet shrinkage and forward-and-backward diffusion [J]. Journal of Computer Applications, 2012, 32(12): 3357-3360.
[13]	YANG Jin LIU Zhi-qin WANG Yao-bin GAO Xiao-ming. Improved anisotropic diffusion ultrasound image denoising method based on logarithmic compression [J]. Journal of Computer Applications, 2012, 32(11): 3218-3220.
[14]	KE Dan-dan CAI Guang-cheng CAO Qian-qian. Anisotropic diffusion denoising method based on image feature [J]. Journal of Computer Applications, 2012, 32(03): 742-745.
[15]	LIU Kui SU Ben-yue ZHAO Xiao-jing. Image inpainting method based on structure tensor [J]. Journal of Computer Applications, 2011, 31(10): 2711-2713.

Wide baseline image matching for 3D objects

立体目标的宽基线图像匹配

PDF

Knowledge

Abstract

Cite this article

share this article

References

Related Articles 15

Recommended Articles

Metrics