Abstract
This article addresses the problem of registering high-resolution, small field-of-view images with low-resolution panoramic images provided by a panoramic catadioptric video sensor. Such systems may find application in surveillance and telepresence systems that require a large field of view and high resolution at selected locations. Although image registration has been studied in more conventional applications, the problem of registering panoramic and conventional video has not previously been addressed, and this problem presents unique challenges due to (i) the extreme differences in resolution between the sensors (more than a 16:1 linear resolution ratio in our application), and (ii) the resolution inhomogeneity of panoramic images. The main contributions of this article are as follows. First, we introduce our foveated panoramic sensor design. Second, we show how a coarse registration can be computed from the raw images using parametric template matching techniques. Third, we propose two refinement methods allowing automatic and near real-time registration between the two image streams. The first registration method is based on matching extracted interest points using a closed form method. The second registration method is featureless and based on minimizing the intensity discrepancy allowing the direct recovery of both the geometric and the photometric transforms. Fourth, a comparison between the two registration methods is carried out, which shows that the featureless method is superior in accuracy. Registration examples using the developed methods are presented.
- Amintabar, A. and Boufama, B. 2008. Homography-based plane identification and matching. In Proceedings of the IEEE International Conference on Image Processing.Google Scholar
- Bay, H., Ess, A., Tuytelaars, T., and Gool, L. V. 2008. SURF: Speeded Up Robust Features. Comput Vision Image Understand. 110, 3, 346--359. Google Scholar
Digital Library
- Boult, T. E., Gao, X., Micheals, R., and Eckmann, M. 2004. Omni-directional visual surveillance. Image Vision Comput. 22, 7, 515--534.Google Scholar
Cross Ref
- Brown, M. and Lowe, D. G. 2007. Automatic panoramic image stitching using invariant features. Int. J. Comput. Vision 74, 1, 59--73. Google Scholar
Digital Library
- Chen, J., Chen, C., and Chen, Y. 2003. Fast algorithm for robust template matching with M-estimators. IEEE Trans. Signal Process 51, 1, 230--243. Google Scholar
Digital Library
- Conroy, T. L. and Moore, J. B. September 1999. Resolution invariant surfaces for panoramic vision systems. InProceedings of the IEEE Conference on Computer Vision.Google Scholar
- Danilidis, K. and Geyer, C. 2000. Omnidirectional vision: Theory and algorithms. In Proceedings of the IEEE International Conference on Patter Recognition. Google Scholar
Digital Library
- Dornaika, F. and Elder, J. 2002. Image registration for foveated omnidirectional sensing. In Proceedings of the European Conference on Computer Vision. Lecture Notes in Computer Science, vol. 2353. Google Scholar
Digital Library
- Dufourneau, Y., Schmid, C., and Horaud, R. 2000. Matching images with different resolutions. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Google Scholar
- Fischler, M. A. and Bolles, R. C. 1981. Random sample consensus: A paradigm for model fitting with applications to image analysis and automated cartography. Comm. ACM 24, 6, 381--395. Google Scholar
Digital Library
- Fletcher, R. 1990. Practical Methods of Optimization. Wiley, New York. Google Scholar
Digital Library
- Harris, C. and Stephens, M. 1988. A combined corner and edge detector. In Proceedings of the Alvey Vision Conference.Google Scholar
- He Q. and Chu, C. H. 2006. Planar surface detection in image pairs using homographic constraints. In Proceedings of the 2nd International Symposium on Advances in Visual Computing. Lecture Notes in Computer Science, Vol. 4291. Google Scholar
Digital Library
- Kanatani, K. and Ohta, N. 1999. Accuracy bounds and optimal computation of homography for image mosaicing applications. In Proceedings of the IEEE Conference on Computer Vision.Google Scholar
- Kim, D. H., Yoon, Y. I., and Choi, J. S. 2003. An efficient method to build panoramic image mosaics. Patt. Recog. Lett 24, 14, 2421--2429. Google Scholar
Digital Library
- Lin, S. S. and Bajcsy, R. 2006. Single-view-point omnidirectional catadioptric cone mirror imager. IEEE Trans. Patt. Anal. Machine Intell. 28, 5, 840--845. Google Scholar
Digital Library
- Lowe, D. 2004. Distinctive image features from scale invariant keypoints. Int. J. Comput. Vision 60, 2, 91--100. Google Scholar
Digital Library
- Ohta, J. 2007. Smart CMOS Image Sensors and Applications. CRC Press, Taylor & Francis Group.Google Scholar
- Peng, G., Xie, S., and Cheng, L. 2005. An HVSM for improving the homing ability of visual robots. Int. J. Intell. Syst. Techn. Appl. 1, 2, 18--31. Google Scholar
Digital Library
- Pilu, M. 1997. A direct method for stereo correspondence based on singular value decomposition. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Google Scholar
Digital Library
- Prime, S. L., Tsotsos, L., Keith, G. P., and Crawford, J. D. 2007. Visual memory capacity in transsaccadic integration. Exp. Brain Resear. 180, 4, 609--628.Google Scholar
Cross Ref
- Scaramuzza, D. and Siegwart, R. 2007. A practical toolbox for calibrating omnidirectional cameras. In Vision Systems: Applications, Intech.Google Scholar
- Schwartz, W. B., Kembhavi, A., Harwood, D., and Davis, L. S. 2009. Human detection using partial least squares analysis. In Proceedings of the IEEE International Conference on Computer Vision.Google Scholar
- Spacek, L. and Burbridge, C. 2007. Instantaneous robot self-localisation and motion estimation with omnidirectional vision. Rob. Auton. Syst. 55, 667--674. Google Scholar
Digital Library
- Swaminathan, R., Grossberg, M. D., and Nayar, S. K. 2006. Non-single viewpoint catadioptric cameras: Geometry and analysis. Int. J. Comput. Vision 66, 3, 211--229. Google Scholar
Digital Library
- Tanaka, K., Sano, M., Ohara, S., and Okudaira, M. 2000. A parametric template method and its application to robust matching. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Google Scholar
- Torre, F., Vallespi, C., Rybski, P. E., Veloso, M., and Kanade, T. 2005. Learning to track multiple people in omnidirectional video. In Proceedings of the IEEE International Conference on Robotics and Automation.Google Scholar
- Traver, V. J. and Bernardino, A. 2010. A review of log-polar imaging for visual perception in robotics. Rob. Auton. Syst. 58, 4, 378--398. Google Scholar
Digital Library
- Wu, Y., Kanade, T., Li, C., and Cohn, J. 2000. Image registration using wavelet-based motion model. Int. J. Comput. Vision 38, 2, 129--152. Google Scholar
Digital Library
- Zitová, B. and Flusser, J. 2003. Image registration methods: a survey. Image Vision Comput. 21, 977--1000.Google Scholar
Cross Ref
Index Terms
Image registration for foveated panoramic sensing
Recommendations
Image Registration for Foveated Omnidirectional Sensing
ECCV '02: Proceedings of the 7th European Conference on Computer Vision-Part IVThis paper addresses the problem of registering highresolution, small FOV images with low-resolution panoramic images provided by an omnidirectional catadioptric video sensor. Such systems may find application in surveillance and telepresence systems ...
Image Registration of Sectioned Brains
The physical (microtomy), optical (microscopy), and radiologic (tomography) sectioning of biological objects and their digitization lead to stacks of images. Due to the sectioning process and disturbances, movement of objects during imaging for example, ...
Non-rigid medical image registration using image field in Demons algorithm
Highlights- A novel non-rigid registration method for medical images is proposed by introducing the image field into Demons algorithm.
Graphical abstractDisplay Omitted
AbstractIn medical imaging application, non-rigid registration is an important step that requires both accuracy and efficiency to align the deformed query image with the reference image. For non-rigid registration, Demons is generally used as ...






Comments