Abstract

Image matching, it's a crucial problem in computer vision and image processing. In order to improve the matching results, a proposed solution consists on employing geometric constraints. In this paper, we propose two effectiveness matching methods, which are called "First Found Is Winner" (FFIW) and "Polarity Coordinates And FFIW" (PC_FFIW). The first one is based on photometric data, while the second one, uses both photometric and geometric data. The proposed methods are based on three-step. Firstly, we detect for each image its corner points. Secondly, descriptors vectors are calculated for each corner points. Thirdly, to match the pair of images P and Q, we apply a matching algorithm optimized to find the best match for each descriptor from the first image with the descriptors of the second image. Experimental results presented to demonstrate that our proposed methods are efficient and give promising results in terms of repeatability and processing time.

References

• Grauman,K. , Darrel, T. , "Efficient Image Matching with Distributions of Local Invariant Features". In proceeding of the IEEE Conference on Computer Vision and Pattern Recognition, 2005.
• Zitova, B. , Flusser, J. , "Image registration methods: a survey, Image and Vision Computing", 21(2003), pp. 977-1000.
• Tingfang, S. , Ziwen, F. , "Digital Image Processing and Pattern Recognition", Beijing: Press of Beijing University of Science and engineering, 1998, PP. 150-151.
• Schmid, C. , Mohr, R. , "local Gray value Invariants for Image Retrieval", IEEE PAMI, 19 (1997): PP. 530-534.
• Harris, C. , Stephens, M. , "A Combined Corner and Edge Detector", In Proc. Alvey Vision Conf. , Manchester (1988) pp. 189-192.
• Lowe, D. G. , "Distinctive Image Feature from Scale-Invariant Key points", International Journal of Computer Vision, 60, 2 (2004), PP. 91-110.
• Tuytelaars, T. , Van Gool, L. ," wide Baseline Stereo Matching Based on Local Affinely Invariant Regions", In: BMVC, (2000), pp. 412-425.
• Shan, B. , Cui, F. , "Image Matching Based on Local Invariant Feature and Histogram-Based Similar Distance". In First International Workshop on Education Technology and Computer Science - Volume 01 (ETCS 2009). Pp 1030-1033.
• Schmid, C. , "Appariement des images par l'invariant local niveau de gris : Application à l'indexation d'une base d'objets" thèse de doctorat, institut nationale Polytechnique de Grenoble ,1996.
• Abbas. A. , "Construction d'une mosaïque à partir d'une séquence vidéo". mémoire présenté Pour l'obtention du grade du maitre en science. Université de Montréal 2002
• Caelli, T. , Kosinov, S. , "An eigenspace projection clustering method for inexact graph matching". 2004. IEEE Trans. Pattern Anal Machine Intell. 26 (4), pp. 515-519;
• Leordanu, M. , Hebert, M. , "A spectral technique for correspondences problems using pairwise Constraint". In: Proc. IEEE. ICCV 2005.
• Kim, G. , Faloutsos, C. , and Hebert, M. , "Unsupervised modeling of object categories using link analysis Techniques". In: Proc, IEEE CVPR 2008.
• Yuan, Y. , Pang, Y. , Wang, K. , and Shang, M. , "Efficient image matching using weighted voting". Pattern Recognition Letters, February 2011.