A Performance Evaluation of SIFT and SURF for Multispectral Image Matching

This paper evaluates the performance of SIFT and SURF for cross band matching of multispectral images. The evaluation is based on matching a reference spectral image with the images acquired at different spectral bands. The reference image possesses scale and (in-plane) rotational differences in addition to spectral variations. Additive white Gaussian noise is also added to compare performance degradation at different noise levels. We use the precision and repeatability criteria for performance evaluation. Experimental results demonstrate that SIFT performs better than SURF in multispectral environment.

[1]  G LoweDavid,et al.  Distinctive Image Features from Scale-Invariant Keypoints , 2004 .

[2]  Robert C. Bolles,et al.  Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography , 1981, CACM.

[3]  Luc Van Gool,et al.  SURF: Speeded Up Robust Features , 2006, ECCV.

[4]  C. Schmid,et al.  Indexing based on scale invariant interest points , 2001, Proceedings Eighth IEEE International Conference on Computer Vision. ICCV 2001.

[5]  David Salesin,et al.  Wavelets for computer graphics: a primer.1 , 1995, IEEE Computer Graphics and Applications.

[6]  Cordelia Schmid,et al.  A Comparison of Affine Region Detectors , 2005, International Journal of Computer Vision.

[7]  Samia Boukir,et al.  Relevance of airborne lidar and multispectral image data for urban scene classification using Random Forests , 2011 .

[8]  David Salesin,et al.  Wavelets for computer graphics: a primer. 2 , 1995, IEEE Computer Graphics and Applications.

[9]  Cordelia Schmid,et al.  Evaluation of Interest Point Detectors , 2000, International Journal of Computer Vision.

[10]  Axel Pinz,et al.  Computer Vision – ECCV 2006 , 2006, Lecture Notes in Computer Science.

[11]  Cordelia Schmid,et al.  A Performance Evaluation of Local Descriptors , 2005, IEEE Trans. Pattern Anal. Mach. Intell..

[12]  Sridha Sridharan,et al.  Multi-spectral fusion for surveillance systems , 2008, Comput. Electr. Eng..