论文信息 - Image registration with luminance transform operation

Image registration with luminance transform operation

In this paper, the robustness of the measures that are used for image registration are tested for common distortions seen in images. Luminance Transform Operation (LTO) is proposed to be used in image registration applications. Performance of the LTO is compared to the Normilized Cross Correlation (NCC) and Mutual Information (MI) measures. Image registration process is carried out between different wavelenght bands of hyperspectral images. Non-linear luminance changes, noise and rotation effects are added to target images, then performances of NCC, MI and LTO are tested for their robustness on registration of corrupted target images. Results depicted that LTO measure is more robust to mentioned distortion effects than other similarity measures.

Sarp Ertürk | M. Kemal Güllü | Alp Ertürk | Deniz Gerçek | Davut Çesmeci

[1] Oguzhan Urhan,et al. Blotch detection and removal for archive film restoration , 2008 .

[2] Alexandre Bernardino,et al. Random features vs Harris Corners in Real-Time Visual Egomotion Estimation , 2010 .

[3] Max A. Viergever,et al. Mutual-information-based registration of medical images: a survey , 2003, IEEE Transactions on Medical Imaging.

[4] K. Revathy,et al. USING MUTUAL INFORMATION AND CROSS CORRELATION AS METRICS FOR REGISTRATION OF IMAGES , 2008 .

[5] David G. Lowe,et al. Object recognition from local scale-invariant features , 1999, Proceedings of the Seventh IEEE International Conference on Computer Vision.

[6] Lisa M. Brown,et al. A survey of image registration techniques , 1992, CSUR.

[7] M. Kemal Gullu. Block motion estimation using luminance transform based matching , 2008, 2008 IEEE 16th Signal Processing, Communication and Applications Conference.

[8] Dimitris G. Manolakis,et al. Detection algorithms for hyperspectral imaging applications , 2002, IEEE Signal Process. Mag..