Registration and Fusion of Blurred Images

We present a maximum a posteriori solution to problems of accurate registration of blurred images and recovery of an original undegraded image. Our algorithm has the advantage that both tasks are performed simultaneously. An efficient implementation scheme of alternating minimizations is presented. A simulation and a real-data experiment demonstrate the superb performance of the algorithm.

[1]  Jan Flusser,et al.  Combined Invariants to Linear Filtering and Rotation , 1999, Int. J. Pattern Recognit. Artif. Intell..

[2]  Shmuel Peleg,et al.  Restoration of multiple images with motion blur in different directions , 2000, Proceedings Fifth IEEE Workshop on Applications of Computer Vision.

[3]  J Flusser,et al.  Recognition of blurred images by the method of moments , 1996, IEEE Trans. Image Process..

[4]  Rick S. Blum,et al.  A hybrid image registration technique for a digital camera image fusion application , 2001, Inf. Fusion.

[5]  Yoram Bresler,et al.  Perfect blind restoration of images blurred by multiple filters: theory and efficient algorithms , 1999, IEEE Trans. Image Process..

[6]  Kiyoharu Aizawa,et al.  Registration and blur estimation methods for multiple differently focused images , 1999, Proceedings 1999 International Conference on Image Processing (Cat. 99CH36348).

[7]  Nikolas P. Galatsanos,et al.  EM-based simultaneous registration, restoration, and interpolation of super-resolved images , 2003, Proceedings 2003 International Conference on Image Processing (Cat. No.03CH37429).

[8]  Patrizio Campisi,et al.  Multichannel blind image deconvolution using the Bussgang algorithm: spatial and multiresolution approaches , 2003, IEEE Trans. Image Process..

[9]  J. Flusser,et al.  Shift-invariant multichannel blind restoration , 2003, 3rd International Symposium on Image and Signal Processing and Analysis, 2003. ISPA 2003. Proceedings of the.

[10]  Zarina Myles,et al.  Recovering affine motion and defocus blur simultaneously , 1996, Proceedings CVPR IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[11]  Jan Flusser,et al.  Multichannel blind iterative image restoration , 2003, IEEE Trans. Image Process..

[12]  J. Flusser,et al.  Invariant-based registration of rotated and blurred images , 1999, IEEE 1999 International Geoscience and Remote Sensing Symposium. IGARSS'99 (Cat. No.99CH36293).

[13]  Gjlles Aubert,et al.  Mathematical problems in image processing , 2001 .

[14]  Jan Flusser,et al.  Robust detection of significant points in multiframe images , 1999, Pattern Recognit. Lett..

[15]  Hung-Ta Pai,et al.  On eigenstructure-based direct multichannel blind image restoration , 2001, IEEE Trans. Image Process..

[16]  Jan Flusser,et al.  Image registration methods: a survey , 2003, Image Vis. Comput..

[17]  Ying Zhang,et al.  Determination of blur and affine combined invariants by normalization , 2002, Pattern Recognit..