1 / 23

ON THE IMPROVEMENT OF IMAGE REGISTRATION FOR HIGH ACCURACY SUPER-RESOLUTION

ON THE IMPROVEMENT OF IMAGE REGISTRATION FOR HIGH ACCURACY SUPER-RESOLUTION. Michalis Vrigkas , Christophoros Nikou , Lisimachos P. Kondi University of Ioannina Department of Computer Science Ioannina , Greece. MOTIVATION. Objective

jiro
Download Presentation

ON THE IMPROVEMENT OF IMAGE REGISTRATION FOR HIGH ACCURACY SUPER-RESOLUTION

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. ON THE IMPROVEMENT OF IMAGE REGISTRATION FOR HIGH ACCURACY SUPER-RESOLUTION MichalisVrigkas, ChristophorosNikou, Lisimachos P. KondiUniversity of IoanninaDepartment of Computer ScienceIoannina, Greece

  2. MOTIVATION • Objective • Reconstruct a high-resolution image from a sequence of low-resolution images. • Improve spatial resolution. • Constraints on low-resolution images • Motion • Rotation • Blurring • Subsampling • Additive noise

  3. APPROACH • MAP scheme for image super-resolution. • Registration in two parts • At first, the LR images are registered by establishing correspondences between robust SIFT (Scale-Invariant Feature Transform) features. • In the second step, the estimation of the registration parameters is fine tuned along with the estimation of the HR image. • Mutual Information Criterion: maximize the mutual information between HR image and each of the upscaled LR images.

  4. FORMULATION MODEL • Let the high-resolution image where • The set of LR images is described as We consider p LR images each of size

  5. FORMULATION MODEL (cont.) • Observation model: • All images are ordered lexicographically • represents zero-mean additive Gaussian noise, • is the degradation matrix, performing the operations of: • motion • blur • down-sampling

  6. MAP ESTIMATOR • The Gaussian prior for the HR image is: • is the Laplacian of the image z • controls the precision and the shape of the distribution • The likelihood of the LR images is Gaussian:

  7. MAP ESTIMATOR (cont.) • MAP approach • Maximize • Which leads to a MAP functional to be minimized with respect to HR image z and the transformation parameters s: • Use gradient descent method • The update equation is given by: where εn is the step size at the n-th iteration.

  8. SCALE INVARIANT FEATUTE TRANSFORM - SIFT • Objective: independently detect corresponding keypoints in scaled versions of the same image. • Idea: Given a keypoint in two images, determine if the surrounding neighborhoods contain the same structure up to scale. • SIFT features are invariant to: • Image scale and rotation • Affine transformations • Changes in illumination and noise [D. G. Lowe. "Distinctive image features from scale invariant keypoints.”International Journal of Computer Vision 60 (2), pp. 91-110, 2004.]

  9. MUTUAL INFORMATION CRITERION • Basics: the mutual information is maximized when the two images are correctly registered. • The mutual information between two images A and B is: • H(A) and H(B) are the marginal entropies of the random variables A and B. • H(A,B)is the joint entropy.

  10. MUTUAL INFORMATION CRITERION (cont.) • Normalized Mutual Information: • Robust measure in order to provide invariance to the overlapping areas between the two images. • Problem: • If mutual information is not initialized close to the optimal solution it is trapped by local maxima. • Good initialization is important. • Solution: • Initialization using SIFT descriptors.

  11. Image Registration • Estimation of registration parameters in two steps. • First step, LR images are registered by employing SIFT features. • Minimization of mean square error between the locations of features between the reference image and the LR images. • Provides good initialization.

  12. Image Registration (cont.) • Second step, the estimation of the registration parameters is fine-tuned along with the estimation of the HR image, by maximization of mutual information criterion. • Iterative scheme. • Contribution: • The registration accuracy is improved at each iteration step. • Refinement of the mutual information registration.

  13. EXPERIMENTAL PARAMETERS • Synthetic data sets. • LR images were created by rotating, translating, blurring, down-sampling and degrading by noise. • Translation: uniformly selected in [-3, 3] (in pixels) • Rotation: uniformly selected in [-5, 5] (in degrees) • Down-sampling factor: 2 (4 pixels to 1) • Blurring: 5x5 Gaussian kernel, standard deviation of 1 • Additive noise: AWGN to obtain SNR of 30 dB and 20 dB

  14. EXPERIMENTAL PARAMETERS (cont.) • First estimate of the HR image • Bicubic interpolation • Total number of realizations for each case: 10 • Convergence: or 70 iterations reached. • Quantitative evaluation: peak signal to noise ratio

  15. COMPARE METHODS

  16. EXPERIMETAL RESULTS • Books (PSNR = 26.06 dB) • 4 LR images used LR image Reconstructed HR image

  17. EXPERIMETAL RESULTS (cont.) • Front page (PSNR = 26.14 dB) • 6 LR images used LR image Reconstructed HR image

  18. EXPERIMETAL RESULTS (cont.) • Car (PSNR = 28.13 dB) • 5 LR images used LR image Reconstructed HR image

  19. EXPERIMETAL RESULTS (cont.) • Eye chart (PSNR = 27.33 dB) • 4 LR images used LR image Reconstructed HR image

  20. EXPERIMETAL RESULTS (cont.) • Statistics for the compared SR methods +1.5 dB on average better results than SIFT.

  21. CONCLUTIONS • Hybrid registration approach • SIFT-based image registration combined with the maximization of mutual information. • High precision registration • High accuracy super-resolved image. • Improvement is 1.5 dB on average higher for both 30 dB and 20 dB. • Proposed algorithm converges faster than the standard solution.

  22. QUESTIONS? ?

  23. THANK YOU ALL FOR YOUR PARTICIPATION AND PATIENCE!

More Related