Motion blur concealment of digital video using invariant features

Authors:
Ville Ojansivu;Janne Heikkilä
Affiliations:
Machine Vision Group, Department of Electrical and Information Engineering, University of Oulu, Finland;Machine Vision Group, Department of Electrical and Information Engineering, University of Oulu, Finland
Venue:
ACIVS'06 Proceedings of the 8th international conference on Advanced Concepts For Intelligent Vision Systems
Year:
2006

Citing 6
Cited 2

Degraded Image Analysis: An Invariant Approach

IEEE Transactions on Pattern Analysis and Machine Intelligence
Video Processing and Communications

Video Processing and Communications
Moment Forms Invariant to Rotation and Blur in Arbitrary Number of Dimensions

IEEE Transactions on Pattern Analysis and Machine Intelligence
Moment relations and blur invariant conditions for finite-extent signals in one, two and N-dimensions

Pattern Recognition Letters
Full-Frame Video Stabilization

CVPR '05 Proceedings of the 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05) - Volume 1 - Volume 01
Multichannel blind deconvolution of spatially misaligned images

IEEE Transactions on Image Processing

Weighted DFT Based Blur Invariants for Pattern Recognition

SCIA '09 Proceedings of the 16th Scandinavian Conference on Image Analysis
Object recognition using frequency domain blur invariant features

SCIA'07 Proceedings of the 15th Scandinavian conference on Image analysis

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper deals with concealment of motion blur in image sequences. The approach is different from traditional methods, which attempt to deblur the image. Our approach utilizes the information in consecutive frames, replacing blurred areas of the images with corresponding sharp areas from the previous frames. Blurred but otherwise unchanged areas of the images are recognized using blur invariant features. A statistical approach for calculating the weights for the blur invariant features in frequency and spatial domains is also proposed, and compared to the unweighted invariants in an ideal setting. Finally, the performance of the method is tested using a real blurred image sequence. The results support the use of our approach with the weighting scheme.