A new auto-regressive (AR) model-based algorithm for motion picture restoration

• Published in: 1997 IEEE International Conference on Acoustics, Speech, and Signal Processing Vol. 4; pp. 2557 - 2560 vol.4
• Main Authors: Kalra, S., Chong, M.N., Krishnan, D.
• Format: Conference Proceeding
• Language: English
• Published: Washington DC IEEE 1997; IEEE Computer Society Press
• Subjects: Applied sciences; Bidirectional control; Equations; Exact sciences and technology; Image processing; Image restoration; Information, signal and communications theory; Low pass filters; Motion estimation; Motion pictures; Parameter estimation; Predictive models; Signal processing; Signal restoration; Telecommunications and information theory; Motion estimation; Autoregressive model; Image restoration
• ISBN: 0818679190; 9780818679193
• ISSN: 1520-6149
• DOI: 10.1109/ICASSP.1997.595310

This paper proposes a new AR model-based restoration algorithm which is able to suppress mixed noise processes and recover lost signals in an image sequence. A drawback of an AR model is the limiting size of the block (of pixels) that can be adequately modeled. Using a single set of AR coefficients to restore a large region of missing data will result in a homogenized texture region. In order to overcome this inherent limitation of the AR model, a block-based divide-and-conquer approach is proposed. In addition, a new Gaussian weighting scheme is used to better estimate the AR coefficients for the interpolation process.