A robust iterative super-resolution mosaicking algorithm using an adaptive and directional Huber-Markov regularization

• Published in: Journal of visual communication and image representation Vol. 40; pp. 98 - 110
• Main Authors: Ghosh, Debabrata, Kaabouch, Naima, Hu, Wen-Chen
• Format: Journal Article
• Language: English
• Published: Elsevier Inc 01.10.2016
• Subjects: Homography; Huber-Markov regularization; Mosaicking; Scale invariant feature transform (SIFT); Steepest descent optimization; Super-resolution; Steepest descent optimization; Mosaicking; Super-resolution; Homography; Scale invariant feature transform (SIFT); Huber-Markov regularization
• ISSN: 1047-3203; 1095-9076
• DOI: 10.1016/j.jvcir.2016.06.008

•We develop a super-resolution algorithm using directional Huber-Markov regularization.•We compare our algorithm with two other state-of-the-art algorithms.•We perform quantitative evaluation using six performance metrics. A robust spatial-domain based super-resolution mosaicking algorithm is proposed. This technique incorporates a mosaicking algorithm, and a super-resolution reconstruction algorithm. The main contribution of this paper is the development of a super-resolution algorithm using a Huber Norm-based maximum likelihood (ML) estimation in combination with an adaptive directional Huber-Markov regularization. Another contribution is the development of a no-reference performance metric based on reciprocal singular value curve for quantitative evaluation of the proposed algorithm. Along with the above-mentioned metric, five other performance measurement metrics are used to assess the efficiency of the algorithm. The performance of this algorithm is compared with the performances of two different algorithms: the Tikhonov regularization-based and the total variation (TV)-based super-resolution mosaicking algorithms. Results show that the proposed algorithm outperforms the other two techniques in terms of lowest amount of blur and noise in the output.