An improved algorithm of image registration based on least squares adjustment. Issue 349 (3rd July 2016)

• Record Type: Journal Article
• Authors: Zhou, C.; Zhu, J.; Fan, D.; Zhou, J.
• Is Part Of: Survey review. Volume 48:Issue 349(2016)
• Page Start: 240
• Page End: 246
• Publication Date: 2016-07-03
• Subjects: Least squares adjustment -- Image registration -- Super-resolution image reconstruction; Surveying -- Periodicals; Great Britain -- Surveys -- Periodicals; 526.9
• DOI: 10.1179/1752270615Y.0000000025
• Languages: English
• ISSNs: 0039-6265

Abstract

Abstract : Image registration is one of the most critical steps in super-resolution (SR) image reconstruction. Correspondingly, it is of considerable importance to improve the accuracy of image registration. Here, we report a novel least squares adjustment theory to significantly improve the image registration algorithm, which combines both the advantages of Vandewalle algorithm and Keren algorithm. Compared with traditional Vandewalle algorithm, Keren algorithm and Li algorithm, the experimental results show that the registration parameters exhibit the highest accuracy in our improved algorithm. Meanwhile, the runtime of the improved algorithm is much faster than that of Keren algorithm and Li algorithm, and the peak signal-to-noise ratio (PSNR) value of the reconstructed image is 1·5 dB higher than that of Vandewalle algorithm and Keren algorithm.