مشروع البحث: A CONVOLUTIONAL NEURAL NETWORK FOR MOTION-BASED MULTIFRAME SUPER-RESOLUTION USING FUSION OF INTERPOLATED FRAMES
| dc.contributor.advisor | Dr. Russell Hardie | |
| dc.date.accessioned | 2026-06-15T07:06:58Z | |
| dc.date.available | 2026-06-15T07:06:58Z | |
| dc.description | we present a novel motion-based multiframe image super-resolution (SR) algorithm using a convolutional neural network (CNN) that fuses multiple interpolated input frames to produce an SR output. We refer to the proposed CNN and associated preprocessing as the Fusion of Interpolated Frames Network (FIFNET). We believe this is the first such CNN approach in the literature to perform motion-based multiframe SR by fusing multiple input frames in a single network. We study the FIFNET using translational interframe motion with both fixed and random frame shifts. | |
| dc.description.abstract | Single frame Super-resolution (SR) is another area receiving significant attention in the literature. With the absence of additional spatial samples, single frame methods rely more heavily on exploiting prior statistical information to address the ill-posed inverse problem of interpolating aliased imagery. Notwithstanding the significant challenge, several approaches have demonstrated performance far superior to simple interpolation. Recently, a popular approach to single frame SR has involved the use of convolutional neural networks (CNNs). Here prior information is learned by training the CNNs on large image databases of highresolution (HR) images. | |
| dc.identifier | 1340 | |
| dc.identifier.uri | https://dspace.academy.edu.ly/handle/123456789/2315 | |
| dc.subject | INTERPOLATED FRAMES | |
| dc.title | A CONVOLUTIONAL NEURAL NETWORK FOR MOTION-BASED MULTIFRAME SUPER-RESOLUTION USING FUSION OF INTERPOLATED FRAMES | |
| dspace.entity.type | Project | |
| project.endDate | 2023 | |
| project.funder.name | هندسة كهربائية | |
| project.investigator | حامد الهادي فرج الورفلي | |
| project.startDate | 2022 |