Robust super-resolution image generation

• Main Author: Fatin Nazurah Roslan
• Other Authors: Lin Weisi
• Format: Final Year Project
• Language: English
• Published: 2019
• Subjects: Engineering::Computer science and engineering
• Online Access: http://hdl.handle.net/10356/78982
• Institution: Nanyang Technological University

Since the beginning of the 1960s, different techniques of digital image processing were developed. Improving and evolving these technologies over the years have been done. Super-Resolution reconstruction was introduced to produces high-quality images to get a better visual of an image. As a result, obtaining this outcome is significant for our project to prove that Super-Resolution improves and provides a better vision of an image. Thus, it can assist in daily application like identifying an object used for accident prevention used in an autonomous vehicle which is briefly explained in this project. The project aims to prove that whether the development of one technique, Super-Resolution, gradually produces a better and higher quality image as compare to image without Super-Resolution. Significantly, improves the visual image for better viewing, human interpretation, and machine perception. To handle this situation, Super-Resolution was used to observe the difference between an image before and after super-resolution is applied. Furthermore, to prove that the image quality increases after Super-Resolution reconstruction. Consequently, this will improve the visual representation of an image for better viewing which can help in daily application like object detection for easier identifying.