Image inpainting

• Main Author: Lim, En Xu
• Other Authors: Anamitra Makur
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2023
• Subjects: Engineering::Computer science and engineering::Computing methodologies::Image processing and computer vision; Engineering::Electrical and electronic engineering
• Online Access: https://hdl.handle.net/10356/167829
• Institution: Nanyang Technological University

Image inpainting is the process of recovering lost pixels of an image. Discrete Cosine Transform (DCT) is one of the methods to do so. A casual observer`s naked eye should not be able to notice the inpainted region after the image has been processed on the damaged or missing pixels of the image. Even though Inpainting is a usual technique to recover the original image data, it is still very difficult to predict the pixels as accurate or close to original image. This project aims to design an algorithm to recover the lost pixels such that , it is best if the restored image can reappear as closely to the original image as possible without having the flawed area of the image to be too obvious to be. This project serves as a starting point and additional function can be built upon this project to make the program more reliable and useful.