Creating high quality photos taken from low quality camera

• Main Author: Tay, Mei Hwee
• Other Authors: Lim Yong Ching
• Format: Final Year Project
• Language: English
• Published: 2016
• Subjects: DRNTU::Engineering
• Online Access: http://hdl.handle.net/10356/67522
• Institution: Nanyang Technological University

Image enlargement using weighted interpolation is one method that creates high quality images, with high resolution. Therefore, when images are viewed using high definition display screens, the images remain clear and visually appealing. Despite the efforts in the numerous studies done to formulate the most ideal and suitable image enlargement algorithm, there are still many limitations that needs to be considered. In this paper, an alternative method of interpolation shall be proposed to process low quality photos, to create high quality images where the clarity and sharpness of the objects in the images are preserved. By assigning weights to four pixels nearest to the point of interpolation, the interpolated value will be more accurate since closer pixels generally have similar values. Also, the sum of the weights adds to one to maintain the brightness of the image, as such only image enlargement is done, without altering other image information. In order to prove the viability of the proposed interpolation method, the nearest-neighbor interpolation and an average interpolation method were used as comparison basis. Comparisons based on human visual analysis, frequency response curves, and image quality metrics are done to justify that the proposed interpolation method creates high quality image from a low quality image.