An Effective Algorithm for Removing Jaggy Artifact Using Directional Blurring

• Main Authors: Ching-YaoChao, 趙勁堯
• Other Authors: Shen-Chuan Tai
• Format: Others
• Language: en_US
• Published: 2014
• Online Access: http://ndltd.ncl.edu.tw/handle/01283610212213566287

碩士 === 國立成功大學 === 電腦與通信工程研究所 === 102 === With the advance of science and technology, the need for the visual quality of images is getting higher. Therefore, technology of super-resolution becomes a hot research topic. The main purpose of up-scaling is to obtain high-resolution images from low-resolution images, and the goal is to make the result of the upscaled image looks like natural image. This technique is also known as super-resolution. It had been widely used in high definition televisions, smart phones, satellite images and surveillance cameras. In general, the popular convolution-besed methods usually induce blurring artifacts and jaggy artifacts along slant edges. Therefore, in order to solved these problems, the proposed algorithm first removes the jaggy artifacts, and then adopts the regression model established from an LR image to reconstruct an HR image. Experimental results show that the proposed algorithm produces HR images with better visual quality.