期刊名称：International Journal of Computer Science & Technology
印刷版ISSN：2229-4333
电子版ISSN：0976-8491
出版年度：2015
卷号：6
期号：4
页码：75-80
语种：English
出版社：Ayushmaan Technologies
摘要：Image inpaintingdeals with the issue of filling-in missing regions in an image. Inpainting images with distortion, damage or corruption is a challenging task. Most existing algorithms are pixel based, which develop a statistical model from image characteristics. One of the primary burdens of these methodologies is that, their viability is constrained by the surrounding pixels of the destroyed part. Subsequently, good performance of these strategies is acquired just when the images have particular consistency. Images in the frequency domain contain sufficient data for image in painting and can be utilized as a part of data recreation e.g., high frequency indicates image edges or textures, which motivates conducting image inpainting in the frequency domain. In this paper, we use KNN method which will utilize the DCT coefficients in the frequency domain to remove the distortions. We look for an adequate representation for the functions and utilize the DCT coefficients of this representation to produce an over-complete dictionary.The two main objective of this paper is inpainting and compressing images with noise (after denoising/inpainting). This paper analyzes coding algorithm of JPEG image and proposes a K-Nearest Neighbor (KNN) approach to perform inpainting in the DCT Coefficients to get a more optimized compression ratio. The proposed methodology is expected to outperform the compression ratio of the Baseline JPEG Algorithm dealing with images having cracks and distortions. The reason behind this is that images having distortions will have anomalies in the distorted parts which will contribute to the size of the image. If those distortions are removed before compression, the output will be more optimized.
