Digital Image Processing

S. Arumugan ¹, K. Senthamarai Kannan ², K. John Peter ³

Affiliations
1 Nandha Engineering College, Tamilnadu, IN
2 Department of Statstics, MS University, Tamilnadu, IN
3 Department of Information Technology, VINS Christian College of Engineering, Tamilnadu, IN

Abstract

In this paper one new compactive demosaicing Algorithm is used for color filter array (CFA). In all the existing algorithm they detect the edges in horizontal-, vertical- or Omni-direction. In the existing algorithm does not detect diagonal edges. We proposed a new approach of similarity-based demosaicing algorithm using unified high-frequency (UHF) map. Similarities between pixels are calculated on a local map called UHF map. A missing sample must be interpolated from the neighboring samples that are highly correlated with the missing sample. This highly correlated neighboring samples are detected by similarity calculation. It is, thus, expected that the proposed algorithm is able to deal with edges of any direction such as diagonal edges. The proposed algorithm improve PSNR and image quality. Again, the proposed algorithm requires fewer resources.

Keywords

Color Filter Array, Demosaicing, Edge-Directed Interpolation, Unified High-Frequency Map.

Full Text

     

K. John Peter ¹, G. Nagarajan ¹, S. Arumugan ², K. Senthamarai Kannan ³

Affiliations
1 VINS Christian College of Engineering, IN
2 Nandha Engineering College, Erode, IN
3 M.S University, IN

Abstract

Most cost-effective color cameras have single image sensors with a fixed repeating pattern of color filters with transmission spectra analogous to the human eye. To reconstruct a full color image a process often referred to as “demosaicking” is necessary. Almost every past method for demosaicking had the intention to create the most pleasant image for the human observer. The quality of demosaicked images is degraded due to the sensor noise introduced during the image acquisition process. According to the proposed Demosaicking algorithm we can extract edge information in terms of the direction of variation and the gradient from the mosaic image directly and accurately, and the extracted more accurate edge information will be utilized to assist the design of our proposed new demosaicking algorithm. The proposed algorithm is also capable of reducing the interpolation errors which are visually objectionable because they tend to correlate with object boundaries and edges. Proposed algorithm find out the missing color components in the mosaic images captured by the CFA and improve the visual quality of the resulting output.

Keywords

Demosaicking, Gradient, Mosaic Image, Correlation, Mosaic, CFA.

Full Text