STM Journals Peer-reviewed research from Consortium e-Learning Network

Journal of Image Processing & Pattern Recognition Progress

Single Sensor Camera Image Denoising Using PCA and Demosaicking with LMMSE

  1. Arti Noor
  2. Y.K. Jain
  3. Ashok Saini

Abstract

Single sensor cameras are used mostly for image acquisition in digital cameras with a colour filter array (CFA). These cameras save image in an unrefined format, since the complete process of image acquisition get lengthy and complex, so it introduces noise in the image and hence requires some efficient technique for denoising. Different techniques have been used to denoise the captured image but efficiency of the algorithms has not been up to the mark because of various limitations. This paper introduces a method for denoising in which the 2D image component is decomposed in high-pass and low-pass component, high-pass component is then denoised using principal component analysis (PCA), and combined with the low-pass component which is then demosaicked and converted to RGB image format. Demosaicking is done through linear minimal mean square error (LMMSE) estimation process. The performance of the system can be estimated with the help of signal-to-noise ratio (SNR) analysis and visual quality.Cite this ArticleNoor A, Jain YK, Saini A. Single Sensor Camera Image Denoising Using PCA and Demosaicking with LMMSE. Journal of Image Processing & Pattern Recognition Progress. 2016; 3(2): 1–10p.

Keywords

Support