The output image of a digital camera is subject to a severe degradation due to noise in the image sensor. This paper proposes a novel technique to combine demosaicing and denoising procedures systematically into a single operation by exploiting their obvious similarities. We first design a filter as if we are optimally estimating a pixel value from a noisy single-color image. With additional constraints, we show that the same filter coefficients are appropriate for CFA interpolation (demosaicing) given noisy sensor data. The proposed technique can combine many existing denoising algorithms with the demosaicing operation. In this paper, a total least squares denoising method is used to demonstrate the concept. The algorithm is tested on color images with pseudo-random noise and on raw sensor data from a real CMOS digital camera that we calibrated. The experimental results confirm that the proposed method suppresses noise (CMOS image sensor noise model) while effectively interpolating the missing pixel components, demonstrating a significant improvement in image quality when compared to treating demosaicing and denoising problems independently.
Joint demosaicing and denoising
IEEE International Conference on Image Processing 2005 ; 3 ; III-309
2005-01-01
466300 byte
Conference paper
Electronic Resource
English
Joint Demosaicing and Denoising
British Library Conference Proceedings | 2005
|Low-Complexity Linear Demosaicing using Joint Spatial-Chromatic Image Statistics
British Library Conference Proceedings | 2005
|Fast Algorithms for Single Chip Video Demosaicing
British Library Conference Proceedings | 2006
|