Lightweight deep dense Demosaicking and Denoising using convolutional neural networks

A single sensor camera uses Color Filter Array (CFA) to capture single-color information at each pixel. Thus, to estimate the missing color samples and then to reconstruct an original image is known as CFA interpolation or demosaicking. Despite remarkable improvements made in the last decade, a fundamental issue remains to be addressed, i.e., how to assure the visual quality of an image in the presence of noise. Hence, the CFA images without denoising lead to the demosaicking artifacts that eventually reduce the image quality. Therefore, based on the aforementioned constraints, the paper presents a novel approach for demosaicking and denoising based on the convolutional neural network (CNN). The proposed technique is using CNN, which consists of four phases. In the first stage, the picture is sorted out. In stage-II, the demosaicking is performed utilizing the profound thick convolutional neural system, which gives us a demosaicked picture. In the stage-III, denoising performs and pass this picture to the last stage. At last, in the stage-IV, the picture goes to the last post-preparing stage delivering a better quality high-resolution image. To test the feasibility of the proposed scheme, Python language is utilized. The proposed conspire beats the few existing strategies regarding throughput delay, inactivity, precision.