Image restoration based on weighted average of multiple blurred and noisy images
Ryo Tanikawa, Takanori Fujisawa, Masaaki Ikehara · 2018 International Workshop on Advanced Image Technology (IWAIT) · 2018
In this paper, we propose a new method for image restoration from a pair of images with different noise and blur artifacts. These images are obtained from the camera with different exposure time and the restored images have higher quality. Some restoration methods using multiple degraded images have been proposed. Most of these methods solve the optimization problem achieving the noise removal and the blur suppression at the same time. However, this approach cannot handle the degree of noise removal and blur suppression easily. This paper proposes a new method for image restoration from a pair of images with different noise and blur artifacts. We take a wighted average of the two images to produce one image for the restoration process. By merging the noisy image, the noise and blur artifact can be efficiently suppressed while keeping useful image information. Then we propose a simple restoration method and obtain a higher quality restored image. Experiment results show that the proposed method can obtain a higher quality restored images which are removed noise and preserved edges.