Weight optimization for multiple image integration
Ryo Matsuoka, Tomohiro Yamauchi, Tatsuya Baba, Masahiro Okuda · 2013
We propose a denoising technique using multiple image integration. When acquiring a dark scene, the detail of the dark area is often deteriorated by sensor noise. A simple image integration inherently has the capability of reducing random noises. In this paper we develop the denoising performance of the multiple image integration by optimizing weight maps. We determine the optimal weight by solving a convex optimization problem. Through some experimental results, we show the weight optimization significantly improves the de-noising performance.