Study on Noise Reduction Method of low Illuminance Image Based on Noise Feature

In the field of image processing, it is often necessary to separate the background and foreground of image for the application for image editing, moving-object recognition, machine learning and so on. In image recognition under low illumination, the random noise is so intense relative to the intensity of the signal. So "How to deal with the noise" is important problem. Therefore, in this research, we propose the method of separating target and noise instead of removing noise. It can be seen from the results of the proposed method that the separation of target and noise is possible.