Abstract: To resolve the problem that the traditional de-noise methods will blur the detail of image, this paper introduces a new method to de-noise the image noise in low light. The method reduces the noise with the advantage of human eyes, according to the different noise level with the corresponding illuminance. The function and flow chart is shown and parametric of function is decided by experiment. And the result of experiment shows that the method reduce the noise and the detail of the image is not blurred.

Key words: auto color, noise reduction, low light, exposure, gain

摘要： 针对传统去噪方法会导致图像细节模糊的问题，提出一种低照度条件下的图像自动降色去噪方法。结合人眼的生理特点，根据环境照度所对应噪声的大小，进行不同程度的噪声降低处理。介绍函数关系及流程，并通过实验数据确立函数值。结果表明，该方法能在不模糊图像细节的基础上有效降低噪声。

关键词: 自动降色, 去噪, 低照度, 曝光, 增益

CLC Number: 

• TN911.73