There is a lack of visibility for low-light image, for solving this problem, put forwards the image enhancement method based on eye intensity response. The guided filter is used to calculate the partial mask by the inverse version of luminance component. With global average value and standard value of luminance channel to calculate key parameter, carry out adaptive curve adjustment for image luminance-channel and improve dynamic range. A color correction is carried out by calculating enhanced luminance channel color saturation gain and offset. Depend on the grey histogram distribution to further stretch the global contrast. Use SSIM, VLD and MOS evaluation index to compare different image enhancement methods. The test results show that the algorithm has good enhancement effect on low-light images, can effectively enhance image brightness and details, and improve visibility of low-light image.