Abstract: The image quality of laser active imaging system degraded dramatically in fog. The obtained infrared image had the features of low contrast and serious noise. Thus an image process method combining dark channel prior and bilateral filter was proposed. The dark channel prior was adopted to remove fog from the image. Then the image was denoised using the bilateral filter. So the contrast of the image was improved and the noise was reduced. Both subjective and objective evaluation of the processed image by different methods show that the proposed method could obtain better visual effect. The standard deviation is more than doubled. The information entropy and information capacity have about a twenty percent improvement. Thus, it's an effective way to improve the quality of infrared image in fog.