Near-infrared radiation is an important component of night sky light. Making optimal use of near-infrared radiation is an effective method to improve the signal-to-noise ratio of true-color night vision imaging systems. However, green plants with high reflectance of near-infrared radiation introduce a strong color cast problem in true-color night vision images. In this study, we analyze the gray value distribution of green plant images in red, green, and blue channels and find that the depth of field of the green plant area is large. The difference between the all-pass channel and the red channel is the sum of the green and blue components, without the near-infrared radiation. In addition, the green and blue components of the green plant region are few in the difference image. Based on these two characteristics, in this study, we propose a color correction method for true-color night vision images. The method extracts green plants with differential depth of field, then uses the extracted green plant area to adjust the values of the red and blue channels, according to the color balance principle. The results show that the proposed method can effectively correct the influence of near-infrared radiation on image color and restore the image color of green plants in night vision images.