The invention discloses a simulation fog image generation method based on depth priori, and mainly solves the problems in the prior art that a simulation fog adding method does not reasonably combine depth information. The technical scheme comprises the steps that 1) the sky area in a no-fog image is separated so that an initial sky area is obtained; 2) the initial sky area is corrected based on edge detection; 3) the depth map of the no-fog image is constructed by using the positional relationship of the corrected sky area and the non-sky area, and the transmittance of each point is calculated; and 4) a simulation fog image after artificial fog adding in the no-fog image is established by using the depth map and the transmittance. The method has the advantages that the sky area of the image is accurately separated and the fog adding process meets the natural law and is genuine and believable so that the method can be used for constructing the depth map for the no-fog image photographed in an outdoor long-distance scene, evaluating the defogging effect, detecting the performance of equipment and performing defogging network model training.