IMAGE DEFOGGING METHOD BASED ON GENERATIVE ADVERSARIAL NETWORK FUSED WITH FEATURE PYRAMID

摘要

Disclosed in the present invention is an image defogging method based on a generative adversarial network fused with a feature pyramid in the technical field of image processing, aiming at solving the technical problems in the prior art that for an image processed using an image enhancement defogging method, information is lost, for an image processed using an image restoration defogging method, the effect of a restored image will be affected if parameters are selected improperly, and if a depth learning-based defogging algorithm is used, the image defogging speed will be affected. The method comprises the following steps: inputting a fog image into a pre-trained generative adversarial network, and acquiring a fog-free image corresponding to the fog image; a generator network of the generative adversarial network is fused with a feature pyramid.