Transform-based single image defogging method and aerial photography fixed-wing unmanned aerial vehicle

The invention discloses a Transform-based single image defogging method and an aerial fixed-wing unmanned aerial vehicle, and belongs to the technical field of digital image processing, and the method comprises the steps: building a foggy-fogless image data set; the method comprises the following steps of: constructing an SCU DehazeFormer network model; and training the network model by using a foggy-fogless data set, obtaining a trained network model after the training is completed, inputting a foggy image into the trained network model, and finally outputting a defogged clear image. A foggy-fogless data set is adopted, an ideal processing result can be achieved for a single-image defogging task, a new SCUDehazeFormer network model is adopted, some negative effects which are not important to a high-level vision task but are crucial to a low-level vision task are avoided, and the method has the advantages of being high in robustness and high in robustness. A moving window division scheme based on reflection f