Insulator corona discharge data enhancement method and system based on ultraviolet image

The invention discloses an insulator corona discharge data enhancement method and system based on an ultraviolet image, and the method comprises the steps: obtaining an ultraviolet image of insulator corona discharge, marking an insulator and a discharge region in the ultraviolet image through employing a polygonal frame according to a label tool, and obtaining a marked target ultraviolet image; and inputting the target ultraviolet image into a pre-constructed generative adversarial network model, wherein the adversarial network model outputs a virtual ultraviolet image corresponding to the target ultraviolet image. A generative adversarial network loss function is optimized, and strong learning region loss is added into the generative adversarial loss function to realize local supervised learning, so that a network model can also obtain a good effect when being trained on a small data set; by adjusting the strong learning weight in the loss function, the supervised learning participation weight can be contro