Workpiece image defogging method and device applied to ion beam precise coating

Abstract

The embodiment of the invention provides an image defogging method and device and electronic equipment, and the method comprises the steps: obtaining a to-be-defogged image and a blind defogging model; inputting the to-be-defogged image into a neural network to obtain a blurred kernel of the to-be-defogged image, the loss function of the neural network being obtained according to the blind defogging model, the blind defogging model comprising dark channel prior; substituting the fuzzy kernel into a blind de-atomization model to obtain a non-blind de-atomization model; and according to the non-blind defogging model, defogging the to-be-defogged image to obtain a clear image. By adopting the technical scheme provided by the embodiment of the invention, when the image is subjected to defogging, the universality of the blind defogging model can be improved, so that the accuracy of image defogging is improved.

Description

technical field [0001] The present invention relates to the technical field of image processing, in particular to an image defogging method, device and electronic equipment. Background technique [0002] During the image capturing process, due to the light problem of the shooting environment or the problem of the object to be shot, it is easy to make the captured image blurred. In many cases, when the captured image is blurred, it is necessary to perform defog processing on the captured image to obtain a clear image. For example, in the process of ion beam sputtering coating, it is necessary to take images of workpieces to be processed, and position the workpieces to be processed according to the captured images, so as to adjust the positions between the ion source, the workpieces to be processed and the target platform, so that the sputtering The ions can be aimed at the workpiece to be processed, but because the light in the operating room of the ion beam sputtering coati...

AI Technical Summary

Problems solved by technology

[0004] However, when using the existing technology to defog the image, since most of the priors of the blur kernels are statistically obtained, this will lead to poor versatility of the blind defog model, making the image defog is also less accurate

Images