Super-resolution image processing method based on probability generation model

Abstract

The invention discloses a super-resolution image processing method based on a probability generation model. The method comprises: (1), inputting an image; (2), obtaining a training sample; (3), obtaining a testing sample; (4), training a probability generation model; and (5), testing the probability generation model. According to the invention, compared with the common method, the super-resolutionimage processing method enables the generated high-resolution image to have higher resolution than that generated by the common method and to have more information; and with the prior information ofthe probability generation model and information implied by a hidden variable, the image super-resolution speed is increased. The method is a high-efficiency super-resolution image processing method.

Description

technical field [0001] The invention belongs to the technical field of image processing, and further relates to an image super-resolution processing method based on a probability generation model in the technical field of image super-resolution processing. The present invention can be used to first generate corresponding high-resolution image blocks by using low-resolution image blocks, then generate high-resolution images corresponding to the original low-resolution images, and then regenerate high-resolution images corresponding to the original low-resolution images. Background technique [0002] Super-resolution optical images can overcome the limitations of low-resolution optical images and have shown promising results in many applications such as medical diagnosis, remote sensing, computer vision, and surveillance. To obtain high-resolution optical images, the most direct way is to use high-resolution image sensors, but due to the limitations of the manufacturing proces...

AI Technical Summary

Problems solved by technology

However, there are still deficiencies: this method does not take into account the prior information of the model, nor does it consider the uncertain factors implied by the hidden variables in the model, so that the method cannot use all the information contained in the original image. Only part of the information is used to generate the corresponding high-resolution image

However, there are still shortcomings: this method only uses a shallow probability model, and only uses the information on the surface of the low-resolution image, and does not take into account the hidden information in the high-resolution / low-resolution image block pair, where The methods involved cannot generate more information to improve the resolution of the image, limiting the resolution of the final high-resolution image

Images