Double Laplacian pyramid and convolutional neural network-based image amplification method

Abstract

The invention discloses a double Laplacian pyramid and convolutional neural network-based image amplification method. The method comprises the steps of obtaining a low-resolution image, and performingbicubic amplification processing to obtain a blurred high-resolution image with a corresponding amplification factor; performing feature extraction processing on the blurred high-resolution image toobtain image features of different scales from large to small; performing feature extraction processing on a low-resolution image to obtain image features of different scales from small to large in layers; and fusing the obtained image features of the different scales from large to small and image features of the different scales from small to large in the layers by reconstructing a super-resolution network to obtain a high-resolution image. By implementing the method, the high-quality high-resolution image can be reconstructed more quickly; the image quality is improved by reconstructing thelow-resolution image; and therefore, the method can be better applied to more scenes.

Description

technical field [0001] The invention relates to the technical fields of machine vision and super-resolution reconstruction, in particular to an image enlargement method based on a double Laplacian pyramid convolutional neural network. Background technique [0002] With the continuous development of electronic information science and the continuous popularization of digital products, while the ability to acquire images is continuously enhanced, the ability to display images is also increasing. The resolution of display devices has been continuously improved, but limited by the resolution of imaging devices in the past, 'old photos' and 'old movies' cannot be displayed well on existing high-definition displays. At the same time, it is difficult to obtain all the information of the displayed scene due to the limitations of imaging equipment and imaging conditions in the generation of remote sensing images and medical CT images. The needs of all the above problems can be attrib...

AI Technical Summary

Problems solved by technology

The learning-based method can reconstruct a high-resolution image at a relatively good price, but it takes a lot of time and calls, and it needs to be learned for each image, and the reconstruction time is long

The current deep learning-based methods have problems such as too many network layers that make it difficult to train, and require a large number of iterations and sample data.

Images