Image super-resolution reconstruction method based on multi-band deep convolutional neural network

Abstract

The invention discloses an image super-resolution reconstruction method based on a multi-band deep convolutional neural network, comprising the following steps: selecting training samples and test samples; inputting training images to the network in batches for feature extraction, and combining long-term and short-term memories to construct a multi-band learning structure, then performing featurerefinement, feature mapping and up-sampling reconstruction; and obtaining network parameters according to the model obtained through training to complete the image reconstruction. The invention enhances high-frequency information reconstruction by introducing multi-band learning, and uses the memory migration operation to make the network have long-term and short-term memories at the same time, which speeds up the image reconstruction speed, enhances image edge and texture detail reconstruction, and obtains image super-resolution reconstruction results of better quality. The invention has strong super-resolution capability, and the reconstructed image is closer to the real image.

Description

technical field [0001] The invention belongs to the field of image technology, and in particular relates to a method for super-resolution reconstruction of a single image based on a multi-band deep convolutional neural network. Background technique [0002] In fields such as medicine, astronomy, monitoring, and military affairs, high-resolution images are of great significance. However, the current traditional image super-resolution technology cannot reconstruct better high-resolution images, and the restoration of texture details and edge features needs to be improved. In recent years, image super-resolution reconstruction methods based on convolutional neural networks have more advantages than traditional methods and have received extensive attention. [0003] Dong et al. first proposed a convolutional neural network-based image super-resolution reconstruction method SRCNN, which greatly improved image resolution by using a three-layer convolutional neural network. Then t...

AI Technical Summary

Problems solved by technology

[0004] The existing image super-resolution reconstruction methods based on convolutional neural networks can effectively improve the performance of image super-resolution reconstruction, but they have the following defects: these methods use the low-frequency information of the image to reconstruct high-frequency information, and the computational complexity is high. The recovery of image details is still not enough, and the edge and texture details of the reconstructed image are not clear enough to meet the needs of reality.

Moreover, due to the network structure, the training is difficult, it is difficult to achieve convergence, the phenomenon of gradient disappearance is prone to occur, and the flexibility is low.

Images