Compressed Sensing Network and Whole Image Reconstruction Method Based on Block Observation

Abstract

The invention discloses a block-based observation-based compressed sensing network and a full-image reconstruction method, which mainly solves the problem of poor image quality restored by the existing network. Its network includes an observation subnetwork and a reconstruction subnetwork. The observation subnetwork consists of the first convolutional layer; the reconstruction subnetwork includes a deconvolutional layer, a second convolutional layer, and 12 residual blocks of the same size. , the fifth convolutional layer and the sixth convolutional layer; each residual block includes the third convolutional layer and the fourth convolutional layer; the output of the first convolutional layer is connected to the input of the deconvolution layer, and the reverse The output of the convolution is connected to the input of the second convolutional layer, and the output of the second convolution is connected to 12 residual blocks in turn, and the output of the 12th residual block is connected to the input of the fifth convolutional layer The terminals are connected, and the output terminal of the second convolution and the output terminal of the fifth convolution are commonly connected to the input terminal of the sixth convolution. The network of the invention avoids the block effect of the reconstructed image, improves the image restoration quality, and can be used for image processing.

Description

technical field [0001] The invention belongs to the technical field of compressed sensing, and mainly relates to a compressed sensing network for sub-block observation and a full-image reconstruction method, which can be used for image processing. Background technique [0002] In a large number of practical problems, people tend to collect as little data as possible, or have to collect incomplete data due to objective conditions. Traditional image compression is based on Nyquist sampling for data collection, and starts from the characteristics of the data itself to find and eliminate the hidden redundancy in the data. The result of this is that data compression must be done after the data is completely collected, and the compression process requires complex algorithms, which is contradictory to the performance of equipment that collects and processes a large number of signals. The concept of compressed sensing is proposed to solve this problem. It can directly collect the c...

AI Technical Summary

Problems solved by technology

This method of dividing a large image into small images will cause the reconstructed image to have obvious block effects, the image recovered at a low observation rate is blurred, the semantic information is not obvious, and it is necessary to ensure a large amount of training data set, the training process takes a long time, and is not suitable for image reconstruction at low observation rates

Images