Natural image compressed sensing reconstruction method based on deep sparse coding

Abstract

The invention discloses a natural image compressed sensing reconstruction method based on deep sparse coding, which is mainly used to solve the problem that it is difficult to reconstruct a natural image with coefficients fast and accurately for the existing method. The method comprises the following steps: (1) segmenting an image and transforming the image in an orthogonal transformation domain, and calculating the observation vector of a transformation coefficient; (2) getting the restoration transformation coefficient of the observation vector through an iterative threshold method, and updating calculation parameters; (3) calculating the observation vector of the transformation coefficient in step (2), and getting the residual between the observation vector and the observation vector in step (1); (4) repeating the steps (1)-(3) to get a trained model, and saving model parameters; (5) inputting test observation data and the model parameters to the trained model to get an image transformation coefficient corresponding to the test observation data; and (6) inversely transforming the transformation coefficient in step (5) to get a final reconstructed natural image. The natural image reconstructed in the invention is clear, and the reconstruction is very fast. The method can be used to restore a natural image.

Description

technical field [0001] The invention belongs to the field of image processing, and in particular relates to a natural image compression sensing reconstruction method, which can be used for sampling natural image restoration. Background technique [0002] With the development of media technology, massive image data is facing huge challenges in real-time transmission and storage. The introduction of compressed sensing technology has opened up new ideas for these problems in theory, and effectively solved the problems. Compressed sensing theory believes that if the signal has sparsity under a certain transformation base, the signal can be randomly projected and observed, and it can be accurately reconstructed through the prior information of the signal with fewer observations. The model is a norm optimization problem under the fidelity constraints of solving observed data. [0003] For the above-mentioned compressed sensing model, different norm constraints represent differen...

AI Technical Summary

Problems solved by technology

Although the above algorithms have further solved the problem, they are all based on optimization theory, and there are still many problems such as limited reconstruction accuracy, complex iterative process, and slow convergence speed.

[0004] In recent years, with the development of deep learning technology, some scholars began to propose Recon-Net, a compressed sensing reconstruction method based on convolutional neural network. Although this method solves the problem of complex iterative process, there are still two problems: 1) The reconstructed image obtained by directly applying the model is noisy, and the reconstructed image with better quality must be denoised; 2) The convergence speed of the model is slow during training. Even on a high-performance computer, it takes a day to train the model. time

Images