A Task-Driven Invertible Projection Matrix Learning Algorithm for Hyperspectral Compressed Sensing

Shaofei Dai,Kaiyu Li,Zhengyi Wang,Wenbo Liu

doi:10.3390/rs13020295

Abstract

The high complexity of the reconstruction algorithm is the main bottleneck of the hyperspectral image (HSI) compression technology based on compressed sensing. Compressed sensing technology is an important tool for retrieving the maximum number of HSI scenes on the ground. However, the complexity of the compressed sensing algorithm is limited by the energy and hardware of spaceborne equipment. Aiming at the high complexity of compressed sensing reconstruction algorithm and low reconstruction accuracy, an equivalent model of the invertible transformation is theoretically derived by us in the paper, which can convert the complex invertible projection training model into the coupled dictionary training model. Besides, aiming at the invertible projection training model, the most competitive task-driven invertible projection matrix learning algorithm (TIPML) is proposed. In TIPML, we don’t need to directly train the complex invertible projection model, but indirectly train the invertible projection model through the training of the coupled dictionary. In order to improve the accuracy of reconstructed data, in the paper, the singular value transformation is proposed. It has been verified that the concentration of the dictionary is increased and that the expressive ability of the dictionary has not been reduced by the transformation. Besides, two-loop iterative training is established to improve the accuracy of data reconstruction. Experiments show that, compared with the traditional compressed sensing algorithm, the compressed sensing algorithm based on TIPML has higher reconstruction accuracy, and the reconstruction time is shortened by more than a hundred times. It is foreseeable that the TIPML algorithm will have a huge application prospect in the field of HSI compression.

Highlights

Spectral images with a spectral resolution in the 10-2 order of magnitude are called hyperspectral images (HSI)
Aiming at the time-consuming and low reconstruction accuracy of the reconstruction algorithm of compressed sensing technology, in the paper, we proposed a task-driven invertible projection matrix learning algorithm
On the basis of this algorithm, we study a hyperspectral compressed sensing algorithm based on a task-driven invertible projection matrix learning algorithm

Summary

Introduction

Spectral images with a spectral resolution in the 10-2 order of magnitude are called hyperspectral images (HSI). One of the main reasons for the higher visibility of hyperspectral imaging is the richness of the spectral information collected by this sensor. This function has positioned the hyperspectral analysis technology as the mainstream solution for land area analysis and the identification and differentiation of visually similar surface materials. Hyperspectral image processing is accompanied by a large amount of data management, which affects real-time performance on the one hand, and, on the other hand, the demand for on-board storage resources. The latest technological advances are introducing hyperspectral cameras with a higher spectrum and spatial resolution to the market.

Methods

Results

Conclusion