Abstract:Requirements of compressed sensing techniques targeted at miniaturized hyperspectral satellite applications include lightweight onboard hardware, high-speed sensing, low sampling rate for compressing the massive volume of typical hyperspectral data, and noise robustness for reliable data transmission to the ground station. We achieve all these aims via deep learning, and neural networks resulted from which can be implemented on-chip, thereby allowing light hardware implementation. Our neural networks were trained from small-scaled data, but, even so, the resulting encoder achieves a very low sampling rate and very high speed. Unlike typical network training, the input–output pairs are not square but stripe-like images, partly because compressed acquisition does not allow performing compression after obtaining complete data cube and partly because stripe-like acquisition well matches the popular pushbroom hyperspectral sensing schemes. Even with such hard restriction caused by nontraditional training, the resulting decoder still reconstructs the image with high accuracy. To match the requirement of pushbroom sensing, a lightweight encoder is proposed to compress the stripe-like images immediately. Meanwhile, multiscale feature fusion block (MFB) and aggregation (MFA) modules are proposed to form our decoder for enhancing the feature representation of the compressed acquisitions. Furthermore, we achieve joint spatial/spectral super-resolution (SR) progressively, ensuring accurate hyperspectral reconstruction via a low-rank-driven decoder. The encoder and decoder are trained in an end-to-end manner, where noise robustness is forced during the training stage. Comprehensive experiments demonstrate the superiority of the proposed hyperspectral compressed sensing method, as well as its one-shot transfer learning (OTL)-based extension, both quantitatively and qualitatively.

Tensor-Train Based Deep Learning Approach for Compressive Sensing in Mobile Computing

Deep Learning Approach Based on Tensor-Train for Sparse Signal Recovery

Understanding Sensor Data Using Deep Learning Methods on Resource-Constrained Edge Devices.

Deep Model Compression for Mobile Platforms: A Survey

On-Demand Deep Model Compression for Mobile Devices

A Dictionary Learning Based Unsupervised Neural Network for Single Image Compressed Sensing

From Patch to Pixel: A Transformer-Based Hierarchical Framework for Compressive Image Sensing.

The Optimally Designed Autoencoder Network for Compressed Sensing.

Scalable Deep Compressive Sensing

A Lightweight Recurrent Learning Network for Sustainable Compressed Sensing

Generalized Tensor Summation Compressive Sensing Network (GTSNET): An Easy to Learn Compressive Sensing Operation

Compressed Tensor Completion: A Robust Technique for Fast and Efficient Data Reconstruction in Wireless Sensor Networks

Efficient Compressed Sensing for Wireless Neural Recording: A Deep Learning Approach

ConvCSNet: A Convolutional Compressive Sensing Framework Based on Deep Learning

Tree-Structured Dilated Convolutional Networks for Image Compressed Sensing

Distributed Compressive Sensing: A Deep Learning Approach

DCSN: Deep Compressed Sensing Network for Efficient Hyperspectral Data Transmission of Miniaturized Satellite

ADMM-CSNet: A Deep Learning Approach for Image Compressive Sensing

Multi-Scale Deep Networks for Image Compressed Sensing

Classification Guided Deep Convolutional Network For Compressed Sensing

Compressed Sensing via a Deep Convolutional Auto-encoder