Abstract:In the compressive spectral imaging (CSI) framework, different architectures have been proposed to recover high-resolution spectral images from compressive measurements. Since CSI architectures compactly capture the relevant information of the spectral image, various methods that extract classification features from compressive samples have been recently proposed. However, these techniques require a feature extraction procedure that reorders measurements using the information embedded in the coded aperture patterns. In this paper, a method that fuses features directly from dual-resolution compressive measurements is proposed for spectral image classification. More precisely, the fusion method is formulated as an inverse problem that estimates high-spatial-resolution and low-dimensional feature bands from compressive measurements. To this end, the decimation matrices that describe the compressive measurements as degraded versions of the fused features are mathematically modeled using the information embedded in the coded aperture patterns. Furthermore, we include both a sparsity-promoting and a total-variation (TV) regularization terms to the fusion problem in order to consider the correlations between neighbor pixels, and therefore, improve the accuracy of pixel-based classifiers. To solve the fusion problem, we describe an algorithm based on the accelerated variant of the alternating direction method of multipliers (accelerated-ADMM). Additionally, a classification approach that includes the developed fusion method and a multilayer neural network is introduced. Finally, the proposed approach is evaluated on three remote sensing spectral images and a set of compressive measurements captured in the laboratory. Extensive simulations show that the proposed classification approach outperforms other approaches under various performance metrics.

Classification-based image-fusion framework for compressive imaging

Region-Based Image-Fusion Framework for Compressive Imaging

Hybrid Compressive Spectral Imaging with Coded Aperture Optimization Based on Coherence Minimization

Multi-Frame Image Fusion Method Combining Spatial-Temporal Saliency Detection and Nsct

Dual Band Image Fusion Method Based on Region Saliency

Image fusion in compressed sensing

PCNN-Based Image Fusion in Compressed Domain

Feature Fusion via Dual-resolution Compressive Measurement Matrix Analysis For Spectral Image Classification

Gradient-based Compressive Image Fusion

ICIF: Image fusion via information clustering and image features

A pixel-level image fusion framework based on structural similarity

A Medical Image Fusion Method Based on SIFT and Deep Convolutional Neural Network in the SIST Domain

Compressive sensing image fusion in heterogeneous sensor networks based on shearlet and wavelet transform

A Classification-Based Image Fusion Scheme Using Wavelet Transform

IFCNN: A General Image Fusion Framework Based on Convolutional Neural Network.

Confocal microscopy multi-focus image fusion method based on axial information guidance

An Image Fusion Method Based on Sparse Representation and Sum Modified-Laplacian in NSCT Domain

SIFusion: Lightweight infrared and visible image fusion based on semantic injection

An image fusion algorithm based on image clustering theory

C2IENet: Multi-branch medical image fusion based on contrastive constraint features and information exchange

A Novel Fusion Strategy and Convolutional Sparse Coding for Robot Multisource Image Fusion