Hyperspectral unmixing (HU) aims to decompose mixed pixels into their constituent spectral signatures and estimate their corresponding fractional abundances. Recently, the nonlinear spectral mixing model (NLMM) has advanced significantly and offered strong physical interpretability. However, the effective integration of physics-driven NLMM with data-driven deep learning (DL) approaches still remains a critical challenge. To address this, we propose a diffusion-enhanced autoencoder (DEAE), a novel unsupervised framework that innovatively incorporates the diffusion model (DM) into nonlinear HU. DEAE introduces the residual second-order attention mechanism to capture global spectral information, adaptively weighting informative bands while compressing redundant bands. Subsequently, we integrate the extended multilinear mixing model (EMLM) into the DM-enhanced decoder, which extracts latent features from the linear autoencoder’s output and generates an enhanced reconstructed image while simultaneously estimating the transition probabilities of EMLM. Finally, a nonlinear decoder outputs the ultimate reconstructed image based on both the enhanced reconstructed image and transition probabilities. Experiments conducted on synthetic and three real-world datasets demonstrate the superior performance of DEAE compared to the state-of-the-art methods based on both LMM and NLMM.

The availability of spectral library makes hyperspectral sparse unmixing an attractive unmixing scheme, and the powerful feature extraction capability of deep learning meets the requirements of estimating abundances with hundreds of channels in sparse unmixing. However, few related researches have been carried out. In this letter, we propose a window transformer convolutional autoencoder (WiTCAE) to address the sparse unmixing problem. In our method, a well-designed transformer encoder for hyperspectral images is applied before convolutional neural network (CNN), aiming at exploring non-local information by a new attention mechanism called window-based pixel-level multihead self-attention (WP-MSA). Three consecutive CNN blocks focus on further joint spatial-spectral feature extraction, and adjust the number of channels to the number of endmembers contained in the spectral library. Moreover, CNN establishes the connections among windows, and smooths out the discontinuities caused by window partition. The decoder is a convolutional layer with the kernel size of 1, and its weights are fixed to a known spectral library. Comparative experiments on both simulated and real datasets confirm the superiority of our proposed network.

Hyperspectral unmixing (HSU), which aims to identify constituent materials and estimate the corresponding proportions in a scene, is an essential research topic in remote sensing. Most deep learning-based methods are data-inspired, relying on massive amounts of data to train black-box-like networks. While a few model-inspired unmixing networks only consider the spectral features of the pixel, ignoring the exploration of spatial information between pixels. In this paper, we design a network topology according to the classical iterative algorithm, and the large number of learnable parameters contained in the network are continuously updated through data fitting. In other words, we integrate the concepts of both model-driven and data-driven and propose a deep interpretable fully convolutional neural network (DIFCNN). The iteration of the classic sparse unmixing algorithm is unfolded to provide guidance for the network structure and incorporate prior knowledge into the network. Meanwhile, two-dimensional (2D) convolutional layers are employed to automatically learn the spatial information at different scales. A known spectral library is used as a prior to initialize network parameters and reconstruct the image. The DIFCNN adopts an end-to-end training strategy, in addition, we establish a new loss function that adds a joint sparse constraint on the abundance result to the cross-entropy loss. Experiments on both synthetic and real datasets show that the performance of the DIFCNN not only outperforms the SUnSAL and its improved algorithms, but also is highly competitive in the state-of-the-art methods of deep learning.

Airborne synthetic aperture radar (SAR) image quality considerably degrades because of motion errors. High-precision motion compensation (MOCO) is necessary in an advanced SAR data processing scheme. Operation complexity and computation burden are both increased as development of ultra-high resolution SAR. There are two main disadvantages for conventional MOCO. Firstly, accurate envelope correction should be performed by complicated interpolation, expending a large number of computing resources. In addition, interpolation makes a separate process, which could not be integrated into imaging algorithms easily. Secondly, since the fixed processing flow of conventional MOCO, phase correction is seriously influenced by the accuracy of envelope correction. An efficient MOCO approach is presented in this paper. A novel calculation formula of line-of-sight (LOS) range displacement is introduced from another perspective. On this basis, modified phase correction turns to be performed before envelope correction, on the premise that the accuracy of signal phase should be guaranteed. Consequently, an approximate envelope correction without interpolation can be adopted using subswath, to improve the processing efficiency. Simulations with point targets and processing of real data are used to confirm the validity of the proposed approach.

Sliding spotlight synthetic aperture radar (SAR) is a rising imaging mode, whose azimuth resolution is higher and imaged area is larger. When processing data, two key problems should be considered. Firstly, system's pulse repetition frequency (PRF) is always insufficient, which introduces aliasing into the azimuth spectrum. Secondly, the effect of motion error enhances because of longer synthetic aperture, consequently the accuracy of motion compensation (MOCO) should be increased. This paper presents a modified imaging scheme based on subaperture. Subaperture method is used to overcome the problem that PRF is insufficient. Meanwhile, processing of subaperture data chooses high precision line-of-sight (LOS) motion compensation, improving focused quality. The presented algorithm can attain 0.1m azimuth resolution and has the value of practice. Point targets simulation and processing of real data are used to confirm the validity of the proposed approach.