Deep learning has recently garnered substantial interest in hyperspectral unmixing due to its exceptional learning capabilities. In particular, unsupervised unmixing methods based on autoencoders have become a research hotspot, with many existing networks focusing on the fusion of spatial and spectral information. However, the diversity of fusion structures makes it challenging to select appropriate modules that meet unmixing requirements, while the issue of endmember variability is often neglected. In this paper, we propose a novel spatial-spectral adaptive fusion network (SSAF-Net) that accounts for endmember variability. The network consists of two cascaded encoders and a deep generative model (DGM) based on a variational autoencoder. The encoders perform local spatial-spectral information fusion through channel and spatial attention mechanisms, respectively, while self-perception loss facilitates global information fusion during the cascading process. In addition, we address endmember variability using a proportional perturbation model (PPM), learning the necessary endmember parameters through an elaborately designed DGM. Our SSAF-Net learns both endmember variability and the corresponding abundances in an unsupervised manner. Experimental results on a synthetic dataset and real-world datasets validate the significant superiority of SSAF-Net compared to other methods. The code for this work is available at https://github.com/yjysimply/SSAF-Net.
3 November 2023 in IEEE Robotics and Automation Letters
Self-supervised monocular depth estimation (MDE) has great potential for deployment in a wide range of applications, including virtual reality, autonomous driving, and robotics. Nevertheless, most previous studies focused on complex architectures to pursue better performance in MDE. In this letter, we aim to develop a lightweight yet highly effective self-supervised MDE model that can deliver competitive performance in edge devices. We introduce a novel MobileViT-based depth (MViTDepth) network that can effectively capture both local features and global information by leveraging the strengths of convolutional neural networks (CNNs) and a vision transformer (ViT). To further compress the proposed MViTDepth model, we employ knowledge distillation, which leads to improved depth estimation performance. Specifically, the self-supervised MDE MonoViT is used as a teacher model to construct the knowledge distillation loss for optimizing a student model. Experimental results on benchmark datasets demonstrate that the proposed MViTDepth significantly outperforms Monodepth2 in terms of parameters and accuracy, thereby indicating its superiority in application to edge devices.
Deep learning techniques have recently shown great promise in the field of bearing fault diagnosis, yet their performance is often limited by practical challenges such as insufficient fault data and varied working conditions in real-world industrial settings. Furthermore, traditional deep transfer learning approaches often require extensive parameter fine-tuning for specific tasks, thus reducing their adaptability in scenarios where rapid deployment is crucial. To address these issues, we propose a novel metric-based meta-learning relation network designed for few-shot bearing fault diagnosis across diverse domains. This method directly targets an industrial need: accurate fault detection with minimal training samples under unseen operational conditions. This capability is critical for predictive maintenance systems in resource-constrained environments. Vibration signals from various working conditions are first transformed into two-dimensional time-frequency images. These samples are then divided into meta-training and meta-testing sets, with each set further split into support and query subsets according to a meta-learning strategy. Following this division, a residual shrinkage non-local feature extraction module is introduced to extract and combine features from both subsets. A neural network with a nonlinear metric is subsequently employed to compute similarity scores between the support and query sets. The proposed method enables rapid and precise bearing fault diagnosis, even with limited data samples and under unknown working conditions, which are typical in maintenance workshops and field operations. Comparative tests on three datasets demonstrate that our approach outperforms existing methods under different working conditions and noise levels, highlighting its potential to reduce unplanned downtime and improve equipment reliability in real industrial applications. The experimental results further confirm the method’s robust generalization and rapid adaptability.
Convergence of the diffusion RLS (DRLS) algorithm to steady-state has been extensively studied in the literature, whereas no analysis of its transient convergence behavior has been reported yet. In this letter, we conduct a theoretical analysis of the transient behavior of the DRLS algorithm for cyclostationary colored inputs, in the mean and mean-square error sense. The resulting analytical models allows us to thoroughly investigate the convergence behavior of the algorithm over adaptive networks in such complex scenarios. Simulation results support the accuracy and correctness of the theoretical findings.
23 April 2025 in IEEE Transactions on Geoscience and Remote Sensing
Hyperspectral unmixing (HU) aims to decompose mixed pixels in remote sensing imagery into material-specific spectra and their respective abundance fractions. Recently, autoencoders have made significant advances in HU due to their strong representational capabilities and ease of implementation. However, relying exclusively on feature extraction from a single-modality hyperspectral image can fail to fully utilize both spatial and spectral information, thereby limiting the ability to distinguish objects in complex scenes. To address these limitations, we propose a multimodal spectral-spatial feature fusion network (MSSF-Net) for enhanced HU. MSSF-Net adopts a dual-stream architecture to extract feature representations from complementary input modalities. Specifically, the hyperspectral subnetwork leverages a convolutional neural network (CNN) to capture spatial information, while the light detection and ranging (LiDAR) subnetwork incorporates an enhanced channel attention mechanism (ECAM) to capture the dynamic changes in spatial information across different channels. Furthermore, we introduce a cross-modal fusion (CMF) module that integrates spectral and spatial information across modalities, leading to more robust feature representations. Experimental results indicate that MSSF-Net significantly outperforms existing traditional and deep learning-based methods in terms of unmixing accuracy. The code is available at https://github.com/Octopus-Squidward/MSSF-Net.
15 August 2023 in IEEE Transactions on Circuits and Systems II: Express Briefs
The data-reuse RLS (DR-RLS) algorithm is a computationally efficient technique that has been recently introduced to improve the tracking capabilities of the RLS. Nevertheless, no analysis of its convergence has been proposed so far. The aim of that work is to fill this gap with theoretical analyzes in the sense of the mean and mean-square error. Both transient and steady-state analytical models are provided. Their effectiveness and accuracy are validated through numerical simulations.