In this article, to bridge the gap between the traditional model-based methods and data-driven deep learning schemes, we propose a physical model-inspired deep unrolling network for solving nonlinear inverse scattering problems, termed PM-Net. The proposed end-to-end network is formed by two consequent steps. First, an augmented Lagrangian method is introduced to transform a constrained objective function to be an unconstrained optimization. In addition, it is further decomposed into four quasi-linear subproblems. Second, we unfold the iterative scheme into a layer-wise deep neural network. Each subproblem is mapped into a module of the deep unrolling network. In PM-Net, these variables including the weight, the regularization of contrast, and other parameters are learned and updated alternately by corresponding network layers. PM-Net effectively combines neural networks with the knowledge of underlying physics as well as traditional techniques. Unlike existing networks, PM-Net explicitly exploits contrast source and contrast modules. Compared to traditional iterative methods, the performance of PM-Net is comparable or even better than subspace-based optimization method in the high noise-level circumstance. Compared to the state-of-the-art learning approaches, not only less network parameters need to be learned, but also better performance is achieved by PM-Net.

Electromagnetic inverse scattering problems (ISPs) aim to retrieve permittivities of dielectric scatterers from the scattering measurement. It is often highly nonlinear, causing the problem to be very difficult to solve. To alleviate the issue, this letter exploits a linear model-based network (LMN) learning strategy, which benefits from both model complexity and data learning. By introducing a linear model for ISPs, a new model with network-driven regularizer is proposed. For attaining efficient end-to-end learning, the network architecture and hyper-parameter estimation are presented. Experimental results validate its superiority to some state-of-the-arts.

Electromagnetic inverse-scattering problems (ISPs) are concerned with determining the properties of an unknown object using measured scattered fields. ISPs are often highly nonlinear, causing the problem to be very difficult to address. In addition, the reconstruction images of different optimization methods are distorted which leads to inaccurate reconstruction results. To alleviate these issues, we propose a new linear model solution of generative adversarial network-based (LM-GAN) inspired by generative adversarial networks (GAN). Two sub-networks are trained alternately in the adversarial framework. A linear deep iterative network as a generative network captures the spatial distribution of the data, and a discriminative network estimates the probability of a sample from the training data. Numerical results validate that LM-GAN has admirable fidelity and accuracy when reconstructing complex scatterers.

In this paper, a novel alternating direction method of multiplier (ADMM) is proposed to solve the inverse scattering problems. The proposed method is suitable for a wide range of applications with electromagnetic detection. In order to solve the internal ill-posed problem of the integral equation and make the reconstructed images more closer to the ground truth, first, the augmented Lagrangian method is introduced to transform the complex constrained optimization problem into the extremum problem of unconstrained cost function. Therefore, two artificial regularization factors of the cost function are optimized. Then, this proposed method decomposes the unconstrained global problem in the inversion process into three linear sub-problem forms of contrast source function, contrast function, and dual variables. And the form of the updated algebra for each sub-problem is not complicated. By cross-iterating and updating contrast source function, contrast function, and dual variables, the global minimization of the cost function can be accurately found. Finally, the proposed method is compared with the existing well-known iterative method for solving the inverse scattering problem. Both the numerical and experimental tests verify the validity and accuracy of the proposed ADMM.

Event cameras provide high temporal resolution, high dynamic range, and low power consumption, making them ideal for high-speed robotic applications such as autonomous driving and aerial navigation. Events-to-video reconstruction aims to generate dense, photorealistic frames from sparse, asynchronous event streams. However, event data are globally sparse but locally bursty in fast-motion scenarios. Naive processing of such data is prone to noise in inactive regions, while excessive focus on dense bursts overlooks sparse global cues, leading to incomplete structures and degraded textures. Moreover, existing reconstruction methods are generally parameter-heavy and computationally demanding, limiting real-time deployment on resource-constrained edge devices. To address these challenges, we propose a lightweight events-to-video reconstruction framework optimized for edge devices. Specifically, a score-guided sparse adaptation module evaluates local event density to enhance salient regions and improve feature quality under sparse inputs. Additionally, a multi-level knowledge distillation strategy transfers reconstruction capability from a high-capacity teacher network to a compact student network, substantially reducing parameters with only minor compromise in visual fidelity. Implemented on the NVIDIA Jetson Orin NX SUPER, the framework achieves an effective balance between reconstruction accuracy, latency, and resource usage. Extensive experiments on a simulated dataset and four high-speed, high-dynamic-range, real-world datasets demonstrate that the proposed method consistently outperforms existing approaches in both image quality and inference time. The results indicate its potential for real-time event-based vision on resource-constrained edge devices.

By leveraging the blur-noise trade-off, imaging with non-uniform exposures largely extends the image acquisition flexibility in harsh environments. However, the limitation of conventional cameras in perceiving intra-frame dynamic information prevents existing methods from being implemented in the real-world frame acquisition for real-time adaptive camera shutter control. To address this challenge, we propose a novel Neuromorphic Shutter Control (NSC) system to avoid motion blur and alleviate instant noise, where the extremely low latency of events is leveraged to monitor the real-time motion and facilitate the scene-adaptive exposure. Furthermore, to stabilize the inconsistent Signal-to-Noise Ratio (SNR) caused by the non-uniform exposure times, we propose an event-based image denoising network within a self-supervised learning paradigm, i.e., SEID, exploring the statistics of image noise and inter-frame motion information of events to obtain artificial supervision signals for high-quality imaging in real-world scenes. To illustrate the effectiveness of the proposed NSC, we implement it in hardware by building a hybrid-camera imaging prototype system, with which we collect a real-world dataset containing well-synchronized frames and events in diverse scenarios with different target scenes and motion patterns. Experiments on the synthetic and real-world datasets demonstrate the superiority of our method over state-of-the-art approaches.