Non-line-of-sight (NLOS) detection is widely encountered in military and civilian fields, such as urban battles and disaster rescues. In this study, a novel multipath imaging approach is proposed to solve the problem of multi-target localization behind an L-shaped corner for a single-channel ultra-wide-band (UWB) radar. Specifically, a multipath propagation model based on an L-shaped corner is established, and the positions of virtual radars are analyzed by the specular reflection feature of electromagnetic wave, which forms a virtual array aperture to achieve enough image resolution. Then, the radar echo is mapped based on each propagation path to form multiple images. Finally, according to the trajectory distribution in each image, the incoherent addition is exploited to fuse the multiple images to focus on the target locations. Simulation and experimental results demonstrate that the estimated target positions are consistent with the actual target locations, and the proposed multipath imaging algorithm is applicable in the case of partial path loss.

Many objects in the naturalenvironment are generated from the background and even transformed by nature or human beings. Thus, they do not have closed and well-defined boundaries in remote sensing imagery. Recently, convolutional neural network (CNN) based object detection achieved great success in the remote sensing field. However, there is no investigation in the literature about the detection of objects with ambiguous boundaries. In this article, taking the case of the potential loess landslide detection, we designed massive experiments to evaluate convolutional neural networks for detecting objects with ambiguous boundaries in remote sensing imagery. We analyzed the evaluated methods comprehensively by comparing the performance on objects with ambiguous boundaries in remote sensing imagery with the performance on ordinary objects in visual imagery. Furthermore, drawing from these analyses, we provided a fundamental principle of object representation and a meaningful suggestion of information learning to detect objects with ambiguous boundaries. We finished this article by presenting several promising directions for detecting objects with ambiguous boundaries to facilitate and spur future research. This article would provide a significant reference and guidance to develop detectors for objects with ambiguous boundaries in remote sensing imagery.

Time-frequency (TF) analysis has received considerable attention in seismic spectral anomaly detection for its superiority in significantly revealing the frequency content of seismic signals changing with time variation. In this letter, a novel method termed second-order horizontal multi-synchrosqueezing transform (SHMSST) is proposed for hydrocarbon reservoir identification. First, a Gaussian-modulated linear chirp model is constructed in the frequency domain for describing the signals with slowly or rapidly linear-varying group delay. Then, the second-order local group delay estimation of the model is deduced in the TF domain through an approach similar to the instantaneous frequency estimation in the synchrosqueezing transform (SST). Finally, the rearrangement procedure is executed iteratively to concentrate the blurry TF energy into the estimated group delay, which provides an energy-concentrated TF representation for the signals with distinct nonlinear and nonstationary features. Synthetic signal and field seismic data illustrate the effectiveness of the proposed SHMSST in time localization and direct hydrocarbon detection.

Existing tampering localization methods perform poorly when localizing scaled images that are missing high-frequency forensic traces. Preserving these traces is critical for accurately localizing tampering because scaling operations can obscure crucial details and introduce artifacts and distortions. To address this issue, we propose a simple yet effective tampering localization framework for scaled images, named SLRID. The framework consists of two components: the Symlet Wavelet Recovery Module (SLR) and the Detector Module (SE-RRU-net). The SLR employs an invertible network of the Symlet Wavelet Transform to simulate the loss of information in a scaled image caused by tampering, achieving high-fidelity restoration of high-frequency forensic traces. The SE-RRU-net network utilizes an end-to-end image segmentation network (RRU-net) for tampering localization, incorporating a Spatial Channel Squeezing and Excitation (SCSE) mechanism to enhance the extraction of forgery-related features in scaled images. Results indicate that the SLRID framework outperforms existing state-of-the-art methods in detecting forged images during scaling operations.

Salient object detection in optical remote sensing images (ORSI-SOD) encounters notable challenges, mainly because of the small scale of salient objects and the similarity between these objects and their backgrounds in images captured by satellite and aerial sensors. Conventional approaches frequently struggle to efficiently leverage multi-scale and multi-stage features. Moreover, these methods usually rely on sophisticated and resource-heavy architectures, which can limit their practicality and efficiency in real-world applications. To overcome these limitations, this paper proposes a novel lightweight network called the Multi-scale Feature Fusion Network (MFFNet). Specifically, a Multi-stage Information Fusion (MIF) module is created to improve the detection of salient objects by effectively integrating features from multiple stages and scales. Additionally, we design a Semantic Guidance Fusion (SGF) module to specifically alleviate the problem of semantic dilution often observed in U-Net architecture. Comprehensive evaluations on two benchmark datasets show that the MFFNet attains outstanding performance in four out of eight evaluation metrics while only having 12.14M parameters and 2.75G FLOPs. These results highlight significant advancements over 31 state-of-the-art models, underscoring the efficiency of MFFNet in salient object-detection tasks.

Recent advances in 3D Gaussian splatting (3DGS) have enabled efficient image-based scene reconstruction, but existing methods that rely heavily on multi-view photometric consistency remain sensitive to dynamic illumination and weakly constrained regions. This issue is especially evident in tunnel scenes, where limited ambient light and localized active illumination cause strong appearance variation and shadowed regions that appear weakly textured in the captured images. As a result, existing methods often suffer from appearance inconsistency, floating artifacts, and unstable Gaussian distributions. To address these challenges, we present dynamic lighting-aware Gaussian splatting (DLG-GS), a real-time framework designed primarily for tunnel-oriented reconstruction under dynamic lighting. DLG-GS includes two complementary components: a dynamic lighting-adaptive appearance modeling strategy that reduces illumination-induced artifacts while preserving local texture details, and a voxel–depth joint constraint that uses monocular depth priors to regularize the spatial distribution of voxel anchors and neural Gaussians, thereby improving optimization stability and suppressing floating artifacts in shadow-induced weak-texture regions. By jointly optimizing appearance adaptation and depth-guided spatial regularization, DLG-GS improves reconstruction stability and rendering quality while maintaining real-time performance. Experiments on a self-collected tunnel dataset show clear improvements over selected baselines, and additional evaluations on public benchmarks indicate competitive performance beyond the target tunnel setting.