Huaixin Chen

Aiming at the problems of glare interference, local overexposure and detail loss caused by artificial light sources such as vehicle lamps and street lamps in nighttime road scenes, as well as the challenges of existing glare suppression models with large parameters, high computational complexity and difficulty in deploying on edge devices, this paper proposes a lightweight glare suppression network (LGSNet) based on ghost depthwise separable convolution and Lightweight Parallel Attention. Based on the U-Net architecture, the network introduces ghost depthwise separable convolution blocks (GhostDSC) in the encoder and decoder, which generates ghost features through cheap linear transformations by exploiting feature map redundancy, significantly reducing model parameters and computational costs while maintaining feature representation ability. Meanwhile, a Lightweight Parallel Attention (LPA) module is designed in the decoder stage, which integrates channel attention and pixel attention in parallel, enhancing the network’s attention to glare regions and edge details with extremely low parameter increment to improve detail recovery accuracy. In addition, a joint loss function consisting of background loss, glare loss and reconstruction loss is constructed to collaboratively optimize glare suppression and detail preservation. Experimental results on the public Flare7K++ dataset and the self-built nighttime road glare dataset NRGD show that the proposed method has only 7.45 M parameters, much lower than standard U-Net and Uformer. It achieves competitive results on full-reference metrics such as PSNR, SSIM, LPIPS and no-reference metrics such as NIQE, BRISQUE, PIQE, and can effectively suppress various types of glare interference and restore obscured scene details. It achieves a superior trade-off between model complexity and enhancement performance, significantly reducing the parameter count and computational overhead compared to heavy baselines, thereby offering a highly efficient solution for resource-aware glare suppression tasks.

Text-to-image person re-identification (T2I-ReID) aims to retrieve pedestrians from images/videos based on textual descriptions. However, most methods implicitly assume that training image–text pairs are correctly aligned, while in practice, issues such as under-correlated and falsely correlated image–text pairs arise due to coarse-grained text annotations and erroneous textual descriptions. To address this problem, we propose a T2I-ReID method based on noise identification and pseudo-text generation. We first extracts image–text features using the Contrastive Language–Image Pre-Training model (CLIP), then employs the token fusion model to select and fuse informative local token features, resulting in token fusion embedding (TFE) for fine-grained representations. To identify noisy image–text pairs, we apply the two-component Gaussian mixture model (GMM) to fit the per-sample loss distributions computed by the predictions of basic feature embedding (BFE) and TFE. Finally, when the noise identification tends to stabilize, we employ a multimodal large language model (MLLM) to generate pseudo-texts that replace the noisy text, facilitating learning more reliable visual–semantic associations and cross-modal alignment under noisy conditions. Extensive experiments on the CUHK-PEDES, ICFG-PEDES, and RSTPReid datasets demonstrate the effectiveness of our proposed model and the good compatibility with other baselines.

Considering the large-area distribution, smooth brightness gradients, and blurred boundaries of Mura defects in real industrial scenarios, as well as the challenge of balancing accuracy and efficiency in existing methods, we propose a lightweight deep learning-based detection method for large-area Mura defects, termed SIFNet. The SIFNet adopts a classical encoder-decoder architecture with MobileNet-V2 as the backbone. Furthermore, we design a Graph-based Semantic Interscale-fusion Block (GSIB) that integrates the Semantic Fluid Aggregation Module (SFAM) and the Semantic Graph Inference Module (SGIM) to collaboratively extract high-level semantic features across multiple scales and establish abstract semantic representations for accurately localizing large-area Mura defects. Specifically, SFAM leverages a global attention mechanism to extract cross-spatial semantic flows, guiding the model to focus on potential brightness anomaly regions in the image and SGIM explicitly models the semantic relationships between multi-scale features using graph convolution, enhancing the model's ability to interpret regions with blurred boundaries and ambiguous structures. To further improve the model’s sensitivity to edges in regions with smooth brightness transitions, we introduce a NeighborFusion Edge Enhancement Module (NEEM). This module integrates depthwise separable convolutions with a spatial attention mechanism and introduces a CrossNorm-based feature alignment strategy to enhance spatial collaboration across feature layers. Additionally, an edge enhancement mechanism is employed to significantly improve the model’s ability to delineate blurred Mura defect boundaries, while keeping computational cost low and strengthening edge perception and representation. Extensive quantitative and qualitative experiments on three large-area Mura defect datasets constructed in this study demonstrate that SIFNet achieves excellent detection performance with only 3.92M parameters and 6.89 GFLOPs, striking an effective balance between accuracy and efficiency, and fully meeting the demands of industrial deployment.

Existing nighttime visibility enhancement methods primarily focus on improving overall brightness under low-light conditions. However, nighttime road images are also affected by glare, glow, and flare from complex light sources such as streetlights and headlights, making it challenging to suppress locally overexposed regions and recover fine details. To address these challenges, we propose a Nighttime Road Glare Suppression Network (NRGS-Net) for glare removal and detail restoration. Specifically, to handle diverse glare disturbances caused by the uncertainty in light source positions and shapes, we designed a gated positional attention (GPA) module that integrates positional encoding with local contextual information to guide the network in accurately locating and suppressing glare regions, thereby enhancing the visibility of affected areas. Furthermore, we introduced an improved Uformer backbone named LCAtransformer, in which the downsampling layers adopt efficient depthwise separable convolutions to reduce computational cost while preserving critical spatial information. The upsampling layers incorporate a residual PixelShuffle module to achieve effective restoration in glare-affected regions. Additionally, channel attention is introduced within the Local Context-Aware Feed-Forward Network (LCA-FFN) to enable adaptive adjustment of feature weights, effectively suppressing irrelevant and interfering features. To advance the research in nighttime glare suppression, we constructed and publicly released the Night Road Glare Dataset (NRGD) captured in real nighttime road scenarios, enriching the evaluation system for this task. Experiments conducted on the Flare7K++ and NRGD, using five evaluation metrics and comparing six state-of-the-art methods, demonstrate that our method achieves superior performance in both subjective and objective metrics compared to existing advanced methods.

With the continuous development of the automotive industry, head-up display (HUD) technology has become widely implemented in modern vehicles. However, in the HUD off-axis three-mirror system, the curved shape of the windshield causes image distortion, negatively affecting the driver’s visual experience. To address this issue, this paper proposes a correction method for HUD projection based on optical path difference. Using the equal optical path difference criterion, the optical path difference from the eye box to the main reflector is calculated to determine the deflection value of the main reflector (M2). Subsequently, Zemax simulation software is used to correct the deflection of M2 and the subsequent overall structure. The subsequent surface positions are determined using the main ray tracing method. The simulation results demonstrate that this approach effectively corrects the projection tilt, yielding high-quality HUD projection images. Moreover, it simplifies the optimization process of system components, improving the overall performance and visual experience. This method offers a practical solution for enhancing HUD image quality in real-world driving conditions.

As the most crucial medium for multimedia presentation, liquid crystal displays (LCDs) is integral to numerous industries, making precise defect detection essential to ensure display quality and user experience. However, high accuracy defect detection of LCDs remains a significant challenge due to large-scale variations and high inter-class similarity, especially under the setting of few-shot learning. To address these challenges, we propose a few-shot defect detection network, namely HiSCAD-Net. Specifically, we design an auxiliary branch for hierarchical scale enhancement, which introduces additional objectness and classification constraints based on object pyramid sampling. Moreover, to tackle the misclassification caused by inter-class similarity, we introduce an object-level contrastive encoding to encourage class-discriminative feature learning, which enforces zero distance between objects of the same class, while ensuring that the distance between objects of different classes remains above a predefined threshold. Finally, we propose an adaptive decoupling module to mitigate interference between classification and regression tasks given limited training samples, thereby improving both tasks during decoding. To support benchmarking in few-shot LCD defect detection, we propose a new dataset named FSLCD. Experimental results on the FSLCD, NEU-DET, and PKU-Market-Phone datasets demonstrate that the proposed model outperforms 18 state-of-the-art methods, validating its effectiveness and generalizability. Notably, in the 10-shot setting, our model achieved a mean Average Precision (mAP) of 62.0%, surpassing the state-of-the-art by 10.4%.

In recent years, numerous advanced lightweight models have been proposed for salient object detection (SOD) in optical remote sensing images (ORSI). However, most methods still face challenges such as performance limitations and imbalances between accuracy and computational cost. To address these issues, we propose SggNet, a novel semantic- and graph-guided lightweight network for ORSI-SOD. The SggNet adopts a classical encoder-decoder structure with MobileNet-V2 as the backbone, ensuring optimal parameter utilization. Furthermore, we design an Efficient Global Perception Module (EGPM) to capture global feature relationships and semantic cues through limited computational costs, enhancing the model’s ability to perceive salient objects in complex scenarios, and a Semantic-Guided Edge Awareness Module (SEAM) that leverages the semantic consistency of deep features to suppress background noise in shallow features, accurately predict object boundaries, and preserve the detailed shapes of salient objects. To further efficiently aggregate multi-level features and preserve the integrity and complexity of overall object shape, we introduce a Graph-Based Region Awareness Module (GRAM). This module incorporates non-local operations under graph convolution domain to deeply explore high-order relationships between adjacent layers, while utilizing depth-wise separable convolution blocks to significantly reduce computational cost. Extensive quantitative and qualitative experiments demonstrate that the proposed model achieves excellent performance with only 2.70 M parameters and 1.38 G FLOPs, while delivering an impressive inference speed of 108 FPS, striking a balance between efficiency and accuracy to meet practical application needs.

The targets in Synthetic Aperture Radar (SAR) images are often tiny, irregular, and difficult to detect against complex backgrounds, leading to a high probability of missed or incorrect detections by object detection algorithms. To address this issue and improve the recall rate, we introduce an improved version of YOLOv8 (You Only Look Once), named MDD-YOLOv8. This model is not only fast but also highly accurate, with fewer instances of missed or incorrect detections. Our proposed model outperforms the baseline YOLOv8 in SAR image detection by utilizing dynamic convolution to replace static convolution (DynamicConv) and incorporating a deformable large kernel attention mechanism (DLKA). Additionally, we modify the structure of the FPN-PAN and introduce an extra detection header to better detect tiny objects. Experiments on the MSAR-1.0 dataset demonstrate that MDD-YOLOv8 achieves 87.7% precision, 76.1% recall, 78.9% mAP@50, and 0.81 F1 score. These metrics show an improvement of 8.1%, 6.0%, 6.9%, and 0.07, respectively, compared to the original YOLOv8. Although, MDD-YOLOv8 increases parameters by about 20% and GFLOPs by 53% more than YOLOv8n. To further validate the model’s effectiveness, we conducted generalization experiments on four additional SAR image datasets, proving that MDD-YOLOv8’s performance is robust and universally applicable. In summary, MDD-YOLOv8 is a robust, generalized model with strong potential for industrial applications.

In real-life complex traffic environments, vehicles are often occluded by extraneous background objects and other vehicles, leading to severe degradation of object detector performance. To address this issue, we propose a method named YOLO-OVD (YOLO for occluded vehicle detection) and a dataset for effectively handling vehicle occlusion in various scenarios. To highlight the model attention in unobstructed region of vehicles, we design a novel grouped orthogonal attention (GOA) module to achieve maximum information extraction between channels. We utilize grouping and channel shuffling to address the initialization and computational issues of original orthogonal filters, followed by spatial attention for enhancing spatial features in vehicle-visible regions. We introduce a CIoU-based repulsion term into the loss function to augment the network’s localization accuracy in scenarios involving densely packed vehicles. Moreover, we explore the effect of the knowledge-based Laplacian Pyramid on the OVD performance, which contributes to fast convergence in training and ensures more detailed and comprehensive feature retention. We conduct extensive experiments on the established occluded vehicle detection dataset, which demonstrates that the proposed YOLO-OVD model significantly outperforms 14 representative object detectors. Notably, it achieves improvements of 4.7% in Precision, 3.6% in AP@0.5, and 1.9% in AP@0.5:0.95 compared to the YOLOv5 baseline.