A knowledge graph-guided 3D visual question answering method was proposed to capture the implicit common-sense semantic relationships between objects in the scene. By introducing external structured knowledge, the model was effectively enhanced in both semantic understanding and reasoning ability. Specifically, key semantic entities were extracted from the question text, and a knowledge graph-guided feature enhancement module was designed to obtain knowledge features using these key semantic entities. The knowledge features were fused with visual features extracted both from the question representation and from the 3D object detection network for answer prediction. Experimental results on the ScanQA dataset showed that the proposed method outperforms existing baseline models on metrics such as EM@1 and BLEU-4.

To address the issue of inconsistent thermal stress between the upper and lower devices of the same bridge arm in converters under conditions of uneven aging or layout deviations, a novel thermal balancing method for upper and lower bridge arms based on dead-time shifting adjustment was proposed. By monitoring the on-state voltage of insulated gate bipolar transistor (IGBT) online, the real-time junction temperature of IGBTs was deduced based on the mapping relationship between the on-state voltage and junction temperature of IGBTs. Then, the temperature difference between the upper and lower devices of the same bridge arm was determined. When the thermal stress difference between the upper and lower bridge arms exceeded the threshold, the effective conduction time of the upper and lower bridge arms was adjusted by fine-tuning the position of the dead time, thereby actively regulating the power loss distribution of the upper and lower bridge arms. This achieved an increase in the loss of the bridge arm with lower temperature and a decrease in the loss of the bridge arm with higher temperature without affecting the total system loss, thus improving the thermal stress consistency of power devices in the converter and enhancing system reliability. Theoretical analysis and experimental results showed that the proposed method did not affect system efficiency and had no significant negative impact on output performance. Under typical operating conditions, the average temperature difference between the upper and lower devices was reduced by approximately 14% after adjustment, effectively improving the thermal stress consistency of the upper and lower bridge arms.

Current industrial image anomaly detection methods generally face challenges such as high dependency on anomalous samples, insufficient realism of synthetic samples, and limited perception capability for complex defects. To An industrial image anomaly detection method based on diffusion synthesis and feature mining was proposed to address these issues. Accordingly, a category-sensitive selective diffusion anomaly synthesis module was designed to generate pseudo-anomaly samples with adjustable intensity and category adaptability through controllable perturbation and category-sensitive loss, which could effectively alleviate data scarcity. Meanwhile, a multi-stage feature mining framework was constructed, including contrast-driven feature selection, multi-dimensional perception attention reconstruction, and residual refinement selection modules, enabling dynamic screening of anomaly-sensitive features and enhancement of structural details. Experimental results demonstrated that the proposed method achieved outstanding performance on the MVTec AD and MPDD datasets, with image-level AUROC scores of 99.7% and 98.4%, and pixel-level AUROC scores of 99.0% and 98.7%, respectively, validating its effectiveness and robustness.

A railway foreign object detection algorithm for complex weather conditions based on an improved RT-DETR, named FRP-DETR, was proposed. A feature complementary mapping module (FCM) and Pzconv units were introduced, which built complementary paths between shallow spatial details and deep semantic information to compensate for the limitations of single-scale feature representation, with the two components working collaboratively to enhance the perception capabilities of small and edge targets. A railway perception modulation fusion module (RMFM) was introduced, which was based on adaptive channel attention and spatial modulation mechanisms to enhance the model’s response to key semantic information in railway scenes. The original downsampling module was replaced with pinwheel-shaped convolution (PSConv), which enhanced edge texture extraction through multi-directional asymmetric padding and separable convolution. A railway foreign object detection dataset containing four weather conditions was constructed based on the SaMam style transfer method, by transferring sunny railway foreign object images to rainy, foggy, and snowy scenes. Experimental results showed that compared to the original RT-DETR-R18 model, this method achieved improvements of 1.65 percentage points and 3.4 percentage points in mAP@0.5 and mAP@0.5:0.95, respectively, with a 63.5% reduction in parameter count and an inference speed of 88 frames per second. The results verified that FRP-DETR achieved high accuracy, lightweight design, and real-time performance in railway foreign object detection under complex weather conditions.

Significant progress in multimodal large language models (MLLMs) has driven advances in visual question answering, visual understanding, and reasoning tasks, and their potential for deployment on resource-constrained edge devices is increasingly recognized. However, large model sizes and the substantial costs of deployment and inference remain major barriers to practical adoption. Optimizing MLLMs for edge devices has become a critical research direction in this field. A comprehensive survey of recent advances in optimizing MLLMs for edge deployment was presented, along with the associated challenges and development trends. The research evolution of MLLMs on edge devices was reviewed, with particular emphasis on model architecture optimization and inference scheduling strategies. In model architecture optimization, techniques including visual information compression, sparse attention, and mixture-of-experts models were specifically analyzed. System-level optimizations involving computation scheduling, hardware adaptation, compilation optimization, and cloud-edge collaboration were investigated to enhance inference efficiency and energy efficiency. Furthermore, the key challenges of these models in practical applications were discussed, and a variety of task scenarios ranging from assistive to collaborative and autonomous types were covered, categorized by the perspective of autonomy levels. Finally, current limitations were summarized and future research directions regarding standardized deployment, efficient computing and storage, and multi-modal fusion optimization were outlined.

A lightweight detector SCGI-DETR was proposed based on an enhanced RT-DETR in order to address challenges in grape leaf disease detection—complex background, missed detection of small target, and resource-constrained deployment. The efficient StarNet backbone was employed to reduce parameter count and computational cost, enabling lightweight deployment. A feature pyramid CGSFR-FPN was designed. Spatial feature reconstruction was combined with multi-scale feature fusion in order to strengthen global context modeling and improve localization of multi-scale lesions in cluttered scenes. The Inner-PowerIoU v2 loss was constructed, which integrated global convergence acceleration and local region alignment in order to speed up bounding-box regression and enhance small-object detection performance. SCGI-DETR attained 91.6% precision, 89.8% recall and 93.4% mAP@0.5 on a grape leaf disease dataset, which improved 2.6, 2.4 and 2.3 percentage points over the baseline, and reduced parameters and computation by 46.2% and 64%, respectively. Results demonstrate that the improved algorithm achieves lightweight implementation while delivering superior detection performance, meeting deployment requirements for mobile and embedded devices.

Autonomous driving simulation platforms play a vital role in the development, testing and validation of autonomous driving systems. A systematic review of the classification and key technical pathways of mainstream simulation platforms was presented, covering aspects such as environment modeling, sensor simulation, vehicle dynamics modeling, perception algorithm evaluation, V2X communication and cloud-based simulation. Core challenges and research progress related to synthetic data generation, cost-effective algorithm training, cross-domain generalization and platform scalability were analyzed, emphasizing simulation technologies based on computer vision and artificial intelligence. Future development trends were discussed. Simulation platforms are evolving toward higher realism, interactivity and closed-loop validation with the advancement of emerging technologies such as generative AI, neural rendering and multimodal learning, gradually forming a comprehensive pipeline that integrates data generation, algorithm training and performance evaluation. Simulation platforms will continue to play an essential role in enhancing the generalization capability of autonomous driving system, accelerating product deployment, and establishing standard testing and validation framework in the future.

A small object detection algorithm FMCM-YOLO based on feature enhancement and fusion attention mechanism was proposed, aiming at the challenges of limited feature extraction, foreground-background confusion, and severe missed and false detections in small object detection in optical remote sensing images. Firstly, a four-head detection model was designed and a small target detection layer was added to detect numerous small objects in optical remote sensing images. Secondly, a feature enhancement module was proposed in the backbone network, which improved feature extraction capability by designing a multi-branch convolutional structure and introducing dilated convolution of different sizes. Thirdly, channel and spatial attention mechanisms were incorporated into the neck network, and a residual structure was introduced to focus on small objects, facilitating the distinction between targets and backgrounds. Finally, MPDIoU was adopted as the model’s loss function to accelerate convergence and enhance detection performance for small objects. Experimental results demonstrated that the mAP50 of the proposed algorithm on the two public datasets, USOD and AI-TOD, reached 89.9% and 60.6% respectively, which were 2.8 and 5.9 percentage points higher than those of the baseline algorithm YOLOv5m. Especially, the mean average precision for extremely tiny, tiny, and small objects increased by 2.1, 6.5, and 5.1 percentage points, respectively. These results proved that the FMCM-YOLO algorithm effectively improved the detection performance of small targets in optical remote sensing images.

Discrete element method (DEM) and Tavares breakage model were employed to investigate the rock fragmentation mechanism and the statistical characteristics of fragments under impact loading. Drop hammer impact crushing process of limestone specimens with different aspect ratios (1.0—2.0) and sizes (diameter 30—50 mm) was systematically simulated. Image J image processing technology was used to quantitatively analyze the fragment count, particle size distribution, and shape indices. The results showed that increasing the aspect ratio and the size significantly raised the crushing energy threshold, which led to a delay in the peak normal force, a reduction in its amplitude, and a slowdown of the overall fragmentation process. The equivalent particle size of fragments showed poor alignment with the sieve analysis curve, which indicated that the fragments commonly exhibited irregular geometric configurations such as flaky and rod-like shapes. Fragments from specimens with lower aspect ratios (1.0) and smaller sizes (diameter 30 mm) had lower roundness, more pronounced angularity, and greater surface roughness. A quantitative relationship model among geometric parameters–crushing response–fragment morphology can provide theoretical basis and regulatory pathways for the design of impact-resistant structures, the optimization of blasting parameters, and the control of particle morphology in rockfill grading.

A chest disease classification network based on convolution-assisted window self-attention was proposed, called CAWSNet, aiming at the issues of varying lesion sizes, complex textures, and mutual interference in chest X-ray images. The Swin Transformer was utilized as the backbone, employing window self-attention to model long-range visual dependencies. Convolution was introduced to enhance local feature extraction capability while compensating for the deficiencies of window self-attention. Image relative position encoding was used to dynamically calculate directed relative positions, helping the network better model pixel-wise spatial relationships. Class-specific residual attention was employed, and the classifier’s focus area was adjusted based on disease categories in order to highlight effective information and enhance multi-label classification capability. Dynamic difficulty loss function was proposed to alleviate the problem of large differences in disease classification difficulty and the imbalance of positive and negative samples in the dataset. The experimental results on the public datasets ChestX-Ray14, CheXpert and MIMIC-CXR-JPG demonstrate that proposed CAWSNet achieves AUC scores of 0.853, 0.898 and 0.819, respectively, confirming the effectiveness and robustness of the network in diagnosing chest diseases through X-ray images.

A lightweight small target detection algorithm called RTA-YOLOv8s was proposed in order to address the challenges of complex backgrounds, small target, and limited device resources in UAV images. The RepVGG module was introduced into the backbone network to enhance feature extraction capabilities. A tri-branch attention mechanism was applied to reduce false positive and false negative rates. A dedicated small target detection head was integrated to improve detection accuracy. The WIoUv3 loss function was adopted to improve localization and robustness. The experimental results showed that the RTA-YOLOv8s algorithm achieved a mAP50 of 44.9% and detection speed of 88.5 frame per second on the VisDrone dataset. mAP50 increased by 6.1%, detection accuracy increased by 4.7%, and params reduced by 13.9% compared with YOLOv8s. The improved algorithm effectively addresses the poor detection performance in complex UAV scenes, and balances accuracy and speed. The user-friendly interface design enables result visualization, making detection tasks more intuitive and easier to operate, and is suitable for UAV target detection.

An improved lightweight small object detection method based on YOLOv8s was proposed to address the challenges of complex background interference, small object sizes, and limited device resources in unmanned aerial vehicle (UAV) image object detection. A high-resolution detection head was reconstructed to improve the sensitivity to tiny objects. A dynamic multi-scale aggregation network was designed to adaptively adjust the receptive field for scale-variant targets, while an efficient multi-scale feature deep fusion module was proposed to integrate features from different semantic levels. An efficient channel-spatial attention module was introduced to further refine fused features. Experiments on the VisDrone2021 dataset demonstrated that the proposed method achieved improvements of over 5.0% in average precision, precision at different intersection over union thresholds, and detection precision for small and medium-sized objects, with a 16.25% reduction in parameters. Tests on the DOTA dataset validated the generalization ability of the proposed model. Results demonstrate that the proposed method improves the detection performance while achieving the model light weighting, indicating its practical value for object detection in UAV aerial images.

A lightweight micro-expression recognition method based on optical flow and convolutional vision Transformer was proposed to solve the problems of short duration, low motion intensity and insufficient sample size of micro-expressions. The optical flow and optical strain of human faces between the onset frame and the apex frame were extracted to highlight the movement of facial muscles, thereby effectively reducing the texture interference and lowering the feature dimension. The adversarial domain adaptation method based on identity domain was adopted to further remove the irrelevant components in the micro-expression features by making full use of the subjects’ labels. A lightweight multi-stage CNN-Transformer hybrid model named MiER-CvT, including the convolutional embedding layer, the convolutional Transformer block and the SeqSoftmax layer, was constructed to enhance the model’s capabilities of local representation and information integration for micro-expressions. The experimental results showed that the proposed method achieved a UF1 score of 0.9171 and a UAR score of 0.9192 on the MEGC 2019 dataset, and the parameter number and computational complexity of MiER-CvT were 7.5 M and 0.1 G, respectively. Compared with the existing methods, such as MiMaNet, the proposed method has the advantages of high precision and light weight.

A collaborative control approach integrating multi-agent proximal policy optimization (MAPPO) and maximum pressure control (MPC) was developed to address traffic control challenges at mixed traffic intersections involving connected autonomous vehicles (CAVs), human-driven vehicles (HDVs), and pedestrians. A hierarchical state space and an action space were designed through the formulation of a decentralized partially observable Markov decision process (Dec-POMDP). Reward functions considering the balance among safety, traffic efficiency, phase switching frequency, and regulatory compliance were introduced. The proposed model was validated under low, medium, and high traffic flow conditions using a centralized training with decentralized execution framework on the SUMO simulation platform. Experimental results demonstrated that the proposed MAPPO-MPC method significantly improved the throughput (by up to 44.3%) while reducing the queue length (by up to 49.3%), average vehicle delay (by up to 43.2%), and pedestrian waiting time (by up to 43.53%). Moreover, the model exhibited more substantial performance advantages when the CAV penetration rate exceeded 50%, outperforming the traditional Webster-based methods and the baseline models.

Aiming at the problem of collaborative multi-task assignment of heterogeneous UAVs under multiple constraints, a three-objective optimization model was constructed, which considered the UAV flight distance cost, time cost, combat effectiveness and multiple constraints. A multi-objective particle swarm optimization algorithm based on hybrid strategies was proposed to solve the model. An average action efficiency index of ammunition was proposed to evaluate the task strike efficiency, and considering the possibility of deadlock during task execution, a calculation of waiting time was proposed in the process of modeling. In order to solve the problem that traditional particle swarm optimization falls into local optimality, and ensure that feasible solutions satisfying constraints are searched, a constraint-based particle dynamic optimal initialization strategy, a dominance relationship-based advantageous individual selection strategy, and a task-based small module particle update and correction strategy were proposed, respectively. The overall performance of the algorithm in terms of convergence accuracy and diversity was effectively improved by these strategies. The validity of the model and the algorithm was verified through multi-scenario simulation experiments and ablation experiments. Results show that the solution sets obtained by the proposed algorithm are more convergent, diverse and evenly distributed than the comparative algorithms, and the collaborative multi-task assignment of heterogeneous UAVs is efficiently realized by the proposed algorithm.

The development and research status of YOLO algorithm in traffic object detection were systematically summarized from the perspective of the three core elements of 'people-vehicle-road' in order to comprehensively analyze the important role of YOLO (You Only Look Once) algorithm in improving traffic safety and efficiency. The commonly used evaluation indexes of YOLO algorithm were outlined, and the practical significance of these indexes in traffic scenarios was elaborately expounded. An overview of the core architecture of YOLO algorithm was provided, its development process was traced, and the optimization and improvement measures in each version iteration were analyzed. The research status and application scenarios of YOLO algorithm for traffic object detection were sorted out and discussed from the perspective of the three traffic objects 'people-vehicle-road'. The limitations and challenges of YOLO algorithm in traffic object detection were analyzed, and corresponding improvement methods were proposed. Future research focuses were anticipated, providing a research reference for the intelligent development of road traffic.

Based on the statistics of typical cases of water and sand gushing in subway tunnel construction stages in China from 2002 to 2019, the disaster characteristics were analyzed from the aspects of disaster occurrence characteristics, disaster geological environment and hazard factors. According to the geological environment, causes and forms of disaster sources and engineering conditions, the disaster-causing structure of water and sand gushing in the subway tunnels was classified into 3 categories including 12 types. The first category is large-scale unfavorable geological bodies, including fault and weak fracture zone type, karst and underground rivers type, interlayer fracture zone type, weathering deep groove type, intrusive rocks type, and underwater sandy stratum type. The second category is water-bearing sand and soft soil stratum, including overlying/invading soft soil type, upper-soft and lower-hard composite stratum type, water-rich sandy stratum type, and ground cavity/water bag and silt stratum type. The third category is artificial underground water-rich space, including underground water transmission pipes type, abandoned mining spaces and air-raid shelters filled with water type. Three typical disaster modes of water and sand gushing in subway tunnels with the soil surrounding rock were proposed based on the mechanical characteristics of soil instability and failure, namely, sliding failure mode, breaking failure mode, and seepage failure mode.

The multi-brain attention mechanism and capsule fusion module based on CapsNet (MBA-CF-cCapsNet) was proposed in order to improve the accuracy of emotion recognition. EEG-fNIRS signals were evoked by emotional video clips to construct TYUT3.0 dataset, and the features of EEG and fNIRS were extracted and mapped to the matrix. The features of EEG and fNIRS were fused by the multi-brain region attention mechanism, and different weights were given to the features of different brain regions in order to extract higher quality primary capsules. The capsule fusion module was used to reduce the number of capsules entering the dynamic routing mechanism and reduce the running time of the model. The MBA-CF-cCapsNet model was used to conduct experiment on the TYUT3.0 dataset. The accuracy of emotion recognition combined with the two signals increased by 1.53% and 14.35% compared with the results of single-modal EEG and fNIRS. The average recognition rate of the MBA-CF-cCapsNet model increased by 4.98% compared with the original CapsNet model, and was improved by 1%-5% compared with the current commonly used CapsNet emotion recognition model.

A thorough analysis and cross-comparison of recent relevant works was provided, outlining a closed-loop process for EEG data analysis based on deep learning. EEG data were introduced, and the application of deep learning in three key stages: preprocessing, feature extraction, and model generalization was unfolded. The research ideas and solutions provided by deep learning algorithms in the respective stages were delineated, including the challenges and issues encountered at each stage. The main contributions and limitations of different algorithms were comprehensively summarized. The challenges faced and future directions of deep learning technology in handling EEG data at each stage were discussed.

To address the core issue of efficiency degradation in concentrator photovoltaics (CPV) caused by high heat flux, microchannel cooling technology has emerged as a critical research direction. Advances in this field were systematically reviewed, with a focus on three mainstream configurations: single-layer, manifold, and jet-impingement structures. The effects of factors such as channel geometry, working fluid, and flow regime on pressure drop, temperature rise, and cooling efficiency were analyzed. Furthermore, the strengths and weaknesses of various flow channel designs, including serpentine, pin-fin, and fractal configurations, were compared. Studies have shown that single-layer straight channels are prone to high pressure drop and temperature non-uniformity. Derived serpentine and fractal channel designs can enhance heat transfer by perturbing the flow. The manifold configuration effectively reduces flow resistance and improves temperature uniformity, often serving as a foundational platform for integrating other cooling technologies. The hybrid jet-impingement and microchannel approach demonstrates potential in addressing localized ultra-high heat flux, though its system complexity and energy consumption still require optimization. Looking ahead, future research on this technology should progressively advance from the millimeter scale to the micro-scale, enabling the development of lightweight and efficient portable concentrator photovoltaic devices. The focus should shift from pursuing “cooling intensity” to emphasizing “intelligent heat transfer architecture,” deeply integrating bionics and artificial intelligence to design adaptive microchannels. Furthermore, the research perspective should transition from optimizing individual components toward a systemic, co-design approach that couples photonic, thermal, electrical, and mechanical multi-physics fields.