Xiting Peng

Person re-identification (Re-ID) aims to match individuals across different cameras, a task complicated by variations in camera positions, resolutions, and lighting conditions. While supervised training improves Re-ID model accuracy, it requires significant annotation efforts. Unsupervised domain adaptation (UDA) methods address this by leveraging unlabeled target domain data but often fail to fully utilize multiple source domains and are constrained by computational resources. This paper introduces a lightweight multi-source domain adaptation method for person Re-ID that combines meta-learning with pseudo-label-based UDA. By employing self-paced meta-learning and style enhancement techniques, the model learns domain-invariant knowledge from easy to difficult source domains, enhancing pseudo-label quality during adaptation. Our approach, based on an omni-scale feature extraction network using deep separable convolution, combines global and partial feature branches to capture richer pedestrian features. Experiments on public and real-world datasets demonstrate that our method achieves competitive performance with significantly fewer parameters and FLOPs compared to state-of-the-art models, proving its effectiveness and practicality.

High-altitude airships (HAS) and unmanned aerial vehicles (UAVs) equipped with Multiaccess Edge Computing (MEC) servers have emerged as promising aerial MEC nodes for providing task offloading (TO) services to intelligent mobile devices (IMDs) in post-disaster scenarios. HAS offers robust computing and energy resources, while UAVs provide flexible, low-altitude coverage for rapid deployment. However, direct task offloading from IMDs to HAS often leads to task failures due to high transmission delays. UAVs with limited onboard resources require to minimize resource waste. Additionally, IMDs in sparse areas face insufficient TO services due to unfair UAV coverage. This paper defines these challenges as a joint optimization problem involving TO, RA, and UAV coverage fairness. It proposes a cooperative aerial Multiaccess Edge Computing (AMEC) framework integrating HAS and UAVs to address the issue. Within this framework, a hybrid TO scheme is first developed to mitigate the high transmission delay between IMDs and HAS. Second, a Distance, Resource, Urgency-based Decision Mechanism (DRUDM) is designed to enhance the accuracy of UAVs in selecting target IMDs for TO services. Third, a Coverage Fairness Guarantee (CFG) strategy is proposed to optimize UAV flight trajectories, ensuring IMDs in sparse areas receive fair TO services. Finally, the joint optimization problem is modeled as a Multi-Agent Partially Observable Markov Decision Process (MA-POMDP), and a DRUDM–CFG algorithm is presented to efficiently solve this complex non convex optimization problem. Experimental results demonstrate that the proposed algorithm outperforms other compared algorithms in task completion rate and average delay, benefiting from the DRUDM mechanism. Meanwhile, the CFG strategy effectively improves TO service fairness for IMDs in sparse areas.

In the context of smart manufacturing, improving the quality and efficiency of process planning, especially in the processing of complex parts, has become a key factor influencing the level of intelligence in manufacturing systems. However, most current process planning methods still heavily rely on manual expertise, leading to problems such as difficulty in knowledge reuse, low planning efficiency, and slow response times, which are inadequate to meet the diverse and changing needs of engineering applications. To address these issues, this paper proposes an algorithm for Assembly Process Reasoning and Decision-making based on Bidirectional Long Short-Term Memory with Attention (AP-BiLSTM-ATT), which aims to deeply explore the hidden relationships between the multi-dimensional features of parts and process plans, thereby achieving probabilistic modeling of process decisions. Specifically, the attributes, geometric features, and historical process plans of parts are first labeled and vectorized, transforming traditional process knowledge into structured data representations suitable for deep learning models. A BiLSTM network model, integrated with a multi-head attention mechanism, is then constructed to capture contextual dependencies and semantic weight distributions between features, enhancing the model’s ability to express complex process relationships. During training, the model learns the mapping distribution between features and processes from a large-scale historical process dataset, enabling intelligent reasoning and recommendation of process plans for new parts. The results show that this method outperforms traditional methods in terms of accuracy, response speed, and generalization ability in process planning, providing effective support for enhancing the intelligence of complex part process planning and laying a foundation for the structured expression and intelligent application of manufacturing process knowledge.

Magnetic particle imaging technology, a novel medical imaging technology, possesses rapid imaging, high penetration depth, and is free from ionizing radiation. However, the system point spread function causes imaging blurring, which can be further exacerbated by external environmental interferences. Although hardware improvements and system optimization can mitigate blurring, these approaches are often expensive and time-consuming, particularly for low-field imaging in large-scale systems. This article proposes a Fast Context-aware Saliency-enhanced Deblurring Network, FCS-edNET, to solve the challenging issue by deblurring the reconstructed images. The network introduces the Multi-scale Global module to enhance the multi-scale feature perception ability. The Multi-scale Denoising Prior algorithm, which employs a low-frequency filter operator to restrict image noise and offers priors for each layer of subnetworks, is designed to improve the model robustness. Finally, proposing a Multi-level Joint loss optimizes model parameters to promote model convergence speed and space distribution simulation capability. Extensive experiments on multiple public and private datasets demonstrate that FCS-edNET outperforms the state-of-the-art methods in MPI image deblurring efficiently, suggesting its potential to support future research toward clinical imaging applications. The code is available at https://github.com/ydz1118/FCS-edNET.

Recently, with the rapid development of autonomous driving technology, it prompts the vehicle detection technology to continuously improve its accuracy, stability and reliability to better meet the needs of self-driving. However, due to the interference of adverse factors in adverse weather, the decrease of detection accuracy of autonomous vehicle is led to the phenomenon of missing and wrong detection, which has a serious impact on the safety of autonomous vehicles. Therefore, we propose REWeather to solve such problems of autonomous vehicles in multiple adverse weather conditions. Firstly, to classify the types of adverse weather, distinguishing among foggy, rainy and snowy weather, Broad Learning System (BLS) which is simple and efficient is used in REWeather. Due to the impact of these adverse weathers on sensors, simple dark channel and guided filtering methods is used to preprocess foggy and rainy images, respectively. Then, we put the processed images into the Real-Enhanced Super-Resolution Generative Adversarial Networks (Real-ESRGAN) for further denoising and enhancing the details of detected objects, enabling the sensor to recognize other targets on the road faster and better in adverse weather. To ensure the best detection results, we also use latest Realtime Detection Transformer (RT-DETR) as the detector to validate our work and the final model is deployed on the edge device. Moreover, we use several public datasets and our own collected data to make a real world dataset containing a variety of adverse weathers to train and test our proposed framework, which makes it closer to the real situation. The results demonstrate that our framework achieves a 3.8% improvement in mAP, significantly enhancing the detection capability of autonomous vehicles under adverse weather conditions.

In the Internet of Autonomous Vehicles (IoAV), task offloading is crucial for managing tasks that require extensive computing power to guarantee vehicle safety under different weather scenarios. However, extreme weather events can lead to infrastructure damage and network disruptions, significantly increasing the computational demands of autonomous vehicles. These vehicles require additional computing resources to navigate complex road conditions and risks, all while facing a high degree of uncertainty, such as fluctuations in vehicle resource utilization and task workloads. To address these challenges, a new and lightweight task offloading decision framework, named 2FDP-BRL, has been first proposed in this paper. This framework not only considers the fast response time required for autonomous driving, but also considers the resource shortage and offloading uncertainty caused by extreme weather. Therefore, we introduce the dynamic pricing idea and the Interval Type-2 Fuzzy Inference System (IT2FIS) utilizing broad reinforcement learning to deal with various dynamic uncertainties in the IoAV under extreme weather. For the authenticity of experimental results, we utilize the VISSIM platform to collect experimental data and conduct simulations. Moreover, to accurately simulate extreme weather scenarios, we also account for the variability of infrastructure and road elements, including reduced transmission rates and decreased efficiency in executing tasks. Furthermore, to enhance the realism of the simulation, we incorporate historical weather data from NOAA for Shenyang in 2024 to model dynamic uncertainties under extreme weather conditions and conduct comparative experimental analyses focusing on task completion rates. Finally, the proposed framework was implemented on both a local setup and the Huawei Atlas 200I DK A2 device, illustrating its efficacy design.

The advancement of the Internet of Autonomous Vehicles has facilitated the development and deployment of numerous onboard applications. However, the delay-sensitive tasks generated by these applications present enormous challenges for vehicles with limited computing resources. Moreover, these tasks are often interdependent, preventing parallel computation and severely prolonging completion times, which results in substantial energy consumption. Task-offloading technology offers an effective solution to mitigate these challenges. Traditional offloading strategies, however, fall short in the highly dynamic environment of the Internet of Vehicles. This paper proposes a task-offloading scheme based on deep reinforcement learning to optimize the strategy between vehicles and edge computing resources. The task-offloading problem is modeled as a Markov Decision Process, and an improved twin-delayed deep deterministic policy gradient algorithm, LT-TD3, is introduced to enhance the decision-making process. The integration of LSTM and a self-attention mechanism into the LT-TD3 network boosts its capability for feature extraction and representation. Additionally, considering task dependency, a topological sorting algorithm is employed to assign priorities to subtasks, thereby improving the efficiency of task offloading. Experimental results demonstrate that the proposed strategy significantly reduces task delays and energy consumption, offering an effective solution for efficient task processing and energy saving in autonomous vehicles.

Introduction: Magnetic particle imaging (MPI), a radiation-free, dynamic, and targeted imaging technique, has gained significant traction in both research and clinical settings worldwide. Signal-to-noise ratio (SNR) is a crucial factor influencing MPI image quality and detection sensitivity, and it is affected by ambient noise, system thermal noise, and the magnetization response of superparamagnetic nanoparticles. Therefore to address the high amplitude system and inherent thermal noise present in conventional MPI systems is essential to improve detection sensitivity and imaging resolution.Method: This study introduces a novel open-loop, narrow-band MPI signal acquisition system based on mixed-frequency harmonic magnetization response. Allowing superparamagnetic nanoparticles to be excited by low frequency, high amplitude magnetic fields and high frequency, low amplitude magnetic fields, the excitation coil generates a mixed excitation magnetic field at a mixed frequency of 8.664 kHz (f_H + 2f_L), and the tracer of superparamagnetic nanoparticles can generate a locatable superparamagnetic magnetization signal with rich harmonic components in the mixed excitation magnetic field and positioning magnetic field. The third harmonic signal is detected by a Gradiometer coil with high signal-to-noise ratio, and the voltage cloud image is formed.Result: The experimental results show that the external noise caused by the excitation coil can be effectively reduced from 12 to about 1.5 μV in the imaging area of 30 mm × 30 mm, which improves the stability of the detection signal of the Gradiometer coil, realizes the detection of high SNR, and makes the detection sensitivity reach 10 μg Fe. By mixing excitation, the total intensity of the excitation field is reduced, resulting in a slight improvement of the resolution under the same gradient field, and the spatial resolution of the image reconstruction is increased from 2 mm under the single frequency excitation (20.7 kHz) in the previous experiment to 1.5 mm under the mixed excitation (8.664 kHz).Conclusions: These experimental results highlight the effectiveness of the proposed open-loop narrowband MPI technique in improving signal detection sensitivity, achieving high signal-to-noise ratio detection and improving the quality of reconstructed images by changing the excitation magnetic field frequency of the excitation coil, providing novel design ideas and technical pathways for future MPI systems.