A5001996145- IoT and Edge/Fog Computing
- Service-Oriented Architecture and Web Services
- Blockchain Technology Applications and Security
- Business Process Modeling and Analysis
- Privacy-Preserving Technologies in Data
- Cloud Computing and Resource Management
- Traffic control and management
- Traffic Prediction and Management Techniques
- Transportation Planning and Optimization
- Software System Performance and Reliability
- Advanced Software Engineering Methodologies
- Advanced Image Processing Techniques
- Infrastructure Maintenance and Monitoring
- Asphalt Pavement Performance Evaluation
- Vehicular Ad Hoc Networks (VANETs)
- Generative Adversarial Networks and Image Synthesis
- Ideological and Political Education
- Face recognition and analysis
- Liver Disease and Transplantation
- IoT Networks and Protocols
- Machine Learning and ELM
- Morinda citrifolia extract uses
- Tea Polyphenols and Effects
- Caching and Content Delivery
- Liver Disease Diagnosis and Treatment
Southwest Forestry University
2018-2025
Kunming University of Science and Technology
2024
Wuhan University
2020
Yunnan University
2010-2020
Northwest A&F University
2017
Shanghai Power Equipment Research Institute
2016
Army Medical University
2012
Southwest Hospital
2012
Shanghai Institute of Technology
2012
Hubei Normal University
2010-2012
Cloud computing is a formidable paradigm to provide resources for handling the services from Industrial Internet of Things (IIoT), such as meteorological industry. Generally, services, with complex interdependent logics, are modeled workflows. When any nodes hosting workflows fail, all sorts consequences (e.g., data loss, makespan enlargement, performance degradation, etc.) could arise. Thus recovering failed tasks well optimizing and load balance still critical challenge. To address this...
Smart city is gradually forming a large scope of Internet Things (IoT) networks with diffusely deployed IoT devices that produce quantities services. Considering the large-scale and widely distributed features networks, edge computing emerged as powerful suitable paradigm to provide abilities for at networks. In computing, services could be placed on units (ECUs) execution, which provides low latency eases burden bandwidth. However, it still challenging improve overall ECU execution...
With the construction and development of smart cities, accurate real-time traffic prediction plays a vital role in urban traffic. However, data has characteristics nonlinearity, nonstationary, complex structure, so always been challenging problem. The traditional statistical model is good at dealing with linear poor nonlinear data. Although ability to capture improved, deep learning approach difficulty meeting requirements prediction. To solve above challenges, we propose novel based on...
The cloud computing paradigm provides massive storage and rich resources for workflow deployment implementation. Nevertheless, applications (e.g., meteorological prediction financial analysis) are usually data intensive, substantial with privacy information tend to be accessed during the Therefore, it remains challenging design a placement method seeking tradeoffs among multiple performance metrics, i.e., resource usage, acquisition time, energy cost, while avoiding conflicts of...
Benefiting from the powerful data analysis and prediction capabilities of artificial intelligence (AI), on edge is often transferred to cloud center for centralized training obtain an accurate model. To resist risk privacy leakage due frequent transmission between cloud, federated learning (FL) engaged in paradigm, uploading model updated server (ES) central aggregation, instead transferring directly. However, adversarial ES can infer update other ESs aggregated may still expose some...
The Internet of Vehicles (IoV) environment consists a number latency-critical and data-intensive application (e.g., real-time video analytics). In this article, we posit the potential leveraging sixth-generation (6G) mobile networks to minimize communication delay, particularly for task execution. particular, 6G-enabled network in boxes (NIBs) deployed vehicles can communicate real time with edge servers or NIBs other vehicles. Although are capable providing dynamic flexible computing...
Choreography-driven microservice composition has provided a better way to integrate components in the Cyber-physical-Social System (CPSS). Choreography is global contract that specifies interactions among microservices participating composite service. After modeling choreography, problem arises here whether choreography specification at design time can be implemented correctly by generated interact with each other via exchanging messages. In this paper, we propose novel approach for...
In face image recognition and other related applications, incomplete facial imagery due to obscuring factors during acquisition represents an issue that requires solving. Aimed at tackling this issue, the research surrounding completion has become important topic in field of processing. Face methods require capability capturing semantics expression. A deep learning network been widely shown bear ability. However, for high-resolution completion, training inpainting is difficult converge, thus...
Abstract The majority of existing face inpainting methods primarily focus on generating a single result that visually resembles the original image. generation diverse and plausible results has emerged as new branch in image restoration, often referred to “Pluralistic Image Completion”. However, most diversity simply use random latent vectors generate multiple results, leading uncontrollable outcomes. To overcome these limitations, we introduce novel architecture known Reference-Guided...
Pavement crack detection is crucial for ensuring road safety and reducing maintenance costs. Existing methods typically use convolutional neural networks (CNNs) to extract multi-level features from pavement images employ attention mechanisms enhance global features. However, the fusion of low-level introduces substantial interference, leading low accuracy small-scale cracks with subtle local structures varying morphologies. In this paper, we propose a computationally efficient deep learning...
Wood species recognition is an important work in the wood trade and commercial activities.Although many methods were presented recent years, existing mainly use shallow models with low accuracy are still unsatisfying for real-world applications.Besides, their generalization ability not strong.In this paper, a novel deep-learning-based method was proposed, which improved greatly.The uses 20X amplifying glass to acquire images, extracts image features ResNet50 neural network, refines linear...
Traffic flow prediction plays a critical role in reducing traffic congestion transportation systems. However, accurate becomes challenging due to the impact of complex spatio-temporal (ST) correlations and diversity ST correlations. When modeling complicated correlations, researchers usu did not take into consideration, resulting poor accuracy. In this paper, we propose ST-InNet, deep Inception network for collectively predicting each city region. Specifically, ST-InNet employs two networks...
Distributed Collaborative Machine Learning (DCML) has emerged in artificial intelligence-empowered edge computing environments, such as the Industrial Internet of Things (IIoT), to process tremendous data generated by smart devices. However, parallel DCML frameworks require resource-constrained devices update entire Deep Neural Network (DNN) models and are vulnerable reconstruction attacks. Concurrently, serial suffer from training efficiency problems due their nature. In this paper, we...
The roots of Codonopis bulleynana Forest ex diels (cbFed), locally known as Tsoong, have been used a tonic food. Tsoong has wide range pharmacological effects, including anticancer efficacy. In the present study, activity and its potential molecular mechanisms were investigated. Isorhamnetin, flavonol aglycone, is important compound metabolite in Tsoong. It can promote apoptosis colon cancer cells through up-regulating apoptosis-related genes (Apaf1, Casp3 Casp9) because it blocks Hsp70...
Collaborative business processes gather a set of with complementary competencies and knowledge to cooperate achieve more successes. To ensure their successful implementation, correctness is key issue that needs be addressed during development. this end, novel enforcement approach proposed support the development collaborative processes. In approach, we first give an algorithm check original process specified by Petri nets. Then, prune its reachability graph obtain core in case partially...
Interactions in microservice systems are complex due to three dimensions: numerous asynchronous interactions, the diversity of communication, and unbounded buffers. Analyzing such interactions is challenging. In this paper, we propose an approach for interaction analysis using model checking techniques, which supported by Process Analysis Toolkit (PAT) tool. First, use Labeled Transition Systems (LTSs) behaviors as sequences send actions under synchronous communications. Second, introduce a...
Burn injuries are severe problems for human. Accurate segmentation burn wounds in patient surface can improve the calculation precision of %TBSA (total area), which is helpful determining treatment plan. Recently, deep learning methods have been used to automatically segment wounds. However, owing difficulty collecting relevant images as training data, those cannot often achieve fine segmentation. A image-generating framework proposed this paper generate image datasets with annotations...