A5041458559- Cloud Computing and Resource Management
- Distributed and Parallel Computing Systems
- IoT and Edge/Fog Computing
- Scientific Computing and Data Management
- Fire Detection and Safety Systems
- Face and Expression Recognition
- Big Data and Business Intelligence
- Non-Destructive Testing Techniques
- Machine Fault Diagnosis Techniques
- Robotics and Sensor-Based Localization
- Auditing, Earnings Management, Governance
- Advanced Data Storage Technologies
- Image Retrieval and Classification Techniques
- Machine Learning and Data Classification
- Quality and Safety in Healthcare
- Concrete Corrosion and Durability
- Data Stream Mining Techniques
- Advanced Database Systems and Queries
- Algorithms and Data Compression
- Reliability and Maintenance Optimization
- Scheduling and Optimization Algorithms
- Neural Networks and Applications
- Accounting and Organizational Management
- Advanced Image and Video Retrieval Techniques
- Coal Properties and Utilization
Wuhan College
2024
Jiaxing Hengchuang Electric Design and Research Institute
2022
Naval University of Engineering
2021
National University of Defense Technology
2013-2016
Harbin Institute of Technology
2016
Remaining useful life (RUL) is a metric of health state for essential equipment. It plays significant role in management. However, RUL often random and unknown. One type physics-based method builds mathematical model using prior principles, but this tough task real-world applications. Another estimates from available information through condition monitoring; known as the data-driven method. Traditional methods require human effort designing features to represent performance degradation, yet...
Timely and accurate detection of the initiation expansion crack is great significance for improving safe operation civil infrastructures. Image-based visual surface inspection has been an indispensable way long-time infrastructure monitoring. However, existing methods generally suffer from interference complex background, leading to obvious performance drops. To tackle this, improved encoder-decoder architecture based on SegNet proposed in this paper, namely crack-SegNet. The encoder network...
Consolidation and overbooking are two important methods to improve the resource utilization in cloud computing. In environment, resources occupied by virtual machine restricted guarantee performance fairness between multi-tenants. However, interaction among co-host applications results degradation because of sharing competition. Previous studies about interference mainly focused on traditional environment didn't take restriction into account. order predict effect which can give suggestions...
Convolutional Neural Network (CNN) is a type of feed-forward artificial neural network, exploiting the unknown structure in input distribution to discover good representations with multiple layers small neuron collections. CNN uses relatively little pre-processing compared other classification algorithms, usually gradient decent updates parameters network. Since was introduced 1997s deal face recognition, it has made much achievement many fields, and been state-of-the-art method speech etc....
Scale Invariant Feature Transform (SIFT) method plays a critical role in wide variety of vision applications. But it is now facing the real-time computational challenge. Parallel computing one most promising solutions to overcome In this paper, we target at parallelizing SIFT by multi-core architecture with per-core SIMD support. We focus on SIMDization data parallel parts fully utilize power. At Orientation Assignment and Key point Descriptor stages, observe that load balance an important...
Hadoop has become an attractive platform to store large-scale data in HDFS and perform analytics using MapReduce framework. However, dataset of multi-field is usually stored just one-dimension. Analytics need process the whole a brute way. In this paper, we introduce HView, extension layout HDFS, according multiple fields. HView provides people with different dimension views same HDFS. does not modify Hadoop, increase DataNode storage occupy or bring Namenode pressure. We exploit use case...
The spontaneous combustion points of coal piles (SCP) in closed yards has the characteristics dynamic expansion. Once multi-SCP arise at same time, choice fire-fighting disposal path will seriously affect loss and power generation cost. In response to this problem, paper proposes a cooling planning method based on genetic algorithm. composition yard environmental factors are used determine diffusion coefficient, combined with parameters fire water cannon movement output, model over time is...
Abstract The phenomenon of spontaneous combustion coal piles often occurs in closed yards power plants, automatic fire-fighting equipments are generally used to detect and locate the point. However, their performance (false alarm rate) largely limited due complicated environment yard. To deal with this problem, paper proposes a composite image recognition based method for detecting positioning point (SCP) Yolov5s is trained efficiently SCPs infrared information fused distinguish realiability...