- Advanced Data Compression Techniques
- Advanced Vision and Imaging
- Video Coding and Compression Technologies
- Fluid Dynamics and Turbulent Flows
- Computational Fluid Dynamics and Aerodynamics
- Image and Signal Denoising Methods
- Air Traffic Management and Optimization
- Aerodynamics and Acoustics in Jet Flows
- Real-time simulation and control systems
- Geological Modeling and Analysis
- Computational Geometry and Mesh Generation
- Aerosol Filtration and Electrostatic Precipitation
- Infrastructure Maintenance and Monitoring
- Concrete Corrosion and Durability
- Industrial Vision Systems and Defect Detection
- Aerodynamics and Fluid Dynamics Research
- Plasma and Flow Control in Aerodynamics
- Asphalt Pavement Performance Evaluation
- Advanced Numerical Analysis Techniques
- Robotics and Sensor-Based Localization
- Software Reliability and Analysis Research
- Remote Sensing in Agriculture
- Simulation Techniques and Applications
- Remote Sensing and Land Use
- Software Engineering Research
Shenzhen Polytechnic
2023-2024
Jilin University
2010-2012
Boeing (United States)
1983-2000
James S. McDonnell Foundation
1996
Cross-project software defect prediction (CPDP) refers to the construction of models by collecting multi-source project data, but heterogeneity data among projects and modern problem “data islands” hinder its development. In response these challenges, we propose a CPDP algorithm based on differential perception combined with inheritance federated learning (FedDPI). Firstly, design an efficient preprocessing scheme, which lays reliable foundation for integrating oversampling optimal feature...
Multiview video (MVV) is multiple sequences that integrated different viewpoints data of the same three-dimensional (3D) scene. Each viewpoint are taken from ordinary camera. Thus, very large for MVV. So compression necessary in order to store and transmit effectively. Based on theory multi-dimensional vector matrix (MDVM), we propose a six-dimensional (6D) orthogonal transform nuclear matrix, prove its orthogonality energy concentration. We apply multiview coding (MVC). This transformation...
This paper extends the theory of multi-dimensional vector orthogonal transformation matrix and presents a new 4D order-4 DCT integer transform operator based on discrete cosine (MD-VMDCT). Meanwhile, orthogonality energy concentration is verified in paper. Also comparison between float 4D-VMDCT carried out. At last, video sequence compressed using our approach. The experimental results show that algorithm correct effective. It better than H.264/AVC under same conditions it slightly lower...
An accurate, comprehensive interactive boundary-layer (IBL) method has been coupled with an analytic-based Euler equation solution for the analysis and design of airfoils. The goal this development effort is efficient process which includes all necessary physics can ultimately be applied to high-lift systems. It takes full advantage efficiency afforded by inverse predicting viscous effects. improved Cebeci-Smith eddy viscosity formulation computes transition onset using either e-method or...
The detection and prevention of concrete cracks spalls is crucial to ensure the structural integrity longevity civil infrastructure. In this research paper, we propose a novel method for crack spall based on YOLOv8 with ByteTrack supervision. proposed exploits advantages object framework, which provides real-time accurate detection, tracking counting various objects. By integrating ByteTrack, state-of-the-art algorithm, increase system performance robustness in over time. Furthermore,...
Nowadays, the 3D discrete cosine transform (DCT) is applied widely in video coding. But matrix of DCT expressed with floating-point numbers, so computational complexity high, and more system resources are occupied. In addition, accomplished by operating 1D to rows, columns pages data successively, which cannot embody overall space performance well. To overcome these drawbacks, integer submatrix (SDCT) method was proposed paper. First, several operation methods were defined. Then, basic...
This paper proposes a technique for generating the quantisation cube and an improved zigzag scanning method suitable to multidimensional vector matrix DCT integer transform (MD-VMICT) codec after studying statistical properties of DC AC coefficients. An e-exponential function is used it verified two-exponential in order easily carrying out by shift operator. After determining proper parameter experiments, proposed scan are tested on various standard test video sequences. The experiments show...
Chip package testing is a key process to eliminate bad products in the electronics industry. Aiming at problems of YOLOv3 chip detection complex environments with low accuracy and large number model parameters, an improved automatic method based on EMO, called EMO-YOLOV3, was proposed. This uses EMO replace backbone Darknet53 YOLOv3, inherits efficiency CNN short-range dependencies dynamic modeling capability Transformer learn long-distance interactions. The results show that has very good...
With the rapid development of artificial intelligence, emergence intelligent vehicles has brought disruptive changes to traditional survey technology. In response limitations geological environmental factors on scope and search progress field exploration, this article utilizes a modular design concept develop system structure framework specific solutions for each functional module multi-functional vehicle. An control platform based Nvidia Jetson TX2 ROS systems been built, software programs...
Nowadays, a well-established video coding method is based on the block-matching algorithm that in core of all MPEG and H.26x standards. However, this involves motion estimation compensation; thus computational complexity high. three-dimensional discrete cosine transform (3D DCT) potential method, scanning order quantisation 3D DCT coefficients play crucial role effect. So, statistical performances are studied paper, then adaptive proposed. The theoretical analysis experiment results show...
One of the main problems related to unsupervised change detection methods based on ‘difference image’ lies in lack efficient automatic techniques for discriminating between changed and unchanged pixels difference image. Such discrimination is usually performed by using empirical strategies or manual trial-and-error procedures, which affect both accuracy reliability process. To overcome such drawbacks, this paper, we propose an (based clustering characteristic 3D histogram) analysis The...