A5054974260- Video Surveillance and Tracking Methods
- Advanced Image Processing Techniques
- Advanced Vision and Imaging
- Image and Signal Denoising Methods
- Digital Media Forensic Detection
- Remote Sensing and Land Use
- Access Control and Trust
- Advanced Database Systems and Queries
- Advanced Neural Network Applications
- Advanced Decision-Making Techniques
- Face and Expression Recognition
- Advanced Algorithms and Applications
- Network Security and Intrusion Detection
- Image Processing Techniques and Applications
- Fault Detection and Control Systems
- Advanced Steganography and Watermarking Techniques
- Video Analysis and Summarization
- Data Mining Algorithms and Applications
- Data Management and Algorithms
- Fire Detection and Safety Systems
East China University of Science and Technology
2024
Wuhan University of Technology
2018-2019
To solve the problem of incipient fault detection, a targeted gated recurrent unit-canonical correlation analysis (CCA) method is proposed. First, this article proposed unit (FTGRU) to establish temporal feature extraction model. The features extracted by FTGRU are more sensitive faults, thus increasing accuracy detection Then, model established CCA method. In addition, in order ensure universality model, multilayer strategy At first layer, basic used. When no detected at second layer...
Digital watermarking technology is of great importance to protect the copyright owners and authenticate security media. As digital images are vulnerable some common attacks during transmission, it very necessary design a algorithm which can resist all kinds attacks. Nowadays, most algorithms published rely much on locations pixels for watermark embedding, results in less robustness. And took measures increase ability attacks, but taken limited embedding capacity. In this paper, robust based...
The Bayesian theory provides a new solution to image super-resolution reconstruction. In view of the poor robustness noise and motion estimation in vast majority superresolution reconstruction algorithms. this paper, we propose an algorithm based on representation. proposed algorithm, uncharted images, parameters unknown model are utilized for modeling hierarchical framework. We adopt degenerate distribution derive analytic solutions applied which also enables robust noises. experimental...
Multi-join query optimization is an important technique for designing and implementing database manage system. It a crucial factor that affects the capability of database. This paper proposes new algorithm to solve problem multi-join based on parallel ant colony optimization. In this paper, details used have been interpreted, including how define heuristic information, implement local pheromone update global design state transition rule. After repeated iteration, reasonable solution...
The basic features extracted by the LBP operator are not complete. It only analyzes symbolic values of local differences. How to improve and how effectively express information that fails achieve a better recognition rate has yet been solved. By selecting FP-tree, grey value image is frequently analyzed, then feature used extract features, so as obtain more comprehensive facial image.
This paper proposes a perceptual hash algorithm based on the combination of Scale-invariant feature transform (SIFT) and cluster analysis (K-means). SIFT is used for extracting local stabilized points images. K-means compressing data obtain image hash. Assess similarity between video frame by means Hamming distance among value. Through experimental data, results show that has good robustness success rate.
Cloud computing security issues are increasingly serious. Behavioral trust is the relationship established by entity based on its past experience of interaction. It can be either direct interaction or indirectly recommended third parties. User behavior assessment using analytic hierarchy process, collecting historical information to obtain user's evidence, according characteristics evidence divided into different attributes which, through process and relative weight each attribute calculated...
Face super-resolution has attracted much attention in recent years. Many algorithms have been proposed. However, these approaches only perform well under the condition that input is noiseless or small noise. These existing think measure of between LR and HR Euclidean distance. distance cannot image's pixel values tend to random fluctuations varying magnitudes image represented as sparse high-dimensional data. In this paper, we propose a novel representation Wasserstein based face approach...