A5100731491- Advanced Image and Video Retrieval Techniques
- Cloud Data Security Solutions
- Advanced SAR Imaging Techniques
- User Authentication and Security Systems
- Synthetic Aperture Radar (SAR) Applications and Techniques
- Advanced Malware Detection Techniques
- Advanced Data Storage Technologies
- Security and Verification in Computing
- Parallel Computing and Optimization Techniques
- Distributed and Parallel Computing Systems
- Neural dynamics and brain function
- Human Pose and Action Recognition
- Advanced Decision-Making Techniques
- Chaos-based Image/Signal Encryption
- Advanced Neural Network Applications
- Caching and Content Delivery
- Pickering emulsions and particle stabilization
- Distributed systems and fault tolerance
- IoT and Edge/Fog Computing
- Simulation and Modeling Applications
- Numerical Methods and Algorithms
- Underwater Acoustics Research
- Blind Source Separation Techniques
- Robotics and Sensor-Based Localization
- Advanced Vision and Imaging
Beijing Institute of Technology
2016-2024
Xi’an University of Posts and Telecommunications
2023
China Electronics Technology Group Corporation
2023
Soochow University
2021
Zhejiang University of Technology
2015
Dalian Maritime University
2014
This letter proposes a synthetic aperture radar (SAR) image registration method named feature-area optimization (FAO). First, the traditional area-based model is reconstructed and decomposed into three key but uncertain factors: initialization, slice set, regularization. Next, structural features are extracted by scale-invariant feature transform (SIFT) in dual-resolution space (SIFT-DRS), novel SIFT-like dedicated to FAO. Then, factors determined based on these features. Finally, solving...
With the rapid development of information technology, Internet Things (IoT) is profoundly impacting various fields socio-economic landscape, driving transformation traditional industries towards intelligence. However, widespread application IoT has also led to a surge in electronic evidence, whose importance judicial field increasingly prominent, but its characteristics such as ease replication and leakage have posed new challenges for privacy protection. This paper focuses on security...
With the increased snapshot operations (creation or deletion), performance of storage system will degrade severely whether in Copy on write (COW) Redirect (ROW). This paper presents a novel scheme which could use resources effectively and keep historical data accurately for SSDHDD-hybrid system. SSD (source volume) stores all active executes read/write without responding to operations. In contrast, HDDs (snapshot store writing sequentially form logs are responsible backup this way, hybrid...
With the development of satellite load technology and very large-scale integrated (VLSI) circuit technology, onboard real-time synthetic aperture radar (SAR) imaging systems have facilitated rapid response to disasters. The state-of-the-art System-on-Programmable-Chip (SoPC) technique, associated with embedded processor other function modules, provides a potential solution satisfy all constraints. However, improvement processing efficiency imagery granularity, implement an entire SAR using...
In order to classify the remote sensing land-use objects efficiently with very high resolution (VHR) images, this letter proposes a object classification method based on bag of visual words (BOV) model. This combines scale invariant feature transform (SIFT) and texture as words. Then, are used for generating word frequency histograms. The histogram is bridge classifier. At last, in classifier design section, intersection kernel (HIK) adopted SVM. We use proposed UC Merced dataset self-made...
The similarity metric in Loop closure detection (LCD) is still considered an old fashioned way, i.e. to pre-define a fixed distance function, leading limited performance. This paper proposes general framework named LRN-LCD, Lightweight relation network for LCD, which combines the feature extraction module and into simple lightweight network. end-to-end framework, can learn non-linear deep detect loop closures from different scenes. Moreover, LRN-LCD supports image sequences as input speed up...
We investigate the timestamp allocation scheme in classical concurrency controls of database management systems (DBMS) on many-core machines. Then we discuss a distributed logical with uniqueness and fairness to improve performance DBMS control algorithms Further, proposed generator is free bottlenecks such as accessing system clock counter, calling for atomic add operation, synchronization. Finally, experiment an optimistic algorithm based other schemes. The results show that outperforms...