A5101662424- Image Enhancement Techniques
- Advanced Image Fusion Techniques
- Advanced Image and Video Retrieval Techniques
- Advanced Measurement and Detection Methods
- Remote-Sensing Image Classification
- Advanced Image Processing Techniques
- Image and Signal Denoising Methods
- Blind Source Separation Techniques
- Energy Load and Power Forecasting
- Software System Performance and Reliability
- Speech and Audio Processing
- Anomaly Detection Techniques and Applications
- Wireless Sensor Networks and IoT
- Advanced Neural Network Applications
- Remote Sensing and Land Use
- Image and Video Stabilization
- Music and Audio Processing
- Video Surveillance and Tracking Methods
- Infrared Target Detection Methodologies
- Advanced Sensor and Control Systems
- Network Security and Intrusion Detection
- Image Retrieval and Classification Techniques
- Advanced Algorithms and Applications
- Time Series Analysis and Forecasting
- Image Processing Techniques and Applications
Xi'an University of Technology
2021
Xidian University
2018-2019
Shandong University of Science and Technology
2019
China Electronics Technology Group Corporation
2012
Hefei University
2011
Northeastern University
2004-2005
Multimodal remote sensing images contain complementary information, thus, could potentially benefit many applications. To this end, the image registration is a common requirement for utilizing multimodal images. However, due to rather different imaging mechanisms, becomes much more challenging than ordinary registration, particular optical and synthetic aperture radar (SAR) In work, we design deep matching network exploit latent coherent features between patch pairs inferring their labels....
Due to the absorption and scattering, clarity observation of depth field image which is obtained by underwater photoelectric imaging will be reduced. This paper introduces a new single enhancement approach based on fusion strategy. The method first applies white balance global contrast technologies original respectively, then taking these two adapted versions as inputs that are weighted specific maps .We obtain enhanced results computing weight sum in per-pixel fashion. Since we do not...
In order to promote progress on fusion and analysis methodologies for multi-source remote sensing data, The Image Analysis Data Fusion Technical Committee organized the 2018 IEEE GRSS contest. this contest, we proposed a two-branch convolution network hyperspectral image classification with data re-sampling strategy semi-supervised learning address three existing problems, i.e. multi-scale feature learning, imbalance, small size of dataset. contest showed that our proposal achieved best...
Deep learning is a hot method in face super-resolution reconstruction recent years, but it needs to further improve the details of reconstructed images and speed up network training. This paper improves deep residual from two aspects unit structure proposes algorithm based on improved model. We by connecting with densely connected convolutional layer removing BN layer, thereby enhancing information flow between inner layers eliminate damage spatial image batch normalization processing. At...
High-precision electricity consumption forecasting can not only ensure the safe and reliable operation of power system, but also bring considerable economic benefits. Aiming at contradiction between demand for high-precision insufficiency traditional methods, this paper uses a short-term model based on long-term memory network (LSTM) to forecast daily an hourly scale in Wanda shopping mall Guangdong. First, analyze characteristics total obtain its dominant influence factors. Secondly,...
Multi-target detection and tracking are the two key issues of video surveillance. This paper presents methods to solve problems. Firstly, a moving object algorithm based on background modeling is given, which can effectively deal with disturbance illumination changes. Then, multi-targets particle filter described. Some special problems in include computational complexity, new target entry, exit occlusion also discussed. Experiment results show that presented method effective.