A5028202518- Seismology and Earthquake Studies
- Image Enhancement Techniques
- Earthquake Detection and Analysis
- Finite Group Theory Research
- Rings, Modules, and Algebras
- Human Pose and Action Recognition
- 3D Shape Modeling and Analysis
- Geochemistry and Geologic Mapping
- graph theory and CDMA systems
- Limits and Structures in Graph Theory
- Advanced Clustering Algorithms Research
- Robot Manipulation and Learning
- Soft Robotics and Applications
- Image Retrieval and Classification Techniques
- Graph Labeling and Dimension Problems
- Computer Graphics and Visualization Techniques
- Video Surveillance and Tracking Methods
- Surface Modification and Superhydrophobicity
- Anomaly Detection Techniques and Applications
- Advanced MRI Techniques and Applications
- Graph Theory and Algorithms
- Statistical and numerical algorithms
- AI in cancer detection
- Remote Sensing and LiDAR Applications
- Advanced Optical Sensing Technologies
Quanzhou Normal University
2021-2023
Quanzhou Institute of Equipment Manufacturing Haixi Institute
2018-2021
Chinese Academy of Sciences
2018-2021
Huazhong University of Science and Technology
2012
Predict (France)
2012
Dalian University of Technology
2005-2007
The classification of benign and malignant based on ultrasound images is great value because breast cancer an enormous threat to women’s health worldwide. Although both texture morphological features are crucial representations tumor images, their straightforward combination brings little effect for improving the since high-dimensional too aggressive so that drown out low-dimensional features. For that, efficient feature combing method proposed improve malignant. Firstly, (i.e., local binary...
Underwater images suffer from serious color distortion and detail loss because of the wavelength-dependent light absorption scattering, which seriously influences subsequent underwater object detection recognition. The latest methods for image enhancement are based on deep models, focus finding a mapping function subspace to ground-truth subspace. They neglect diversity conditions leads different background colors images. In this paper, we propose Class-condition Attention Generative...
The El Niño Southern Oscillation (ENSO) is one of the most important drivers interannual variations marine ecosystems, especially monsoonal ecosystems. channel effect in a strait amplifies impact monsoons and may be an ideal area to study ecosystem responses ENSO events. Previous studies have shown that physical processes Taiwan Strait (TWS) vary during different phases, but biological are still unclear. In this study, we found clear differences physical, chemical, characteristics TWS...
At present, most of the 3D printing facilities are 3 degrees freedom (DOF) that formed by planar stacking, which cannot achieve free-form surface spraying or other high DOF required operations. However, in a way, this problem can be solved using 6-axis manipulator, is future development trend facilities. In paper, UR manipulator used to realize function, especially coating procedure biomaterials, such as hydroxyapatite on artificial models. By dismantling steps our methodology, we firstly...
In order to learn and grasp the predetermined workpieces for robot actively, a recognition algorithm based on machine learning is proposed. Compared with traditional algorithms, we replenish MTSM (multi threshold space model) getting clearer workpiece shapes. To automatically compactly knowledge, both shape gradient features are designed express specific object by aid of contour mask, meanwhile, compound descriptors fed into SVM classifier they trained jointly minimize classification loss....
Recently, computer vision is widely used in the field of automatic production system, such as capsule detection, ceramic tile extraction, sole location, etc. However, changes lighting, texture background and colour material will have negative impacts on performance system. In this paper, a robust object segmentation algorithm being able to accurately extract target different conditions presented. First, image enhancement techniques are applied reduce effects illumination enhance global...
In this work, we propose a robust tracking algorithm based on context-aware correlation filter framework. order to improve the richness of feature representation, proposed hyper-feature which contains linearly weighted mixture hand-crafted features (such as HOG, color histogram) and hierarchical deep convolutional VGGNet). The final output response map is optimized by Gaussian constrained optimization method control follow distribution, gain robustness target appearance variations. addition,...
The analysis of 3D meshes with deep learning has become prevalent in computer graphics. As an essential structure, hierarchical representation is critical for mesh pooling multiscale analysis. Existing clustering-based hierarchy construction methods involve nonlinear discretization optimization operations, making them nondifferential and challenging to embed other trainable networks learning. Inspired by superpixel image processing, we extend from 2D images proposing a novel differentiable...