A5061102708- Face and Expression Recognition
- Image Retrieval and Classification Techniques
- Face recognition and analysis
- Advanced Image and Video Retrieval Techniques
- Remote-Sensing Image Classification
- Sparse and Compressive Sensing Techniques
- Biometric Identification and Security
- Video Surveillance and Tracking Methods
- Advanced Image Processing Techniques
- Neural Networks and Applications
- Computer Graphics and Visualization Techniques
- Game Theory and Applications
- Image Processing Techniques and Applications
- Domain Adaptation and Few-Shot Learning
- Advanced Neural Network Applications
- Digital Image Processing Techniques
- Multimodal Machine Learning Applications
- Spectroscopy and Chemometric Analyses
- Image and Signal Denoising Methods
- Machine Learning and ELM
- Medical Image Segmentation Techniques
- Multi-Agent Systems and Negotiation
- Auction Theory and Applications
- Advanced Statistical Methods and Models
- Remote Sensing and Land Use
Nanjing Audit University
2020-2024
ORCID
2021
Nanjing University of Posts and Telecommunications
2015-2020
Nanjing University of Science and Technology
2011-2018
Harbin Engineering University
2018
Minjiang University
2017
Yangzhou University
2009-2014
Chinese University of Hong Kong
2010-2013
Carnegie Mellon University
2002-2003
We envision a future economy where e–markets will play an essential role as exchange hubs for commodities and services. Future should be designed to robust manipulation, flexible, sufficiently efficient in facilitating exchanges. One of the most important aspects designing e–market is market mechanism design. A defines organization, information process, trading procedure, clearance rules market. If we view multi–agent system, also structure environment which agents (buyers sellers) game....
The effect of high temperatures (above 25°C) on starch concentration and the morphology granules in grains wheat (Triticum aestivum L.) were studied. Wheat plants cultivars Yangmai 9 (weak-gluten) 12 (medium-gluten) treated with for 3 days at different times after anthesis. results showed that given a heat-shock treatment above 30°C lower than those developing normal temperature both cultivars. High lowered due to decrease amylopectin. Under same temperature, heat shock from 6 8 anthesis...
We present a robust and efficient approach to directly slicing implicit solids. Different from prior techniques that reconstruct contours on the plane by tracing topology of intersected line segments, which is actually not robust, we generate guaranteed contour extraction binary images sampled given solids subsequent simplification algorithm has preserved geometric error controlled. The resultant are free self-intersection, topologically faithful r-regular with shape bounded. Therefore,...
How to capture distinctive features from facial images when there are large variations in illumination, poses, and expressions is important for the face recognition problems. This paper introduces a novel algorithm called fuzzy linear regression discriminant projection (FLRDP) recognition. The proposed FLRDP seeks generate an efficient subspace LRC method could effectively handle between images. To be specific, first computes gradual membership degrees of each sample corresponding classes,...
Agent-based online negotiation technology has the potential to radically change way e-business is conducted. We present a formal model for autonomous agents negotiate on Internet. In our model, process driven by internal beliefs of participating agents. empirically identify relative strength group belief updating methods and show how an agent can its behavior adjusting some critical parameters. The advantage that it flexible easy implement. To different "personalities", one only needs plug...
To reduce the dependence on labeled data, there have been increasing research efforts learning visual classifiers by exploiting web images. One issue that limits their performance is problem of polysemy. solve this problem, in work, we present a novel framework solves polysemy allowing sense-specific diversity search results. Specifically, first discover list possible semantic senses to retrieve Then merge similar and prune noises using retrieved Finally, train classifier for each selected...
Recently, position-patch-based face hallucination methods have received much attention, and obtained promising progresses due to their effectiveness efficiency. A locality-constrained double low-rank representation (LCDLRR) method is proposed for effective in this paper. LCDLRR attempts directly use the image-matrix based regression model compute coefficients maintain essential structural information. On other hand, imposes a constraint on adaptively select training samples that belong same...