title band














主页 | Main Page























































  

Projects

     The Current Projects  |   The completed Projects      




























  

  The Completed Projects:






























1. The research on middle-level vision representation and high-level action recognition for video
 
2014.01-2016.12
Project Funding: National Natural Science Foundation of China

Contact
Dr. Zhi Han















In this  project , we take refining middle-level vision representation as the core destination, and study a comprehensive, effective low-level primitive dictionary and component tracking method based on human vision system mechanism, such as Gestalt principle. Given interrelated middle-level vision representation and features in both spatial and temporal tunnels, the middle-level representation can be applied to video coding, compression and synthesis.With further study on the connection with high-level vision characteristics, a middle-level based action template can be obtained with machine learning and an effective action recognition algorithm will be proposed.































2. Fine robot teleoperation and human-computer interactive control system
    
2013.01-2015.12

Contact: Mr.Wentao Li, Prof. Dr. Yandong Tang

Project Funding: The Ministry of Science and Technology of P.R. China















Development of Visual Servo System including 3D scene reconstruction,target relative positioning and orientating































3. Unified Abnormal Event Detection via Sparse Representation and Dictionary Selection
2012.01-2014.12
Project Funding: National Natural Science Foundation of China

ContactProf. Dr.Yang Cong















AB































4. Physical and imaging properties of outdoor shadow and its removal methods
 
2012.01-2014.12
Project Funding: National Natural Science Foundation of China

ContactDr. Jiandong Tian, Assistant Professor, (PI)















SR SR1































5. The Gastroscope automatic robot operation and assisted diagnostic system

2012.01-2014.12
responsible for development of computer assisted diagnosis.
 
Project Funding: The Ministry of Science and Technology of P.R. China
 
Contact Prof. Dr. Yang Cong, Prof. Dr. Yandong Tang
             















NK  nk1































6. Video based fault detection system of High Voltage Transmission Lines
2011.01-2013.12

Project Funding: State Grid Corporation of China

 
Contact Dr. Yandong Tang, Professor, (PI)
             
Dr. Yang Cong, Associate Professor (Co-PI)















SG
































7. Vision based locating and measuring for remote control

2010.01-2013.12
 
Project Funding: Chinese Academy of Sciences
 
ContactDr. Yandong Tang, Professor; Dr. Yingkui Du, Associate Professor
















RC VD































8. The visual detection technology on high-speed precision manufacturing production line















development of high-speed object detection and recognition
    
2009.01-2012.12
Project Funding: National Natural Science Foundation of China

ContactProf. Dr. Yandong Tang  
















NSFC1





























 

9. Fundus image analysis based on variational principal and level sets
    
2009.01-2011.12
Project Funding: National Natural Science Foundation of China

ContactProf. Dr. Yandong TangDr. Linlin Zhu















  
   For the development of key technologies in the pathology research and computer assisted diagnosis of eye diseases, in this project, with level set theory and partial differential equations we proposed more accurate and reliable the methods for automatic fundus image segmentation,image matching,and 3D reconstruction. These methods include vessel segmentation, papilla locating and segmentation and optic cup segmentation. By computing the ratio of optic cup and optic disc, the width variation of vessel across the boundary of optic disc, the corresponding curves of evolution of eye diseases can be given to assist early diagnosis of ophthalmologist.















     fundus analysis































10. The research on high-speed Visual Odometry

2009.01-2011.12

Project Funding: National Natural Science Foundation of China
 
ContactDr. Yandong Tang, Professor, Dr. Yingkui Du,
















  In this project, we completed the development of a visual odometry sysytem for robot navigation. This system was testified in various environments, such as on sands, asphalt pavement and lawn. the experiments show that the relative locating error is less then 3% for the mileage of 50 meters. 
















    vodometry































 
11. Flame Front and Feature Extraction from PLIF Image based on Level Sets
 
2009.01-2011.12

Project Funding: National Natural Science Foundation of China
 
ContactDr. Huijie Fan,  Dr. Wei Dong

Project cooperation partner: The Institute of Engineering Thermophysics, Chinese Academy of Sciences
















  
   For flame sequential images obtained by Planar Laser Induced Fluorescence (PLIF) imaging, we establish several different segmentation methods,such as line enhanced active contour and local scalable active contour, to obtain accurate flame front with the accurate gap boundaries in flame front. These methods perform well on various types of flame images with sharp edges, cusps and long gaps. The the method curvature calculation of flame front is designed with level set theory. With the methods mentioned above, the system of automatic flame front extraction and statistical analysis is developed as a analysis tool for turbulent combustion.
















     PLIF































12. Onboard automatic target tracking system for UAV

2008.01-2009.12

Project Funding: The State Key Laboratory of Robotics

ContactProf. Dr. Yandong Tang,  Dr. Yingkui Du, Dr. Baojie Fan















   
  With DSP platform, we completed the development of Onboard automatic target tracking system for UAV. Its efectiveness was testified by real flying tests on tracking some moving objects, such as moving car, and other static objects on the ground. 















     atr































13. The system of automatic early forest smoke/fire detection and alarming

2007.01-2008.12
Project Funding: The State Key Laboratory of Robotics

Contact
Prof. Dr. Yandong TangMr. Qunhui Zheng
















Functions:
1) Tower-based automatic & reliable early recognition
of forest smoke and fire;2) Automatic recognition of smoke clouds and fire with a minimal size 25x25 pixels by day and night; 3) The Farthest monitoring at a distance of up to 5 km.
















     smoke detection 















































































logo