-
- MyTSU |
- AskTSU |
- Make a Gift |
- Contact Us |
- Webmail |
- Directory
Vinayak Elangovan
Ph.D Candidate & Research Assistant
|
Tennessee State University |
|
Education:
• Master’s in Computer Information and Systems Engineering, Tennessee State University (August 2008)
• Bachelor of Engineering in Instrumentation & Control at Madras University, India ( April 2003)
Professional Experience:
• Software Programmer, Peri Software Solutions, Newark, New Jersey ( Sept 2008 – June 2009)
• Research Assistant, Tennessee State University, Nashville, Tennessee (Jan 2006 – July 2008)
• Instrumentation Engineer, Engineering Construction Contracts (ECC), Larsen & Toubro (L&T), India (Nov 2003 - Feb 2005)
Project Scope:
The improved Situational awareness in Persistent Surveillance Systems (PSS) is an ongoing research effort of the Department of Defense. Most PSS generate huge volume of raw data and they heavily rely on human operators to interpret and inference data in order to detect potential threats. Human Activity recognition is a challenging problem in due to complexity involved in low-level processing, data alignment from different sensor sources, and fusion of soft (e.g., human intelligence) and hard (physical sensors) data, and interpretation of fragmented, yet spatiotemporally correlated and associated information that enhance Situational Awareness (SA). Human activities may involve human-human interactions, human-object interactions, human-environment interactions or multi-human-object interactions or multi-human-multi-object interactions. Identifying human interactions as a group for high level activities have not been well addressed by previous researchers. High level activities may include operational activities involving Loading and unloading of objects, exchange of baggage, exchange of vehicles etc. Vehicles have been used as a primary source of transportation for pursuing many outdoor suspicious activities. Less attention has been devoted by previous researchers to discovery of human vehicle interactions via fusion of soft and hard sensor data.
The goal of this research work is to contribute to the advancement of Human Group Activity Interaction discovery and recognition in PSS for related DOD and DHS applications. In our research, image processing technique is used as the primary source of sensing modality. For detecting and characterizing the events, and converting detected events to recognizable observations, a Casual Event State Inference (CESI) model is proposed. A Modified Sequential - HMM (MS-HMM) in conjunction with developed ontologies is used for predicting the group activities. The MS-HMM model maps the sequential observations with the ontology library in building up its model. The end results as a semantic messages generated describing the human activities.
Research Applications:
Developed model can be applied for Battlefield Intelligence, Homeland Security applications, Border Monitoring and many other civilian applications.
Publications:
-
Elangovan,V., and Shirkhodaie,A., “Recognition of Human Activity Characteristics based on State Transitions Modeling Technique,” SPIE Defense, Security and Sensing, Baltimore, Maryland, April (2012).
pdf
-
Elangovan,V., Bandaru,V.K., and Shirkhodaie,A., “Team Activity Analysis and Recognition Based on Kinect Depth Map and Optical Imagery Techniques,” SPIE Defense, Security and Sensing, Baltimore, Maryland, April (2012).
pdf
-
Elangovan,V., and Shirkhodaie,A., “Context-Based Semantic Labeling of Human-Vehicle Interactions in Persistent Surveillance Systems,” SPIE Defense, Security and Sensing, Orlando, Florida, April (2011).
pdf
- Shirkhodaie,A., and Elangovan,V., “Acoustic Semantic Labeling and Fusion of Human-Vehicle Interactions,” SPIE Defense, Security and Sensing, Orlando, Florida, April (2011).
-
Elangovan,V., and Shirkhodaie,A., “A Survey of Imagery Techniques for Semantic Labeling of Human-Vehicle Interactions in Persistent Surveillance Systems,” SPIE Defense, Security and Sensing, Orlando, Florida, April (2011).
pdf
- Sabatto, S.Z., Elangovan, V., Chen, W., Mgaya, R. (2009). Localization Strategy for Large-Scale Airborne Deployed Wireless Sensors. Submitted paper for Institute of Electrical and Electronics Engineers (IEEE) Symposium Series on Computational Intelligence (SSCI) in Multicriteria Decision Making (MCDM), Nashville, TN.
- Elangovan, V., and Sabatto, S. Z. (2008). Development of Software for Localization of Stationary Wireless Sensor Nodes. Published abstract and oral presentation at Institute of Electrical and Electronics Engineers (IEEE) Southeast Conference, Huntsville, AL.
Research Advisor:
Dr. Amir Shirkhodaie
Director, Center of Excellence for Battlefield Sensor Fusion
Tennessee State University
Dept. of Mechanical and Manufacturing Engineering
3500 John A. Merritt Blvd., Nashville, TN 37209
Tel: 615-963-5396
webpage contact:
ITMRL
Tennessee State University is an EEO/AA employer and does not discriminate on the basis of race, color, national origin, gender, age, religion or disability in its programs and activities. | more...
UA-23157491-1