Loading...

Visual Event Recognition and Description

Kaviani, Razie | 2014

586 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 46470 (05)
  4. University: Sharif University of Technology
  5. Department: Electrical Engineering
  6. Advisor(s): Gholampour, Iman; Tabandeh, Mahmoud
  7. Abstract:
  8. Increasing needs for traffic video surveillance and intelligent video analysis is a hot topic that has attracted significant interests in recent days. Scene understanding is one of the most important aspects of an intelligent video surveillance system. Statistics shows that rate of vehicle accidents has been continuously increasing over the past years. This necessitates the need for efficient traffic analysis and abnormality detection systems. It is desirable to develop fully automated surveillance systems, with which people will be free from boring monitoring tasks. In this thesis our goal is to describe and detect abnormal event like accident or variation of traffic density in traffic video. For this purpose, we have used unsupervised approach and focus on topic models to automatically extract the semantic relation between motion patterns from the traffic video scenes. Then, we use these learned motion patterns to detect abnormal activities. The strength and weakness of the models are studied through a variety of measures. Using topic models allows us to define various abnormality measures based on the interpretation of the models. We compare the performance of different topic models using video data captured from two busy traffic road junctions and Tehran traffic control center dataset. The final accuracy of system on traffic junction datasets are 72.6% and 93%. The obtained accuracy on Tehran traffic control center dataset is above 97%
  9. Keywords:
  10. Intelligent Transportation System (ITS) ; Machine Vision ; Event Detection ; Event Description ; Abnormal Event Detection ; Topic Model

 Digital Object List

 Bookmark

No TOC