Loading...

Scene Detection and Analysis by Image Classification in Specific Classes

Abbasi Dinani, Mina | 2013

1379 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 45383 (05)
  4. University: Sharif University of Technology
  5. Department: Electrical Engineering
  6. Advisor(s): Gholampour, Iman
  7. Abstract:
  8. Traffic density estimation is one of the most challenging problems in Intelligent Transportation Systems. One of the important traffic information that is broadcasted to drivers is Traffic Density information. In many traffic control centers; human operators are responsible for estimating traffic density from captured video data. Increasing traffic cameras and constraint number of operators introduce an updating delay to broadcasted information. So it is important to have an automatic traffic density estimation system. In this thesis, machine vision is used to solve this problem. Supervised Image classification is our approach. In supervised Image classification, images are classified to thematic classes. The traffic density is usually divided into three main classes: light, medium and heavy. So the data is manually labeled to these three classes. Two groups of dynamic and static features are represented to describe these classes. Static features are calculated based on a single video frame. Dynamic features are extracted from a set of consecutive frames in a traffic video. Two parameters of “sample length” and “difference interval” are introduced for extracting dynamic features. These parameters are optimized for increasing classification accuracy through simulations. The chosen classifier is Support Vector Machine. Simulations have done on two traffic data sets: Tehran traffic control center dataset and University of South California dataset. The final accuracy of system on Tehran traffic control center dataset is 100%. This accuracy degrades in bad light condition to 96.6%. The obtained accuracy for university of South California dataset is 82.7%
  9. Keywords:
  10. Machine Vision ; Scene Classification ; Support Vector Machine (SVM) ; Intelligent Transportation System (ITS) ; Traffic Density Estimation

 Digital Object List

 Bookmark

...see more