Loading...
| Friend's email | |
| Your name | |
| Your email | |
| enter code | |
This page was sent successfuly
- Type of Document: M.Sc. Thesis
- Language: Farsi
- Document No: 42256 (19)
- University: Sharif University of Technology
- Department: Computer Engineering
- Advisor(s): Kasaei, Shohreh
- Abstract:
- Nowadays, with high performance computers and progress in video recorder devices and their related technologies, price of these devices is being balanced and recording the videos is conventional now. Processing of these videos has been a challenge in industry and science. Control and surveillance of places such as airports and roads and reducing their sizes in memory are some of their research area. Usually one of the initial preprocesses of such applications specially the object tracking is background modeling and background subtraction. This preprocess has an essential effect on the other algorithms that process based on these processes’ results. If this section conducted with fewer errors, the next process will have better performance and fewer errors. In this thesis we define three methods for background modeling and subtraction that it is better than previous algorithms. In the first method we model background with five dimensional Gaussian probability functions. Three of these dimensions are color features and others are in spatial domain. In the second method we try to combine one of the famous pixels wise methods in background modeling with our region-based foreground modeling. Finally in the third method we do some changes in the second method. With these changes, foreground model is more related to our new frame rather than previous one. In the all of these methods, especially the first method we can model highly dynamic scene and this feature is important advantage of this method. The mentioned methods in this thesis, with respect to background subtraction errors in scenes with foreground objects changes in scale and orientation and dynamic objecs, is more robust than common background modeling methods. Also, these methods have capability to define foreground objects in different applications. Algorithms are implemented in Matlab and their results have been reported in this thesis
- Keywords:
- Tracking ; Background Subtraction ; Background Modeling ; Foreground
Digital Object List
-
محتواي پايان نامه
- view
Bookmark