Title page for 945201028


[Back to Results | New Search]

Student Number 945201028
Author Wen-Tsai Sheu(³\¤å°])
Author's Email Address No Public.
Statistics This thesis had been viewed 1695 times. Download 161 times.
Department Electrical Engineering
Year 2006
Semester 2
Degree Master
Type of Document Master's Thesis
Language English
Title Foreground Object Detection based on Multi-model Background Maintenance and Its DSP Implementation
Date of Defense 2007-06-15
Page Count 68
Keyword
  • background image
  • background maintenance
  • multi-model background maintenance (MBM)
  • multi-model Gaussian distribution
  • principal features
  • Abstract Foreground object detection in a scene, often referred to as ¡§background subtraction¡¨, is a critical early in step in most computer vision applications in domains such as video surveillance, traffic monitoring, human motion capture and human-computer interaction. Background subtraction is a widely used approach for detecting moving objects from the difference between the current frame and a reference frame, often called the ¡§background image¡¨, or ¡§background model¡¨. As a basic, the background image must be a representation of the scene with no moving objects and must be kept regularly updated so as to adapt to the varying luminance conditions and some problems described in the introduction. For this reason, how to maintain a background image is very important issue.
     In the thesis, in order to acquire accurate foreground object detection with above some problems, a Multi-model Background Maintenance (MBM) algorithm is proposed. A MBM framework contains two principal features to construct a practice background image with time-varying background changes. Under this framework, the background image is represented by the most significant and recurrent features, the principal features at each pixel. Principal features consist of static and dynamic features to represent background pixels. A MBM includes two major procedures, background maintenance and foreground extraction. Experiments show proposed method provides good results on different kinds of sequences. Quantitative evaluation and comparison with the existing method show that the proposed method provides much improved results with lower complexity. Finally, we use IEKC64x platform to implement MBM algorithm for obtaining real time foreground object detection.
    Table of Content Content................................................iv
    List of Figures........................................vi
    List of Tables.........................................viii
    Chapter 1 Introduction...................................1
      1.1  Introduction...................................2
      1.2  Thesis Organization............................5
    Chapter 2 Background and Relative Research...............7
      2.1  Background: a review of background subtraction.8
        2.2.1 Nonparametric Approach......................10
        2.2.2 Parametric Approach.........................13
    Chapter 3 Proposed Multi-model Background Maintenance Algorithm................................................16
      3.1  Overview of Proposed Algorithm.................17
        3.1.1 Design Strategy.............................17
        3.1.2 Flowchart of Proposed Algorithm.............19
      3.2  Background Maintenance.........................20
        3.2.1 Change Classification.......................21
        3.2.2 Learning and Updating for Dynamic Change....22
        3.2.3 Learning and Updating for Static Point......23
      3.3  Foreground Extraction..........................25
    Chapter 4 Experimental Result and Analysis...............26
      4.1  Visual Interpretation..........................27
      4.2  Quantitative Evaluations.......................33
      4.3  Computation Complexity and Run-time Analysis...34
    Chapter 5 Introduction of DSP Platform and DSP Realization of Our Proposed Algorithm................................37
      5.1  Introduction to ATEME IEKC64x Platform.........38
        5.1.1 The TI TMS320C6146 DSP Chip.................39
        5.1.2 Central Processing Unit.....................40
        5.1.3 Memory......................................42
      5.2  TI TMS320C6416 DSP Features for Optimization...43
        5.2.1 Introduction to the Code Composer Studio Development Tools........................................43
        5.2.2 Code Optimization Flow......................45
        5.2.3 Compiler Optimization Options...............46
      5.3  Implement Our Proposed Algorithm...............50
        5.3.1 Simulation Environment......................50
        5.3.2 Implementation and Acceleration of Our Proposed MBM Algorithm on TI TMS320C6416 DSP.............52
        5.3.3 Experimental Result on DSP Implementation and Acceleration.............................................61
        5.3.4 Profiling Analysis on DSP Implementation and Acceleration.............................................62
    Chapter 6 Conclusion.....................................63
    Reference................................................65
    Reference [1] D. Gavrila, ¡§The visual analysis of human movement: A survey,¡¨ Computer. Vis. Image Understanding, vol. 73, no. 1, 1999, pp. 82-98.
    [2] E. Durucan and T. Ebrahimi, ¡§Change detection and background extraction by linear algebra,¡¨ Proc. IEEE, vol.89, Oct. 2001, pp. 1368-1381.
    [3] K. Toyama, J. Krumm, B. Brumitt, and B. Meyers¡§Wallflower: Principlesand practice of background maintenance,¡¨ in Proc. IEEE Int. Conf.Computer Vision, Sept. 1999, pp. 255¡V261.
    [4] M. Harville, G. Gordon, and J. Woodfill, ¡§Foreground segmentation using adaptive mixture model in color and depth,¡¨ in Proc. IEEE Workshop Detection and Recognition of Events in Video, July 2001, pp.3¡V11.
    [5] C. Stauffer and W.E.L. Grimson, ¡§Adaptive Background Mixture Models for Real-Time Tracking,¡¨ Proc. IEEE CS Conf. Computer Vision and Pattern Recognition, vol. 2, 1999, pp. 246-252.
    [6] R. Jain and H. Nagel, ¡§On the Analysis of Accumulative Difference Pictures from image Sequences of Real World Scenes,¡¨ IEEE Trans. Pattern Analysis and Machine Intelligence, 1979.
    [7] A. J. Lipton, H. Fujiyoshi, and R. S. Patil, ¡§Moving target classification and tracking from real-time video,¡¨ in Proc. IEEE Workshop Application of Computer Vision, Oct. 1998, pp. 8¡V14.
    [8] L. Li and M. Leung, ¡§Integrating intensity and texture differences for robust change detection,¡¨ IEEE Trans. Image Processing, vol. 11, Feb. 2002, pp. 105¡V112.
    [9] C. Wren, A. Azarbaygaui, T. Darrell, and A. Pentland, ¡§Pfinder: realtime tracking of the human body,¡¨ IEEE Trans. Pattern Anal. Machine Intell., vol. 19, July 1997, pp. 780¡V785.
    [10] D. Koller, J. Weber, T. Huang, J. Malik, G. Ogasawara, B. Rao, and S. Russel, ¡§Toward robust automatic traffic scene analysis in real-time,¡¨ in Proc. Int. Conf. Pattern Recognition, 1994, pp. 126¡V131.
    [11] K.-P. Karmann, A. Brandt, and R. Gerl, ¡§Using Adaptive Tracking to classify and Monitor Activities in a site,¡¨ Time Varying Image Processing and Moving Object Recognition, 1990.
    [12] N. Friedman and S. Russell, ¡§Image segmentation in video sequences: a probabilistic approach,¡¨ in Proc. 13th Conf. Uncertainty Artificial Intelligence, 1997.
    [13] P. KaewTraKulPong and R. Bowden, ¡§An Improved Adaptive Background Mixture Model for Real-Time Tracking with Shadow Detection,¡¨ Proc. European Workshop Advanced Video Based Surveillance Systems, 2001.
    [14] Z. Zivkovic, ¡§Improved Adaptive Gaussian Mixture Model for Background Subtraction,¡¨ Proc. Int¡¦l Conf. Pattern Recognition, vol. 2, 2004, pp. 28-31.
    [15] Q. Zang and R. Klette, ¡§Robust Background Subtraction and Maintenance,¡¨ Proc. Int¡¦l Conf. Pattern Recognition, vol. 2, 2004, pp. 90-93.
    [16] Dar-Shyang Lee, ¡§Effective Gaussian mixture learning for video background subtraction¡¨ IEEE Trans. Pattern analysis and machine intelligence, vol.27, no.5, MAY 2005, pp. 827-832.
    [17] A. Elgammal, D. Harwood, and L. Davis, ¡§Background and Foreground Modeling Using Nonparametric Kernel Density Estimation for Visual Surveillance,¡¨ Proc. IEEE, 2002.
    [18] K. Kim, T.H. Chalidabhongse, D. Harwood, and L. Davis, ¡§Background Modeling and Subtraction by Codebook Construction,¡¨ Proc. IEEE Int¡¦l Conf. Image Processing, vol. 5, 2004, pp. 3061-3064.
    [19] Y. Sheikh and M. Shah. Bayesian object detection in dynamic scenes. IEEE Conf. Computer Vision and Pattern Recognition, San Diego, CA, June 2005.
    [20] R. Cucchiara, C. Grana, M. Piccardi, and A. Prati, ¡§Detecting moving objects, ghosts, and shadows in video streams,¡¨ IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 25, no. 10, pp. 1337-1342, 2003.
    [21] B. Shoushtarian and H. E. Bez, ¡§A practical adaptive approach for dynamic background subtraction using an invariant colour model and object tracking,¡¨ Pattern Recognition Letters, vol. 26, no. 1, pp. 5-26, 2005.
    [22] I. Haritaoglu, D. Harwood, and L. S. Davis, ¡§ : Real-time surveillance of people and their activities,¡¨ IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 22, no. 8, pp. 809-830, 2000.
    [23] ¡§TMS320C6000 CPU and Instruction Set Reference Guide,¡¨ Texas Instruments, Dallas, TX, Literature Number spru189F, October 2000.
    [24] ¡§TMS320C6000 Programmer¡¦s Guide,¡¨ Texas Instruments, Dallas, TX, Literature Number spru198, August 2002.
    [25] ¡§TMS320C6000 Optimizing Compiler User¡¦s Guide¡¨ Texas Instruments, Dallas, TX, Literature Number spru1871L, MAY 2004.
    [26] F. Meyer and S. Beucher, ¡§Morphological segmentation,¡¨ J. Visual Commun. Image Representation, vol.1, pp. 21¡V46, Sept. 1990.
    [27] Luc Vincent,¡¨Morphological Grayscale Reconstruction in Image Analysis: Application and Efficient Algorithms.¡¨ IEEE Transactions on Image Processing, vol. 2, no. 2, April 1993.
    [28] ¡§TMS320C64x Image/Video Library Programmer's Reference¡¨ Texas Instructments, TX, Dallas, 2002.
    [29] ¡§TMS320C6000 Optimizing Compiler User's Guide¡§Texas Instructments, TX, Dallas, 2004.
    Advisor
  • Tsung-han Tsai(½²©vº~)
  • Files
  • 945201028.pdf
  • approve in 2 years
    Date of Submission 2007-07-16

    [Back to Results | New Search]


    Browse | Search All Available ETDs

    If you have dissertation-related questions, please contact with the NCU library extension service section.
    Our service phone is (03)422-7151 Ext. 57407,E-mail is also welcomed.