computer vision for traffic sign recognition

Download Computer Vision for Traffic Sign Recognition

If you can't read please download the document

Upload: thevijayps

Post on 25-May-2015

724 views

Category:

Education


7 download

TRANSCRIPT

  • 1. Trafc Sign Recognition for Computer Vision Project-Based LearningTripti Kumari and Vijay Pratap Singh S7, Computer Science & Engineering School of Engineering Cochin University of Science & Engineering

2. Contents Introduction Why CV for TSR Objectives Traffic Sign Classes Detection Method Complexity Approach Image Processing in Matlab Algorithms & Sudo Code Conclusion References 3. Introduction Trafc sign detection and recognition have received an increasing interest in the last years. This is due to the wide range of applications that a system with this capability provides, like Driving Assistance System. This is an attempt to make a self learning system that can itself understand and interpret the meaning of new traffic signs. 4. Why Computer Vision for Traffic Sign Highway maintenance: Check the presence and condition of signs along major roads. Sign inventory: Creating an inventory of signs in city environments. Driver support systems: Assist the driver by informing of current restrictions, limits, and warnings. Intelligent autonomous vehicles: Any autonomous car that is to drive on public roads must have a means of obtaining the current trac regulations. This can be done through TSR. 5. Objectives 1. The system has to be able to detect trafc signs independently of their appearance in the image. Because of that, it has to be invariant to: Perspective distortion. Lighting changes. Partial occlusions. Shadows.2. In addition, it has to provide information about the presence of possible problems: Lack of visibility. Bad condition. Bad placement 6. Traffic Sign Classes Guide and recreational traffic signs Regulatory traffic signs Road construction traffic signsWarning traffic signs 7. Detection Methods Trafc sign detection methods are inherently dependent on the nature of data for which they were developed. The approaches that we are going to follow in the detection stage have traditionally been divided into two kinds: 1. Color based methods. 2. Shape based methods. 8. Color Based Detection Understanding of different colors Color based segmentation 9. Shape Based Detection 10. Complexity Input type: videos or static images? Scope of the method: is the method applicable for a single trafc sign class or for multiple classes? Filming conditions: is the data shot in broad daylight, in nighttime or both? Are there adverse weather conditions such as rain, snow, fog? 11. Complexity Sensor type: high resolution or low resolution camera, grayscale or color? Multiple cameras? Other sensors? Processing requirements: should the signs be detected in realtime or is ofine processing acceptable? Acceptable true positive and false positive rates: determined by the nature of the problem. 12. Approach The general ow followed by typical sign detection algorithms. 13. Approach Collection of frames using video segmentation 14. Block Diagram 15. Flow Chart 16. Image Processing in MATLAB Basic Commands Imread Imshow Rgb2gray Imhist Imadjust im2bw Useful Libraries for this project Image Processing Toolbox Image Acquisition Toolbox Computer Vision Toolbox 17. Algorithms & Sudo Code 18. Algorithms & Sudo Code Video Segmentation in MATLABCapturing Video from External Camera (using imaqtool command) 19. Algorithms & Sudo Code Memory management for processing large video files. 20. Algorithms & Sudo Code Image Compression for minimizing the memory use. Approach 1: Read in image 2: Convert to gray scale 3: Display 4: Take Discrete Cosine Transform 5: Toss out higher order terms 6: Compare results to original picture 7: The built in function dct2 uses an FFT-like algorithm to compute transform. 21. Conclusion The algorithm that has been used for trafc signs it can be generalized to deal with other kinds of objects. The known difculties that exist for object recognition in outdoor environments have been considered. This way the system is immune to lighting changes, occlusions and object deformation being useful for Driver Support Systems. Due to this knowledge of the sign status, it is believed that the system is useful for other applications such as maintenance and inventories of trafc sign in highways and or cities. 22. References [1] Karla Brkic, Department of Electronics, Microelectronics, Computer and Intelligent Systems, Electrical Engineering and Computing, Unska 3, 10000 Zagreb, Croatia, 2011 [2] Andreas Mgelmose, Vision, Graphips, and Interactive Systems, AAU, Computer Vision and Robotics Research Lab, UCSD, 2012 [3] David Gernimo, Joan Serrat, Antonio M. Lpez, Member, IEEE, and Ramon Baldrich, Trafc Sign Recognition for Computer Vision Project-Based Learning, 2012 23. Thank-you