1 / 24

Real-Time Canny Edge Detection Parallel Implementation for FPGAs

17th IEEE International Conference on Electronics, Circuits, and Systems ICECS 2010 – Athens - Greece. Real-Time Canny Edge Detection Parallel Implementation for FPGAs. Christos Gentsos 1 , Calliope-Louisa Sotiropoulou 1 , Spiridon Nikolaidis 1 and Nikolaos Vassiliadis 2.

kareem
Download Presentation

Real-Time Canny Edge Detection Parallel Implementation for FPGAs

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. 17th IEEE International Conference on Electronics, Circuits, and Systems ICECS 2010 – Athens - Greece Real-Time Canny Edge Detection Parallel Implementation for FPGAs Christos Gentsos1, Calliope-Louisa Sotiropoulou1, Spiridon Nikolaidis1 and Nikolaos Vassiliadis2 1Electronics Lab, Physics Dept., Aristotle Univ. of Thessaloniki, Greece 2Micro2Gen Ltd. , NCSR Demokritos, Greece

  2. Outline • Motivation • Canny Algorithm • Proposed Canny Implementation • Simulations – Results • Conclusions C.-L. Sotiropoulou – Real-Time Canny FPGA Implementation – AUTH-eLab

  3. Outline • Motivation • Canny Algorithm • Proposed Canny Implementation • Simulations – Results • Conclusions C.-L. Sotiropoulou – Real-Time Canny FPGA Implementation – AUTH-eLab

  4. Motivation • Necessity of Edge Detection • First step in many computer vision algorithms • Identification of sharp discontinuities in an image • Why use the Canny algorithm • Good performance in images contaminated by noise • The need for Real-Time/High Throughput Implementation • Multiplication of camera resolutions in recent years • Real-time applications • The performance of modern FPGA devices • Powerful, efficient, availability of memory resources C.-L. Sotiropoulou – Real-Time Canny FPGA Implementation – AUTH-eLab

  5. Outline • Motivation • Canny Algorithm • Proposed Canny Implementation • Simulations – Results • Conclusions C.-L. Sotiropoulou – Real-Time Canny FPGA Implementation – AUTH-eLab

  6. Canny Algorithm SmoothingFilter Calculation of gradient Localization Elimination of spurious responses C.-L. Sotiropoulou – Real-Time Canny FPGA Implementation – AUTH-eLab

  7. Outline • Motivation • Canny Algorithm • Proposed Canny Implementation • Simulations – Results • Conclusions C.-L. Sotiropoulou – Real-Time Canny FPGA Implementation – AUTH-eLab

  8. Implementation Approach • Introduction of a 4-pixel parallel computation design • Pipelined architecture using on-chip BRAM memories for caching • Very efficient design with the same memory requirements as with a design without parallelism • In addition, complex arithmetical operations were substituted with shifts and additions/subtractions C.-L. Sotiropoulou – Real-Time Canny FPGA Implementation – AUTH-eLab

  9. Pipeline Implementation • Exploitation of minimum buffering before starting the computation C.-L. Sotiropoulou – Real-Time Canny FPGA Implementation – AUTH-eLab

  10. Gaussian Smoothing • 5 x 5 convolution • Introducing 4-pixel parallelism • Substitution of the multiplications and divisions with shifts additions and subtractions 25 pixels 40 pixels C.-L. Sotiropoulou – Real-Time Canny FPGA Implementation – AUTH-eLab

  11. Gaussian Smoothing C.-L. Sotiropoulou – Real-Time Canny FPGA Implementation – AUTH-eLab

  12. Gaussian Smoothing • Exploitation of on-chip BRAMS • Use of one BRAM to accommodate one image line aligned in 4-pixel words • Each BRAM has a size of image width / 4 x 32 bits (grayscale) C.-L. Sotiropoulou – Real-Time Canny FPGA Implementation – AUTH-eLab

  13. Gaussian Smoothing • Start of calculations  As soon as the first 2 lines of the cache are filled • All non-existing lines and columns of pixels necessary for the calculations are considered to be black C.-L. Sotiropoulou – Real-Time Canny FPGA Implementation – AUTH-eLab

  14. Sobel Gradient Calculation • 3 x 3 convolution • Introducing 4-pixel parallelism • Substitution of the multiplications and divisions with shifts additions and subtractions • Use of fixed point arithmetic 9 pixels 18 pixels C.-L. Sotiropoulou – Real-Time Canny FPGA Implementation – AUTH-eLab

  15. Sobel Gradient Calculation • Start of calculations  As soon as the first line of the cache are filled • All non-existing lines and columns of pixels necessary for the calculations are considered to be black C.-L. Sotiropoulou – Real-Time Canny FPGA Implementation – AUTH-eLab

  16. Non Maximum Suppression • Same principles as in Sobel Gradient Calculation • Requires the 3 x 3 neighboring pixels 9 pixels 18 pixels C.-L. Sotiropoulou – Real-Time Canny FPGA Implementation – AUTH-eLab

  17. Double Thresholding and Hysterisis • Double Thresholding is a double comparator • No caching required for this stage • Hysterisis normally requires the 3 x 3 neighboring pixels • Reduced to 4 neighboring pixels • Second pass in the opposite direction C.-L. Sotiropoulou – Real-Time Canny FPGA Implementation – AUTH-eLab

  18. Hysterisis Block C.-L. Sotiropoulou – Real-Time Canny FPGA Implementation – AUTH-eLab

  19. Outline • Motivation • Canny Algorithm • Proposed Canny Implementation • Simulations – Results • Conclusions C.-L. Sotiropoulou – Real-Time Canny FPGA Implementation – AUTH-eLab

  20. Simulations - Results • Synthesis results for 3 different FPGAs C.-L. Sotiropoulou – Real-Time Canny FPGA Implementation – AUTH-eLab

  21. Simulations - Results • Timing results for Spartan-6 C.-L. Sotiropoulou – Real-Time Canny FPGA Implementation – AUTH-eLab

  22. Outline • Motivation • Canny Algorithm • Proposed Canny Implementation • Simulations – Results • Conclusions C.-L. Sotiropoulou – Real-Time Canny FPGA Implementation – AUTH-eLab

  23. Conclusions • Real-time canny implementation • Parallel architecture with 4-pixel calculation • Increased throughput without increased need for memory resources • 240 frames per second achieved for 1Mpixel images on a Spartan-3E with 28% of the area • 580 frames per second on a Virtex-5 • 396 frames per second on a Spartan-6 C.-L. Sotiropoulou – Real-Time Canny FPGA Implementation – AUTH-eLab

  24. Acknowledgement The research activities that led to these results, were co-financed by Hellenic Funds and by the European Regional Development Fund (ERDF) under the Hellenic National Strategic Reference Framework (ESPA) 2007-2013, according to Contract no. MICRO2-49-project LoC. Thank you very much for your attention! C.-L. Sotiropoulou – Real-Time Canny FPGA Implementation – AUTH-eLab

More Related