Midterm Presentation Vitaliy Orekhov Imaging Robotics Intelligent Systems
Midterm Presentation Vitaliy Orekhov Imaging, Robotics, & Intelligent Systems Laboratory The University of Tennessee March 07, 2006 Slide
Outline • Task 1 – Math 472 – Modern Transforms • Task 2 – ECE 571 – Pattern Classifications • Task 3 – ECE 574 –Test, Evaluate, and Transfer Camera Calibration Code to C++ • Task 4 – ECE 671 –Survey on Wide Angle Camera Calibration and Image Correction 11/30/2020 Slide 2/15
Correcting Distortion on an image sequence in real-time IQeye 3 Network Camera • 1. 8 M pixels/sec JPEG • 1/2" 1288 x 968 progressive scan CMOS digital imager • 50 fps (frames per second) at 160 by 120 • 19 fps at 320 by 240 • 5. 4 fps at 640 by 480 • 3. 6 fps at 800 by 600 • 1. 3 fps at 1288 by 968 www. oksolar. com YV 2. 2 x 1. 4 A-SA 2 Lens • Fish-eye vari-focal lens with the horizontal field angle of 185 -94 degrees (when used on 1/3 cameras) www. fujinon. de 11/30/2020 Slide 3/15
Image Sequence Correction Diagram 11/30/2020 Slide 4/15
Camera Calibration • Capturing sequence of images with calibration pattern • Extracting coordinates using Harris corner detections • Use MATLAB calibration code written by Chris Broaddus – Extract camera matrix – Distortion parameters • Radial and tangential distortion – Single polynomial approximation used to model the lens projections. 11/30/2020 Slide 5/15
Image Correction • Code written in C++ • Creates a look up table based the camera matrix and distortion coefficients. • Code will work with most commonly used distortion models: • Perspective • Stereographic • Equidistant • Orthogonal • Logarithmic • Radial polynomial • Look up table is used to correct each image captured from the camera in real-time 11/30/2020 Slide 6/15 Image size: 620 x 620 185 deg. fov Image size: 620 x 620 80 deg. fov
Wide Angle Camera Calibration • • Image size 312 x 312 Frame rate from camera: 14 fps. Frame rate with distortion correction: 6 fps. Smallest angle-of-view setting (about 80 deg. ) 11/30/2020 Slide 7/15
Wide Angle Camera Calibration • • Image size 312 x 312 Frame rate from camera: 14 fps. Frame rate with distortion correction: 6 fps. Widest angle-of-view setting (185 deg. ) 11/30/2020 Slide 8/15
Wide Angle Camera Calibration • • • Original Image size 312 x 312 Restored Image size 428 x 428 Frame rate: 14 fps. Distortion corrected not in real time Smallest angle-of-view setting (about 80 deg. ) 11/30/2020 Slide 9/15
Processing Time Comparison Implementation Clock speed Time (ms) 312 x 312 pixels Time (ms) 620 x 620 pixels Pentium 4 with lookup table 1. 4 GHz 150 460 Clock speed Time (ms) 600 x 600 pixels Pentium 4 with lookup table 1. 7 GHz 760 Pentium 4 with real time calculation 1. 7 GHz 780 Various FPGA implementations 10 MHz – 34 MHz 360 -11 Implementation 1* 1* D. Eadie, F. Shevlin, A, Nisbet “Correction of Geometric Image Distortion Using FPGA’s” SPIE Proceeding Opto-Ireland, Vol. 4877, pp. 28 -37, March 2003. 11/30/2020 Slide 10/15
Task 4 Survey on Wide Angle Camera Calibration and Image Correction Fall 2005 Calibrating distortion in camera’s with wide field of view. Spring 2006 Complete calibration methods. – Test-range calibration – Non-metric calibration – Self calibration 11/30/2020 Slide 11/15
Test-range Calibration • Methods which use calibration objects whose 2 D or 3 D coordinates are accurately known. Z. Zhang, “A flexible new Technique for camera calibration”, Technical Report MSR-TR-98 -71, Microsoft Research, Dec. 1998. Z. Zhang, “A flexible new Technique for camera calibration”, IEEE Tans. Pattern Analysis and Machine Intelligence, Vol. 22, No. 11, 1330 C 1334, Nov. 2000. • Planar calibration pattern from at least two different views. • Two major steps • Closed form solution • Nonlinear optimization C. Broaddus, “Universal Geometrci Camera Calibration with Statistical Model Selection, ” Masters Thesis, Department of Electrical Engineering, The Unverstity of Tennessee, Knoxville, TN 2005. 11/30/2020 Slide 12/15
Non-metric and Self Calibration Non-metric calibration – These methods use invariants of image features. Self calibration – – Methods which use camera intrinsic constraints, camera motion constraints or scene constraints. Self calibration methods which use pure rotation have been shown to be very sensitive to radial distortion. B. Tordoff and D. Murray, “The impact of radial distortion on the selfcalibration of rotating cameras, ” CVIU, Vol. 96, No. 1, Oct. 2004, pp. 17 -34. 11/30/2020 Slide 13/15
Summary Task 3: Test, Evaluate, and Transfer Camera Calibration Code to C++ Accomplished: • Calibrated IQEye 3 camera for different field of views and image sizes. • Generated real time correction of image sequence. • Started working on GUI to control and display the real time correction of image sequence. Future Objectives: • Finish working on the GUI. • Work on improving frame rate of the undistorted image sequence. • Generate results for IPIX camera. • Work on optimization step in camera calibration for C++ code. Task 4: Survey on Wide Angle Camera Calibration and Image Correction Accomplished: • Have been exploring on methods with complete calibration implementations. 11/30/2020 Slide 14/15 Future Objectives: • Continue working on survey paper. • Reexamine and revise content previously written.
Thank you Suggestions/Comments/Questions 11/30/2020 Slide 15/15
- Slides: 15