Stereoscopic%20Imaging%20for%20Slow-Moving%20Autonomous%20Vehicle%20Senior%20Project%20Proposal%20Bradley%20University%20ECE%20Department

1
Stereoscopic Imaging for Slow-Moving Autonomous
VehicleSenior Project ProposalBradley
University ECE Department

• By Alex Norton
• Advisor Dr. Huggins
• November 15, 2011

2
Presentation Outline

• Introduction to stereoscopic imaging
• Project goals
• Previous work
• Project description
• Preliminary lab work
• Equipment list
• Schedule of tasks for spring

3
What is Stereoscopic Imaging?

• The use of two horizontally aligned, slightly
  offset cameras taking a pair of images at the
  same time
• By matching corresponding pixels between the two
  images, the distances to objects can be
  calculated using triangulation
• This depth information can be used to create a
  3-D image and terrain map

4
Project Goals

• Learn theory of 3D stereoscopic imaging
• Investigate existing software (OpenCV and MATLAB)
• Control cameras
• Calibrate cameras
• Take and store images
• Process images for objects
• Correlate objects
• Compute distance to objects
• Compute terrain map

5
Previous Work

• BirdTrak (Brian Crombie and Matt Zivney, 2003)
• Bradley Rover(Steve Goggins, Rob Scherbinski,
  Pete Lange, 2005)
• NavBot (Adam Beach, Nick Wlaznik, 2007)
• SVAN (John Hessling, 2010)

6
Project Description

• System block diagram
• Subsystem block diagrams
• Cameras
• Laptop
• Software
• Mode of operation
• Calibration mode
• Run mode

7
System Block Diagram
8
Cameras Subsystem
9
Laptop Subsystem
10
Software
11
Calibration Mode

• Initial mode of operation
• Ensures the accuracy of the terrain map generated
  in run mode by correcting for lens distortion
• Cameras will take images of a chessboard in
  multiple orientations
• Camera intrinsic and distortion parameters can be
  determined, which are used to correct for
  distortion in images from un-calibrated cameras

12
Run Mode

• Primary mode of operation entered once cameras
  are calibrated
• Cameras capture a set of images after receiving
  signals from the laptop
• A disparity map is created using the two images
  and distances to objects are calculated
• This information is used to generate a terrain
  map which is stored in a text file to be used to
  navigate an autonomous vehicle

13
Preliminary Lab Work

• Current test camera setup

14
Preliminary Lab Work
Left Camera image Right Camera Image
15
Preliminary Lab Work
Edge Detection of Left Image
Edge Detection of Right Image
16
Preliminary Lab Work
Represents the differences in corresponding
pixels between the left and right cameras
Disparity Map Formed Using Left and Right Images
17
Equipment List

• Two Logitech Quickcam Express webcams
• Compaq Presario CQ60 laptop
• Mathworks Matlab
• Microsoft Visual Studio 2008
• OpenCV
• Equipment to be ordered Two webcams compatible
  with Windows 7 and Linux

18
Preliminary Lab Work
MATLAB code that sets up the webcams to receive
image data from them
19
Preliminary Lab Work
MATLAB code that gets an image from each camera,
filters them using the median filter function,
uses the canny edge detection function, and
displays the filtered images and edge detected
images
20
Schedule of Spring Tasks
21
Questions?