Final Presentation

Report
Shane Tuohy

In 2008, rear end collisions accounted for
almost 25% of all injuries sustained in road
traffic accidents on Irish roads [RSA Road Collision
Factbook 2008]

Effective distance determination can go a
long way to reducing injuries

Mercedes Pre-Safe
Audi Pre-Sense Plus
Toyota Pre-Collision System

All are RADAR systems


 Expensive
 Cannot detect humans, animals
 Susceptible to interference

Front facing standard optical camera
 Cheap
 Many uses
 Simple to install

Begun by Intel, currently maintained by
community, under stewardship of Willow
Garage

Extensive library of Computer Vision
functions

C, C++, Python, Java

No need to continually ‘reinvent the wheel’
Capture
Image
Process
(OpenCV)
Feedback
To User
Capture
Image
Process
(OpenCV)
Feedback
To User
Threshold
Image
Warp
Perspective
Determine
Distance

Remove road surface and highlight objects
 Sample road surface in front of vehicle
 Remove pixels ±35 of sampled value
 Apply binary threshold
Threshold
Image
Warp
Perspective
Determine
Distance

Distance in image does not change linearly as
vehicle changes position

Inverse Perspective Mapping

Geometric transform which allows us to
remove perspective effect
Threshold
Image
Warp
Perspective
Determine
Distance

All road pixels are zero

Analyze area in front of car

Find first non zero pixels

Translate to distance using scaling factor

How can we know this ‘scaling factor’?

Need to calibrate for particular camera setup

Can be done once for given environment and
parameters
 Lay 1m object on road surface
 Use chessboard pattern of known size

Roughly calculated for project testing
Capture
Image
Process
(OpenCV)
Feedback
To User

Provide graphical feedback to user
1.
Threshold to remove road surface.
Generate transformation matrix
2.
Transform image to IPM view
3.
Distance determination
4.
Graphics overlay
5.
Modify algorithm for use on a real time video
stream

Further work possible
 Improve thresholding for different road conditions
 Improve performance of IPM algorithm
 Automatic calibration implementation

Paper submitted to ISSC 2010, awaiting review
 S. Tuohy, D. O Cualain, M. Glavin, E. Jones:“Distance
Determination for an Automobile Environment using
Inverse Perspective Mapping in OpenCV”

Successful implementation of proposed
algorithm
Demonstration

similar documents