Detect Cars

This project is a master's thesis that consisted of an investigation in the detection of cars and highway lines using Python and the OpenCV API. The code is free to be used and modified by anybody who wishes to do so.

##UTPL ###Professor:

• Rodrigo Barba lrbarba@utpl.edu.ec

###Students:

• Marcelo Bravo mdbravo4@utpl.edu.ec
• Galo Celly gscelly@utpl.edu.ec
• Nicholas Earley nearley@utpl.edu.ec

System Requirements

• An i3 or better processor. The faster the better, especially at high video resolutions.
• 2 GB or more RAM.
• At least 100 MB Free Disk space
• Windows 7 or later, OS X 10.8 or later (has only been tested on 10.9), Linux 3.0+

Installation

1. First, one should install the following libraries:

• OpenCV version 2.4.10+
• Python 2.7.9 (or any later Python 2.x) (See Installation on OS X if using a Mac)
• Numpy 1.9.2+
• Scipy 0.15.1+
• Now download and extract this repository with one of several options:
  • Clone the repository with $ git clone https://github.com/VAUTPL/Deteccion.git
  • Download the repository as a .zip or .tar.gz and then extract it.

Installation on OS X

Apple uses a prior version of Python that does not support the latest Python libraries. One work around is to install Python with Homebrew:

$ brew install python

Replacing Apple's system Python with an unsupported version may break things. Therefore we linked Homebrew's Python into the system path without replacing the system Python:

$ ln -s /usr/local/Cellar/python/2.x.y/bin/python /bin/hbpython

Where 2.x.y is the version number of your Python.

Running

From a command line in the folder of the repository:

$ python detect.py [-v path/to/video] [-c path/to/cascade.xml]

Videos should be *.mov on Mac OSX and *.avi on Windows.

###On UNIX... You may add a shebang (#!) line to the top of detect.py with the path to the appropriate Python. Example:

#! /bin/python (#! /bin/hbpython in the case of Mac) at the top of detect.py

####Make the script executable $ chmod +x detect.py

####To Run: $ ./detect.py [-v path/to/video] [-c path/to/cascade.xml]

If no video is specified, OpenCV attempts to open the Webcam, see line number 73: camera = cv2.VideoCapture(1) in detect.py

###Examples: Windows: C:\path\to\repository> python detect.py -v data/Loja.avi

UNIX and UNIX-Like: $ python detect.py -v data/Loja.mov or $ ./detect.py -v data/Loja.mov

• OR: $ python detect.py -v data/Loja.avi or $ ./detect.py -v data/Loja.avi

Mac OSX: $ hbpython detect.py -v data/Loja.mov or $ ./detect.py -v data/Loja.mov

###Other Arguments Default values can be found in config.py, but can be temporarily overwritten with the following arguments:

• -c or --car Path to Car Haar Cascade
• -v or --video Path to the (optional) video file -- default is Live Webcam Capture
• -S or --scale-factor Haar Cascade Scale Factor
• -N or --min-neighbors Haar Cascade Minimum Neighbors
• DEPRECIATED: -X or --roi-x Region Of Interest top left corner X coordinate
• DEPRECIATED: -Y or --roi-y Region Of Interest top left corner Y coordinate
• DEPRECIATED: -W or --roi-width Region Of Interest Width
• DEPRECIATED: -H or --roi-height Region Of Interest Height
• -x or --max-line-gap Max Line Gap
• -n or --min-line-length Min Line Length
• -1 or --threshold-1 Canny 1st Threshold
• -2 or --threshold-2 Canny 2nd Threshold
• -a or --aperture-size Canny Aperture Size
• -r or --rho Hough Rho
• -t or --theta Hough Theta in Radians
• -T or --threshold Hough Threshold
• -D or --delay Delay for operating system to sleep between frames, default is no delay (process as fast as possible)
• -h or --help Show help along with default values used for unspecified arguments

Depreciated Arguments

Region of Interest is calculated based on the size of the video (full width, bottom half of the height). Therefore arguments which define the ROI are depreciated.