Train your own classifier using this guide: http://docs.opencv.org/2.4/doc/user_guide/ug_traincascade.html
This classifier is used to detect cars in images and videos.
This is first Iteration of Classifier
Ran the classifier with LBP flag on opencv_traincascade
Works well but there are still inaccuracies in detection
Will need to train again with better images and for longer to get better results
Enter following command in terminal to test classifier
python opencv.py --image test_image.jpg
RESULTS SO FAR:
http://imgur.com/yxDj4Tn --- Worked well in this instance
http://imgur.com/JZPaZOP --- Some errors with the edges of detection. Will be trying to solve such errors with tuning training parameters
http://imgur.com/IRooDC1 --- Seems that in this case its quite off from the desired area. Might be because of pick up truck or the angle
http://imgur.com/PWjvfCU --- A little outside the desired range
http://imgur.com/uOegeGb --- A little outside the range as well
Demo I created to show the stage I plan to reach: