Vehicle Detection

Quick Instructions:

To train the vehicle and non-vehicle data folders must be located in folder:

C:\Udacity\SDCND\term1\resources\training-data\vehicle-detection_data

Run from windows CMD:

$ activate carnd-term1
$ vdt  (train and annotate video)
	<or>
$ vdt preload (video annotation only)

Run from windows bash shell:

$ source activate carnd-term1
$ vdt  (train and annotate video)
	<or>
$ vdt preload (video annotation only)

This project used a software pipeline to detect vehicles in a video (project_video.mp4).

See the detailed writeup of the project folder for more info.

The Project

The goals / steps of this project are the following:

• Perform a Histogram of Oriented Gradients (HOG) feature extraction on a labeled training set of images and train a classifier Linear SVM classifier
• Optionally, you can also apply a color transform and append binned color features, as well as histograms of color, to your HOG feature vector.
• Note: for those first two steps don't forget to normalize your features and randomize a selection for training and testing.
• Implement a sliding-window technique and use your trained classifier to search for vehicles in images.
• Run your pipeline on a video stream (start with the test_video.mp4 and later implement on full project_video.mp4) and create a heat map of recurring detections frame by frame to reject outliers and follow detected vehicles.
• Estimate a bounding box for vehicles detected.

Here are links to the labeled data for vehicle and non-vehicle examples to train your classifier. These example images come from a combination of the GTI vehicle image database, the KITTI vision benchmark suite, and examples extracted from the project video itself. You are welcome and encouraged to take advantage of the recently released Udacity labeled dataset to augment your training data.

Some example images for testing the pipeline on single frames are located in the test_images folder. See examples of the output from each stage of the pipeline in the folder called ouput_images.

As an optional challenge Add in your lane-finding algorithm from the last project to do simultaneous lane-finding and vehicle detection!

If you're feeling ambitious (also totally optional though), don't stop there! We encourage you to go out and take video of your own, and show us how you would implement this project on a new video!