thoomi / vehicle-detection

Vehicle detection pipeline utilizing standard computer vision techniques for feature extraction and a trained linear classifier.

Vehicle Detection

Overview

This is the fifth project I am doing as part of Udacity's Self-Driving-Car Nanodegree.

Project Goals

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.
• Implement a sliding-window technique and use your trained classifier to search for vehicles in images.
• Run your pipeline on a video stream 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.

Results

For a more detailed insight on the project please see the full Writeup / Report.

Training images

The images used to train the linear classifier are a mix from various datasets which are listed below:

• Vehicles
• Non-Vehicles
• GTI vehicle image database
• KITTI vision benchmark suite