You can use this file as a template for your writeup if you want to submit it as a markdown file, but feel free to use some other method and submit a pdf if you prefer.
Vehicle Detection 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.
Rubric Points
Here I will consider the rubric points individually and describe how I addressed each point in my implementation.
1. Provide a Writeup / README that includes all the rubric points and how you addressed each one. You can submit your writeup as markdown or pdf. Here is a template writeup for this project you can use as a guide and a starting point.
You're reading it!
1. Explain how (and identify where in your code) you extracted HOG features from the training images.
The code for this step is contained in cells 1-5 of the Python notebook.
I started by reading in all the vehicle and non-vehicle images. Here is an example of one of each of the vehicle and non-vehicle classes:
I then explored different color spaces and different skimage.hog() parameters (orientations, pixels_per_cell, and cells_per_block). I grabbed random images from each of the two classes and displayed them to get a feel for what the skimage.hog() output looks like.
Here is an example using the YUV color space and HOG parameters of orientations=8, pixels_per_cell=(8, 8) and cells_per_block=(2, 2):
I tried various combinations of parameters and eventually settled on this configuration:
color_space = 'YUV' # Can be RGB, HSV, LUV, HLS, YUV, YCrCb
orient = 9 # HOG orientations
pix_per_cell = 8 # HOG pixels per cell
cell_per_block = 2 # HOG cells per block
hog_channel = 1 # Can be 0, 1, 2, or "ALL"
spatial_size = (16, 16) # Spatial binning dimensions
hist_bins = 16 # Number of histogram bins
spatial_feat = True # Spatial features on or off
hist_feat = True # Histogram features on or off
hog_feat = True # HOG features on or off
3. Describe how (and identify where in your code) you trained a classifier using your selected HOG features (and color features if you used them).
I trained a linear SVM using LinearSVC() and you can see the code in Cell 9
I spent some time tweaking parameters, eventually setling on the YUV colorspace and the U channel (channel 1)
1. Describe how (and identify where in your code) you implemented a sliding window search. How did you decide what scales to search and how much to overlap windows?
This where I spent most of the time. I tried to balance out the accuracy with the speed it took to create each video.
I tried window sizes of 24, 48, 96, 128 and 256.
I tried all possible combinations of the overallping, until I realized that if I use only X overalaping, it gives me very good accuracy and speeds up the process a lot.
Eventually I settled on 96 and 128 which I found to be a good compromise i finding the cars close and far
2. Show some examples of test images to demonstrate how your pipeline is working. What did you do to optimize the performance of your classifier?
Ultimately I searched on two scales using YCrCb 3-channel HOG features plus spatially binned color and histograms of color in the feature vector, which provided a nice result. Here are some example images:
1. Provide a link to your final video output. Your pipeline should perform reasonably well on the entire project video (somewhat wobbly or unstable bounding boxes are ok as long as you are identifying the vehicles most of the time with minimal false positives.)
Here's a link to my video result
2. Describe how (and identify where in your code) you implemented some kind of filter for false positives and some method for combining overlapping bounding boxes.
Described overalpping above.
Here is the output of scipy.ndimage.measurements.label() on the integrated heatmap from all six frames:
1. Briefly discuss any problems / issues you faced in your implementation of this project. Where will your pipeline likely fail? What could you do to make it more robust?
By far the biggest challenge was to eliminate the false positives. My pipeline worked pretty good from the beginning when it comes to detecting vehicles, but was also picking up a lot of junk.
Since the turn-over times were high (between 20 and 40 min to create a video), it took a lot of time for trial and error until finding a good balance of accuracy and performance.