Finding Lane Lines on the Road
The goals / steps of this project are the following:
- Make a pipeline that finds lane lines on the road
- Reflect on your work in a written report
1. Describe your pipeline. As part of the description, explain how you modified the draw_lines() function.
My pipeline consisted of 5 steps:
- Convert the images to grayscale
- Apply Canny to find all the edges
- Mask a region of interest around the lanes
- Apply Hough function to find all the lines in the ROI
- Add Hough lines to the image
In order to draw a single line on the left and right lanes, I modified the draw_lines() function by separating the left and right lines the Hough function returns by calculating their slopes using (y2-y1)/(x2-x1). Positive values were right lines and negative values were left. I also filtered the lines using a range of slopes to try and ignore any horizontal lines. Once I had the lines I want I calculated their average positions to get a single line for the left and right lane lines. These lines represent the average position and slope of all the left and right lines. To extend these lines I was able to take the two lines and calculate the y-intercept using b=y-mx since I had m,x and y. Once I had b I found the two endpoints of the solid lines I need by using x=(y-b)/m, plugging in 540 and 320 as the y positions.
One potential shortcoming would be what would happen when the lane lines are not clearly marked or do not exist? Finding the average single lines would crash the program with a divide by zero error.
Another shortcoming could be if the resolution of the image is different. I would get a bad looking region of interest and solid lines because I am hardcoding the number of pixels I want.
A possible improvement would be to calculate the vertices and limits of my lines in a more robust manner that is independent of resolution.
Another potential improvement could be to find a way to make the solid lines look less jittery in the video.
- Clone project
- If you have conda you can load my environment by running
conda env create -f environment.yml - I use the same environment for all my Udacity Self-Driving Car course python projects
- Run
python main.py - Check the test_images_output and test_videos_output directories