Skip to content

Latest commit

 

History

History
62 lines (37 loc) · 2.6 KB

README.md

File metadata and controls

62 lines (37 loc) · 2.6 KB
Raw

Finding Lane Lines on the Road

In this project, I wrote the code to identify lane lines on the road, first in an image, and later in a video stream (really just a series of images). To complete this project I used the tools that I learned in the lesson.

The first goal was to write code including a series of steps (pipeline) that identify and draw the lane lines on a few test images. Once I successfully identifed the lines in an image, I used the code provided to run on a video stream.

Then I refined the pipeline with parameter tuning and by averaging and extrapolating the lines.

Finally, I made a brief writeup report.

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

Reflection

1. Describe your pipeline. As part of the description, explain how you modified the draw_lines() function.

My pipeline consisted of 5 steps.:

  • I converted the images to grayscale.
  • Then, manually using cv2.GaussianBlur, I removed the noise using a kernel_size of (5,5)
  • Then I applied the cv2.Canny filter in order to get the edges between a certain threshold, in this case (50,150)
  • Then I select the region of interest creating a mask and using cv2.fillPolly.
  • Then I select some lines based on the arguments passed to cv2.HoughLinesP
  • Then I drew the lines of lines found with cv2.HoughLinesP
  • Finally I merged the red lines drew with the rest of the image black (0,0,0) with the original image

In order to draw a single line on the left and right lanes, I modified the draw_lines() function by:

  • First, I found the slope in each set of lines.
  • Then, I split the lines of each segment of the road (left lines, right lines)
  • Then, I calculated x coordinates using y = mx + b and m = (y2-y1)/ (x2-x1)
  • Finally, I drew the lines.

alt text

alt text

2. Identify potential shortcomings with your current pipeline

One potential shortcoming would be what would happen when the region detects as in the challenge video the divider is too near to the lane lines and there is a shadow at the other side, so in both cases it caused a lot of interference.

3. Suggest possible improvements to your pipeline

A possible improvement would be to using a different approach like find a way to deal with very curved scenarios, also trying to detect a larger portion of the landmark without interference with cars or other objects.