https://github.com/garyli1019/udacity-carnd-p1

Udacity Self Driving Car Nano Degree Project 1 Line finder
https://github.com/garyli1019/udacity-carnd-p1

Last synced: 26 days ago
JSON representation

Udacity Self Driving Car Nano Degree Project 1 Line finder

• Host: GitHub
• URL: https://github.com/garyli1019/udacity-carnd-p1
• Owner: garyli1019
• License: mit
• Created: 2018-03-12T00:08:12.000Z (almost 7 years ago)
• Default Branch: master
• Last Pushed: 2018-03-12T00:28:20.000Z (almost 7 years ago)
• Last Synced: 2024-01-21T10:22:38.578Z (11 months ago)
• Language: Jupyter Notebook
• Size: 50.3 MB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# **Finding Lane Lines on the Road** 

[//]: # (Image References)

[image1]: ./test_images_output/draw_line_0.png "Grayscale"

---

### Reflection

### 1. Description of pipeline.

My pipeline consisted of 5 steps. 

1. Blur the gray scale image

2. Apply canny edge

3. Apply mask region

4. Convert to binary image

5. Hough transform

6. Combine the image

![alt text][image1]

My draw_lines() function:

First I used the slope of the lines to separate left line and right line.

Then I selected the longest line in each side.

After that, I calculated the b value of the normal line function(y = kx + b) and used the function to locate the point that the line will intersect with the bottom of the image. By doing this, my lines will always start from the bottom of the image, even if the line in the image are dotted.

Finally, draw the lines on the image.

### 2. Potential shortcomings

If the line was curved, it might not work well.

If the line was not clear, or the weather was bad, the line might not be located.

### 3. Suggest possible improvements 

Find a better way to improve draw_lines() function

Find out how to dealing with curved line.