dr-lab / carnd-term1

Car Lane Detection, for learning only

Finding Lane Lines on the Road

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. Pipe line of how to detect edges and draw lines.

Filter colors, only keep White and Yellow, gray out and blur the image.

When we filter the color, we only keep Yellow and White, it is little aggressive, since some countries may use other colors for lane. But this method help lot o the line detection precision.

Filter White and Yellow Color

Gray Out Image

Blur the Image

Use Canny edge detection, detect the lines

Gt the lines array by Canny edge detection algorithms.

Scope to the region of interest, which includes the lane lines

Crop the image to region which we interested in for lane detection. How to crop the region is very empirical, the vertices we use are based on the images/video we have in this assignment. It may not fit all conditions.

![image5]:(./test_images_output/solidWhiteRight_region_of_interest.jpg "region of interest")

draw lines within the region of interest

We first draw the lines we detected in GREEN, then use linear regression, based on the slope, get two average slope lines from bottom to upper region, one on left, another on right.

mask the lines with the original color image. For better debug and understanding the algorithm

1) Use GREEN mark the lines, which may solid, may dotted
2) Use RED mark the linear regression line

2. Identify potential shortcomings with your current pipeline

1. So far the algorithm based on Color, to detect lines. This has limitations, and not fit all kind of road condition. Some of the lines maybe not White nor Yellow, which will be filtered out and cause issue.
2. We use several empirical settings, e.g. region of interest, which may not fit all conditions.
3. Line draw algorithms use linear line, which cannot solve curve road (lane lines) good

3. Suggest possible improvements to your pipeline

1. We need to use some other ways to remove the noise, then use Gray out algorithms may can help. The way to remove noise various, and not quit easy.
2. and tune the parameter to get the region of interest more accurate. This depends where the camera installed, and whether the camera is static (not tilt).
3. The line draw algorithm need to support curve line