#Finding Lane Lines on the Road

---

Goal & Steps

The goals / steps of this project are the following:

• Make a pipeline that finds lane lines on the road
• Test the pipeline on test images
• Test the pipeline on test videos
• Test the pipeline on challenge video and make adjustments

---

Reflection

Pipeline description

• Convert the image to grayscale
• Apply Gaussian blur
• Apply Canny edge detection
• Mask out edges image with region of interests
• Apply Hough algorithm on the masked edge detected image to obtain line segments.
• Draw the line segments on the original image

How I modify draw_lines()

• Divide line segments into two groups: left side and right side.
• For each side, find out the maximum/minimum x and y value among points of all line segments.
• create a new line segment with the corner points found out in previous step.
• append the new line segment to the original line segment array

Source

• ipython notebook

Results of test images

• solidWhiteCurve '
• solidWhiteRight '
• solidYellowCurve '
• solidYellowCurve2 '
• solidYellowLeft '
• whiteCarLaneSwitch '

results of test videos

• white.mp4(click View Raw to download)
• yellow.mp4(click View Raw to download)
• extra.mp4(click View Raw to download)

###2. Identify potential shortcomings with your current pipeline

• It can't identify thin lane lines on the far end.
• Corner point calculation is buggy. It draws random straight line sometimes.
• It completely fails on the challenge video.

###3. Suggest possible improvements to your pipeline

• Make use of opencv's built-in box-bound function.
• Add some color filters to make it able to handle special road color.