julianshi/carnd-advanced-lane-lines

Advanced Lane Finding

This is the repositor of the codes to detect lanes on roads. The library in LanePipeline.py provids two methods in detecting lanes from images and videos respectively. Examples are as follows

from LanePipeline import LanePipeline
import cv2

lanePipeline = LanePipeline()

# Detect the lane on a 720 by 1280 by 3 image
image = cv2.imread('test_images/test1.jpg')
image_output = lanePipeline.lane_in_image(image) 

# Detect lanes from video 
lanePipeline.lane_in_video('project_video.mp4')

• Please refer to writeup.md for the checking of the project rubric points and solution images
• Please refer to output/project_video.mp4 for the result video.
• Please refer to camera_calibration.py for calibrating the camera. The data is saved in wide_dist_pickle.p for quick loading.

The LanePipeline.py library

The input image goes through the following pipeline:

• Apply a distortion correction to raw images.
• Use color transforms, gradients, etc., to create a thresholded binary image.
• Apply a perspective transform to rectify binary image ("birds-eye view").
• Detect lane pixels and fit to find the lane boundary.
• Determine the curvature of the lane and vehicle position with respect to center.
• Warp the detected lane boundaries back onto the original image.

---

Comments: this project is forked from a Udacity respository.