wolfgang-stefani/advanced_lanelines_detection

In this project, a software pipeline was written to identify the lane boundaries in a video from a front-facing camera on a car using computer vision techniques with openCV.

Advanced Lane Finding

In this project, a software pipeline was written to identify the lane boundaries in a video from a front-facing camera on a car. The full code can be found in this IPython Notebook.

Field test on german highway A99 in Munich

I wanted to know if this algorithm runs on real videos taken by myself. So, what did I do? I took my wife out on the next highway from our apartment. She had to drive the car, so I was able to adjust the camera and record a video. Then I tested the algorithm on this video. Here ist the result:

Additionally, in this longer clip the radius of the curvature and the offset of the car with respect to the center are provided:

The steps of this project are the following:

• Compute the camera calibration matrix and distortion coefficients given a set of chessboard images.
• 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 a polynomial 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.
• Output visual display of the lane boundaries and numerical estimation of lane curvature and vehicle position.

The images for camera calibration are stored in the folder called camera_cal. The images in test_images are for testing the pipeline on single frames. If you want to extract more test images from the videos, you can simply use an image writing method like cv2.imwrite(), i.e., you can read the video in frame by frame as usual, and for frames you want to save for later you can write to an image file.

Pipeline in a nutshell

(for a very detailed description including how camera calibration is implemented, see the file "writeup.pdf")

Step 1: Read in an image

Step 2: Distortion correction

Step 3: Thresholding

Various combinations of color and gradient thresholds were tested. (Note: This step is visualized with “test4.jpg” instead of “test6.jpg” as for all the other steps because here you can see the advantages of s-channel when road has bad sun/shadow conditions)

Steps in detail:

Step 4: Perspective Transform

First, identifying four source points src (pick four points in a trapezoidal shape (similar to region masking) and after four destination points dst.

Step 5: Grayscale

Step 6: Detect lane pixels (sliding windows method) and fit a polynomial to find the lane boundary

Explicit Decision which pixels are part of the lines and which belong to the left respectively to the right line.

Steps in detail:

Step 7: Drawing

This includes: a) Draw the lines on a blank (zeroed) image b) Warp this image back to original image space using inverse perspective Matrix (Minv) c) Combine the result with the original image