学习笔记 | CarND Project - Advanced Lane Finding

(Udacity Learning Note: CarND-Term1 Project 4)

Camera Calibration & Distortion Correction

Two Types of Distortion

• Radial Distortion
• Tangential Distortion

Using OpenCV

• cv2.findChessboardCorners()
• cv2.drawChessboardCorners()
• cv2.calibrateCamera()
• cv2.undistort()

Example

• Finding chessboard corners (for an 8x6 board):

ret, corners = cv2.findChessboardCorners(gray, (8,6), None)

• Drawing detected corners on an image:

img = cv2.drawChessboardCorners(img, (8,6), corners, ret)

• Camera calibration, given object points, image points, and the shape of the grayscale image:

ret, mtx, dist, rvecs, tvecs = cv2.calibrateCamera(objpoints, imgpoints, gray.shape[:-1], None, None)

• Undistorting a test image:

dst = cv2.undistort(img, mtx, dist, None, mtx)

Perspective Transform

Commonly Used View

• Front-facing View (Normal View)
• Top-down View (Bird-eye View)

Why?

• Because ultimately we want to measure the curvature of the lines, and to do that, we need to transform to a top-down view.

Apply Perpective Transform using OpenCV

• Compute the perspective transform, M, given source and destination points:

　　M = cv2.getPerspectiveTransform(src, dst)

• Compute the inverse perspective transform:

　　Minv = cv2.getPerspectiveTransform(dst, src)

• Warp an image using the perspective transform, M: (warp v. 扭曲)

　　warped = cv2.warpPerspective(img, M, img_size, flags=cv2.INTER_LINEAR)

Gradient Threshold

Sobel Operator

• The Sobel operator is at the heart of the Canny edge detection algorithm.

Example use of Sober operator

• Calculate the derivative in the x direction (the 1, 0 at the end denotes x direction):

  sobelx = cv2.Sobel(gray, cv2.CV_64F, 1, 0)

•  Calculate the derivative in the y direction (the 0, 1 at the end denotes y direction):

  sobely = cv2.Sobel(gray, cv2.CV_64F, 0, 1)

•  Calculate the absolute value of the x derivative:

  abs_sobelx = np.absolute(sobelx)

•  Convert the absolute value image to 8-bit:

  scaled_sobel = np.uint8(255*abs_sobelx/np.max(abs_sobelx))

Direction of Gradient

• The direction of the gradient is simply the inverse tangent (arctangent) of the y gradient divided by the x gradient: 
  • arctan(sobel​y​​/sobel​x​​).

Color Space

• RGB
• HSV
• HLS

Reference

1. Wikipedia: HSL and HSV