def find_corners(self, img):
"""
Finds the chessboard corners of a image
"""
gray = utils.grayscale(img)
ret, corners = cv2.findChessboardCorners(gray, (self.nx, self.ny),
None)
return ret, corners, gray
def gray_threshold(self, img, thresh=(130, 255)):
"""
Grayscales and threshold an image.
Attributes:
img: the image.
thresh: threshold lower and upper bounds tuple.
"""
gray = utils.grayscale(img)
return self.threshold(gray, thresh)
def unwarp(self, img, grayscale=False):
"""
Unwarps an image
"""
if grayscale:
gray = utils.grayscale(img)
else:
gray = img
img_size = (gray.shape[1], gray.shape[0])
M, Minv = self.get_matrices()
return cv2.warpPerspective(img, Minv, img_size, flags=cv2.INTER_LINEAR)
def direction(self, img, kernel=3, thresh=(0.7, 1.3)):
"""
Applies sobel x and y, computes the direction of the
gradient and then applies a threshold.
Attributes:
img: the image.
kernel: The sobel kernel.
thresh: threshold lower and upper bounds tuple.
"""
gray = utils.grayscale(img)
# Take the gradient in x and y separately
sobelx = cv2.Sobel(gray, cv2.CV_64F, 1, 0, ksize=kernel)
sobely = cv2.Sobel(gray, cv2.CV_64F, 0, 1, ksize=kernel)
# Take the absolute value of the x and y gradients
absgraddir = np.arctan2(np.absolute(sobely), np.absolute(sobelx))
return self.threshold(absgraddir, thresh)
def sobel(self, img, orient='x', kernel=3, thresh=(20, 100)):
"""
Applies sobel filter in the *orient* direction.
Attributes:
img: the image.
orient: orientation for to apply the filter.
thresh: threshold lower and upper bounds tuple.
"""
gray = utils.grayscale(img)
if orient == 'x':
sobel = cv2.Sobel(gray, cv2.CV_64F, 1, 0, ksize=kernel)
else:
sobel = cv2.Sobel(gray, cv2.CV_64F, 0, 1, ksize=kernel)
# Take the absolute value of the derivative or gradient
abs_sobel = np.absolute(sobel)
# Scale to 8-bit (0 - 255) then convert to type = np.uint8
scaled_sobel = np.uint8(255 * abs_sobel / np.max(abs_sobel))
return self.threshold(scaled_sobel, thresh)
def grad_magnitude(self, img, kernel=3, thresh=(30, 100)):
"""
Applies sobel on both orientations (x and y) and then
computes the gradient magnitude and applies a threshold.
Attributes:
img: the image.
kernel: The sobel kernel.
thresh: threshold lower and upper bounds tuple.
"""
gray = utils.grayscale(img)
# Take the gradient in x and y separately
sobelx = cv2.Sobel(gray, cv2.CV_64F, 1, 0, ksize=kernel)
sobely = cv2.Sobel(gray, cv2.CV_64F, 0, 1, ksize=kernel)
# Calculate the magnitude
gradmag = np.sqrt(sobelx**2 + sobely**2)
# Scale to 8-bit (0 - 255) and convert to type = np.uint8
scale_factor = np.max(gradmag) / 255
gradmag = (gradmag / scale_factor).astype(np.uint8)
return self.threshold(gradmag, thresh)