Canny

Finds edges in an image using the [Canny86] algorithm.

cv2.Canny(image, threshold1, threshold2[, edges[, apertureSize[, L2gradient]]]) → edges

• image (np.ndarray): Single-channel 8-bit input image.
• threshold1 (float): First threshold for the hysteresis procedure. See notes.
• threshold2 (float): Second threshold for the hysteresis procedure. See notes.
• edges (optional; np.ndarray): Output edge map; single channels 8-bit image, which has the same size as image.
• apertureSize (optional; int): Aperture size for the Sobel() operator. Default is 3.
• L2gradient (optional; bool): A flag, indicating whether a more accurate L2 norm \(L_2 =\sqrt{(dI/dx)^2 + (dI/dy)^2}\) should be used to calculate the image gradient magnitude (L2gradient=True), or whether the default L1 norm \(L_1 = |dI/dx|+|dI/dy|\) is enough (L2gradient=False).

The function finds edges in the input image and marks them in the output map, edges, using the Canny algorithm. The smallest value between threshold1 and threshold2 is used for edge linking. The largest value is used to find initial segments of strong edges.

• The threshold parameters determine if a pixel is selected as an edge pixel based on the pixel's gradient. To do this the function examines the intensity gradient following a procedure analgous to Sobel().
  • If the gradient is larger than the higher threshold, it will be selected as an edge pixel.
  • If the gradient is smaller than the lower threshold, it will not be considered.
  • If the gradient is between the threshold values, then it will only be selected if it is connected an existing edge pixel.
  • Canny recommended a upper:lower ratio between 2:1 and 3:1.

• OpenCV Documentation
• OpenCV Tutorial: Canny Edge Detector
• [Canny86]: J. Canny. A Computational Approach to Edge Detection, IEEE Trans. on Pattern Analysis and Machine Intelligence, 8(6), pp. 679-698 (1986).
• Wikipedia: Canny Edge Detector