说明：  Canny 算法包含许多可以调整的参数，它们将影响到算法的计算的时间与实效。 高斯滤波器的大小:第一步所用的平滑滤波器将会直接影响 Canny 算法的结果。较小的滤波器产生的模糊效果也较少，这样就可以检测较小、变化明显的细线。较大的滤波器产生的模糊效果也较多，将较大的一块图像区域涂成一个特定点的颜色值。这样带来的结果就是对于检测较大、平滑的边缘更加有用，例如彩虹的边缘。 阈值:使用两个阈值比使用一个阈值更加灵活，但是它还是有阈值存在的共性问题。设置的阈值过高，可能会漏掉重要信息;阈值过低，将会把枝节信息看得很重要。很难给出一个适用于所有图像的通用阈值。目前还没有一个经过验证的实现方法。
(The Canny algorithm contains many parameters that can be adjusted, which will affect the time and effect of the algorithm's computation. The size of the gaussian filter: the smoothing filter used in the first step will directly affect the results of the Canny algorithm. Smaller filters also produce less blurring, which can detect smaller, more variable lines. The larger filter produces more fuzzy effects, and the larger image area is painted as a specific point color value. The result is more useful for detecting large, smooth edges, such as the edge of a rainbow. Threshold: using two thresholds is more flexible than using a threshold, but it still has a common problem with thresholds. The set threshold is too high and may miss important information; If the threshold is too low, it will be important to see the details. It is difficult to give a generic threshold for all images. There is currently no proven implementation method.)