Modules used:

使用的模塊：

For this, we will use the opencv-python module which provides us various functions to work on images.

為此，我們將使用opencv-python模塊，該模塊為我們提供了處理圖像的各種功能。

Download opencv-python

下載opencv-python

General Way:
pip install opencv-python
Pycharm Users:
Go to the project Interpreter and install this module from there.

opencv-python Module:

opencv-python模塊：

opencv-python is a python library that will solve the Computer Vision Problems and provides us various functions to edit the Images.

opencv-python是一個python庫，它將解決計算機視覺問題并為我們提供編輯圖像的各種功能。

Note: The edge Detection is possible only in grayscale Image.

注意：只能在灰度圖像中進行邊緣檢測。

What we will do in this script?

我們將在此腳本中做什么？

To detect the edges of the images we will use opencv-python various Functions and Provide thresholds.

為了檢測圖像的邊緣，我們將使用opencv-python的各種功能并提供閾值。

In this article we will detect the edge of the Image with the help of various functions and the accuracy of edge increases as we go down,

在本文中，我們將借助各種功能來檢測圖像的邊緣，并且當我們下降時邊緣的精度會提高，

• Sobel Function: This Function will create the Horizontal and vertical edges and after that, we will use the Bitwise or operator to combine them

  Sobel函數 ：此函數將創建水平邊緣和垂直邊緣，然后，我們將使用按位或運算符將它們組合

• Laplacian Function: This Function is the simplest Function in which we just have to put the Grayscale Variable into it, and we will get the edge detected image.

  拉普拉斯函數 ：此函數是最簡單的函數，只需要將灰度變量放入其中，就可以得到邊緣檢測到的圖像。

• Canny Function: This is the most powerful function for edge detection and most accurate.

  Canny功能 ：這是邊緣檢測功能最強大且最準確的功能。

Let's see the code:

讓我們看一下代碼：

1)使用Sobel函數 (1) Using Sobel Function)

# importing the module
import cv2
# read the image and store the data in a variable
image=cv2.imread("/home/abhinav/PycharmProjects/untitled1/b.jpg")
# make it grayscale
Gray=cv2.cvtColor(image,cv2.COLOR_BGR2GRAY)
# Make it with the help of sobel
# make the sobel_horizontal
# For horizontal x axis=1 and yaxis=0
# for vertical x axis=0 and y axis=1
Horizontal=cv2.Sobel(Gray,0,1,0,cv2.CV_64F)
# the thresholds are like 
# (variable,0,<x axis>,<y axis>,cv2.CV_64F)
Vertical=cv2.Sobel(Gray,0,0,1,cv2.CV_64F)
# DO the Bitwise operation
Bitwise_Or=cv2.bitwise_or(Horizontal,Vertical)
# Show the Edged Image
cv2.imshow("Sobel Image",Bitwise_Or)
cv2.imshow("Original Image",Gray)
cv2.waitKey(0)
cv2.destroyAllWindows()

Output:

輸出：

2)拉普拉斯函數 (2) Laplacian Function)

# importing the module
import cv2
# read the image and store the data in a variable
image=cv2.imread("/home/abhinav/PycharmProjects/untitled1/b.jpg")
# make it grayscale
Gray=cv2.cvtColor(image,cv2.COLOR_BGR2GRAY)
# Make Laplacian Function
Lappy=cv2.Laplacian(Gray,cv2.CV_64F)
cv2.imshow("Laplacian",Lappy)
cv2.imshow("Original",Gray)
cv2.waitKey(0)
cv2.destroyAllWindows()

Output:

輸出：

3)使用Canny函數 (3) Using Canny Function)

# importing the module
import cv2
# read the image and store the data in a variable
image=cv2.imread("/home/abhinav/PycharmProjects/untitled1/b.jpg")
# make it grayscale
Gray=cv2.cvtColor(image,cv2.COLOR_BGR2GRAY)
# Make canny Function
canny=cv2.Canny(Gray,40,140)
# the threshold is varies bw 0 and 255
cv2.imshow("Canny",canny)
cv2.imshow("Original",Gray)
cv2.waitKey(0)
cv2.destroyAllWindows()

Output:

輸出：

翻譯自: https://www.includehelp.com/python/edge-detection-of-image-using-opencv-cv2.aspx