如何使用 OpenCV 在Python中检测图像中的形状？

先决条件： OpenCV

OpenCV 是一个开源库，主要用于处理图像和视频以识别形状、对象、文本等。它主要与Python一起使用。在本文中，我们将了解如何检测图像中的形状。为此，我们需要 OpenCV 的cv2.findContours()函数，并且我们将使用cv2.drawContours()函数在图像上绘制边缘。轮廓是形状的轮廓或边界。

方法

导入模块
导入图片
将其转换为灰度图像
对图像应用阈值，然后找出轮廓。
在轮廓范围内运行一个循环并遍历它。
在这个循环中绘制形状的轮廓（使用 drawContours() ）并找出形状的中心点。
根据检测到的形状所具有的轮廓点的数量对检测到的形状进行分类，并将检测到的形状名称放在形状的中心点。

使用的函数

cv2.findContours()：基本上这个方法找出图像中形状的所有边界点。

Syntax: cv2.findContours(src, contour_retrieval, contours_approximation)

Parameters:

src: input image n-dimensional (but for in our example we are going to use 2 dimensional image which is
mostly preferred.)
contour_retrieval:
- cv.RETR_EXTERNAL:retrieves only the extreme outer contours
- cv.RETR_LIST:retrieves all of the contours without establishing any hierarchical relationships.
- cv.RETR_TREE:retrieves all of the contours and reconstructs a full hierarchy of nested contours.
contours_approximation:
- cv.CHAIN_APPROX_NONE: It will store all the boundary points.
- cv.CHAIN_APPROX_SIMPLE: It will store number of end points(eg.In case of rectangle it will store 4)

Return value: list of contour points

编程需要懂一点英语

cv2.drawContours() ：此方法绘制轮廓。如果您提供边界点，它还可以绘制形状。

Syntax: cv.DrawContours(src, contour, contourIndex, colour, thickness)

Parameters:

src: n dimensional image
contour: contour points it can be list.
contourIndex:
- -1:draw all the contours
To draw individual contour we can pass here index value
- color:color values
- thickness: size of outline

编程需要懂一点英语

输入：

程序：

Python3

import cv2
import numpy as np
from matplotlib import pyplot as plt
  
# reading image
img = cv2.imread('shapes.png')
  
# converting image into grayscale image
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
  
# setting threshold of gray image
_, threshold = cv2.threshold(gray, 127, 255, cv2.THRESH_BINARY)
  
# using a findContours() function
contours, _ = cv2.findContours(
    threshold, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)
  
i = 0
  
# list for storing names of shapes
for contour in contours:
  
    # here we are ignoring first counter because 
    # findcontour function detects whole image as shape
    if i == 0:
        i = 1
        continue
  
    # cv2.approxPloyDP() function to approximate the shape
    approx = cv2.approxPolyDP(
        contour, 0.01 * cv2.arcLength(contour, True), True)
      
    # using drawContours() function
    cv2.drawContours(img, [contour], 0, (0, 0, 255), 5)
  
    # finding center point of shape
    M = cv2.moments(contour)
    if M['m00'] != 0.0:
        x = int(M['m10']/M['m00'])
        y = int(M['m01']/M['m00'])
  
    # putting shape name at center of each shape
    if len(approx) == 3:
        cv2.putText(img, 'Triangle', (x, y),
                    cv2.FONT_HERSHEY_SIMPLEX, 0.6, (255, 255, 255), 2)
  
    elif len(approx) == 4:
        cv2.putText(img, 'Quadrilateral', (x, y),
                    cv2.FONT_HERSHEY_SIMPLEX, 0.6, (255, 255, 255), 2)
  
    elif len(approx) == 5:
        cv2.putText(img, 'Pentagon', (x, y),
                    cv2.FONT_HERSHEY_SIMPLEX, 0.6, (255, 255, 255), 2)
  
    elif len(approx) == 6:
        cv2.putText(img, 'Hexagon', (x, y),
                    cv2.FONT_HERSHEY_SIMPLEX, 0.6, (255, 255, 255), 2)
  
    else:
        cv2.putText(img, 'circle', (x, y),
                    cv2.FONT_HERSHEY_SIMPLEX, 0.6, (255, 255, 255), 2)
  
# displaying the image after drawing contours
cv2.imshow('shapes', img)
  
cv2.waitKey(0)
cv2.destroyAllWindows()

输出：