python - 如何从Vitruvian Man OPENCV查找和轮廓化圆形和矩形

Question

我必须在OpenCV中执行此图像之间的圆形和矩形的检测以及检测:

我正在尝试:

import cv2
import numpy as np

cv2.namedWindow("jan", cv2.WINDOW_AUTOSIZE)
image = cv2.imread("vit.jpg", cv2.IMREAD_ANYCOLOR)
gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
edge = cv2.Canny(gray,500,500)

contours, hierarchy = cv2.findContours(edge,cv2.RETR_TREE,cv2.CHAIN_APPROX_SIMPLE)[-2:]
print(contours)

cv2.drawContours(image, contours, -1, (0,0,255), 10)
cv2.imshow("jan", image)

cv2.waitKey(0)
cv2.destroyAllWindows()

但是我得到这个结果:

如何仅使用圆形和矩形执行此操作？

Answer 1

您可以绘制的圆圈如下:

image = cv2.imread("vit.jpg", cv2.IMREAD_ANYCOLOR)
grayImage = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
edgeImage = cv2.Canny(grayImage, 50, 100)

circles = cv2.HoughCircles(edgeImage,cv2.HOUGH_GRADIENT,1,240,
                            param1=50,param2=20,minRadius=0,maxRadius=0)

if circles is not None:
    for circles in circles[0, :]: 
        x = circles[0]
        y = circles[1]
        radius = int(circles[2])
        cv2.circle(image, (x,y), radius, (255,0,0), 5)   

和您的返回: