python - 如何绘制遮挡/重叠物体的轮廓？

Question

我试图让 OpenCV 识别出红色区域是一个轮廓，而不是将其分割成两个单独的轮廓。我尝试过破坏层次结构，但没有成功。有什么想法吗？

下面提供了图像和代码。

import cv2
import numpy as np
black = (0,0,0)
# All of these are in hsv format
lb=np.array([0,181,0])
ub=np.array([6,255,255])
# Load img
img = cv2.imread('test.jpg')
img=cv2.resize(img,(400,400))

kernelOpen=np.ones((5,5))
kernelClose=np.ones((20,20))
#convert BGR to HSV
imgHSV= cv2.cvtColor(img,cv2.COLOR_BGR2HSV)
# create the Mask
mask = cv2.inRange(imgHSV, lb, ub)
maskso=cv2.morphologyEx(mask,cv2.MORPH_OPEN,kernelOpen)
masksc=cv2.morphologyEx(maskso,cv2.MORPH_CLOSE,kernelClose)
conts,hiers=cv2.findContours(masksc.copy(),cv2.RETR_EXTERNAL,cv2.CHAIN_APPROX_NONE)
# Create convex hull
hull = []
for i in range(len(conts)):
    # creating convex hull object for each contour
    hull.append(cv2.convexHull(conts[i], False))
# Draw contour
cv2.drawContours(img,hull,-1,black,2)

cv2.imshow("image", img)
cv2.waitKey(5000)
cv2.destroyAllWindows()

Answer 1

替换此代码片段

# Create convex hull
hull = []
for i in range(len(conts)):
    # creating convex hull object for each contour
    hull.append(cv2.convexHull(conts[i], False))
# Draw contour
cv2.drawContours(img,hull,-1,black,2)

与

# Create convex hull
hull = np.zeros((0, 1, 2), np.int32)
for i in range(len(conts)):
    hull = np.concatenate((hull, conts[i]), axis=0)
hull = cv2.convexHull(hull, False)
# Draw contour
cv2.drawContours(img,[hull],-1,black,2)

我得到以下输出，我认为这是您想要的:

我不是收集每个轮廓的单个凸包，而是收集所有轮廓，然后获取所有轮廓点的凸包。

希望有帮助!

----------------------------------------
System information
----------------------------------------
Platform:    Windows-10-10.0.16299-SP0
Python:      3.8.1
NumPy:       1.18.1
OpenCV:      4.2.0
----------------------------------------