如何使用 OpenCv + Python 找到 blob 的轮廓

Question

Goal

I want to use OpenCV in Python to find the precise contour of the blob , but ignore all the tentacles in the picture below:

I need to find the precise contour so using dilate/erode is not ideal.

Answer 1

You can use cv2.morphologyEx() with the morph open operation ( cv2.MORPH_OPEN ) and an elliptical shaped kernel ( cv2.MORPH_ELLIPSE ) created with cv2.getStructuringElement() . This is essentially an erosion followed by a dilation so this operation will remove the noise and keep the spherical shape of the blob. You can experiment with various kernel sizes to remove more or less of the blob and the strength of the operation by changing the number of iterations. Here's the result after performing a morph open:

From here we find the bounding box coordinates and draw a rectangle around the blob.

import cv2

image = 255 - cv2.imread('1.png')
gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
kernel = cv2.getStructuringElement(cv2.MORPH_ELLIPSE, (3,3))
opening = cv2.morphologyEx(gray, cv2.MORPH_OPEN, kernel, iterations=4)

x,y,w,h = cv2.boundingRect(opening)
cv2.rectangle(image, (x, y), (x + w, y + h), (36,255,12), 2)

cv2.imshow('opening', opening)
cv2.imshow('image', image)
cv2.waitKey()