需要帮助使用开放的CV Python从图像中删除轮廓

Question

我正在尝试删除此矩形轮廓来自此照片。

我试图使用此教程 https://pyimagesearch.com/2015/02/09/removing-contours-image-image-using-python-opencv/ ，但这并不产生所需的结果，我对什么并不清楚什么是什么语法确实如此。

thresh = cv2.adaptiveThreshold(img, 255, cv2.ADAPTIVE_THRESH_MEAN_C, cv2.THRESH_BINARY_INV, 61, 10)
cntrs, hiearchy = cv2.findContours(thresh.copy(), cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
largest_areas = sorted(cntrs, key=cv2.contourArea)
largest_cntr = [largest_areas[-1]] #rectangular contour or largest contour
    
mask = np.ones(thresh.shape[:2], dtype="uint8") * 255    
cv2.drawContours(mask, largest_cntr, -1, 0, -1)
thresh = cv2.bitwise_and(thresh, thresh, mask=mask)
cv2.imshow("Mask", mask)
cv2.imshow("After", thresh)

最后，我得到了一个黑色的图像。如何正确从图像中删除轮廓？

完整代码：

import cv2,os, glob
from imutils.perspective import four_point_transform
from imutils import contours
import numpy as np


folder_dir = os.getcwd()
print(folder_dir)
img = cv2.imread(folder_dir + "/imgur.jpg")
img = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
thresh = cv2.adaptiveThreshold(img, 255, cv2.ADAPTIVE_THRESH_MEAN_C, cv2.THRESH_BINARY_INV, 31, 10)
cntrs, hiearchy = cv2.findContours(thresh.copy(), cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
largest_areas = sorted(cntrs, key=cv2.contourArea)
largest_cntr = [largest_areas[-1]] #rectangular contour or largest contour
    
mask = np.ones(thresh.shape[:2], dtype="uint8") * 255    
cv2.drawContours(mask, largest_cntr, -1, 0, -1)
thresh = cv2.bitwise_and(thresh, thresh, mask=mask)
cv2.imshow("Mask", mask)
cv2.imshow("After", thresh)
cv2.waitKey(0)
cv2.destroyAllWindows()