opencv-python: how to crop image with bounding box coordinates [duplicate]

Question

I recognized pink wood in an image which is identified in the image below with the green box. Now I want to crop the image based on the box coordinates. I describe what I mean with a picture.

bounding box coordinate: x, y, w, h = (50, 1034, 119, 72)

input image

output expected (Manually cropped)

image1 - crop coordinates from the beginning of the image to the beginning of pink wood(bounding box)

image2 - crop coordinates from the end of the pink wood(bounding box)to the end of the image

for image 1 I wrote the blow code but it is wrong.

x, y => beginning of the image (0,0)

x, y => beginning of pink wood (50 1034)

from PIL import Image
img = Image.open("img.png")
img2 = img.crop((0, 0, 50, 1034))
img2.save("1.png")

x, y, w, h = (50, 1034, 119, 72) does not seem correct to me for the image you show. 50 should be more like about 10. Open your image in GIMP or Photoshop and measure the bounding box values and see if they measure correctly. — fmw42
– fmw42, Commented Nov 27, 2021 at 21:27

db_max · Accepted Answer · 2021-11-28 02:50:07Z

1

You can select your area of interest as an array:

import cv2

imagepath = yourPath
im = cv2.imread(imagepath)

imh, imw, _ = im.shape
x, y, w, h = (50, 1034, 119, 72)

img1 = im[:y, :]
img2 = im[y+h:imh, :]

cv2.imwrite('img1.tif', img1)
cv2.imwrite('img2.tif', img2)

answered Nov 28, 2021 at 2:50

db_max

2304 silver badges16 bronze badges

Sign up to request clarification or add additional context in comments.

Collectives™ on Stack Overflow

opencv-python: how to crop image with bounding box coordinates [duplicate]

1 Answer 1

Comments

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Linked

Related