How to remove black bounding box from the image using openCV or Python

Question

I have this retinal scanned image with a black rectangular box around this. How do I remove that portion alone using python or opencv since I have to carry out the operation for hundreds of images

The resultant image should look like this:

The cropped final and required image,basically after removing the bounding black box

It would be helpful if you can also provide an example of what you want to have as a result of removing the bounding box. — mahesh
I want the final image to be just the retinal eye(circular one) and the entire black color bounding the circle to be eliminated. — chatbot_chakra
@chatbot_chakra So you just want to change the black portion outside the retina's disc to white? I'd suggest flood fill. — beaker
It's not white actually. It's just the circle which i want. This is for feeding the images into a neural network , so the purpose is to feed only the useful part and leaving out the rest. — chatbot_chakra

mahesh mahesh · Accepted Answer · 2018-06-06T16:16:54

Well, here is something to get started:

import cv2

threshold = 25

img = cv2.imread('D:\\img.jpg', 0) # load grayscale version

# the indeces where the useful region starts and ends
hStrart = 0
hEnd = img.shape[0]
vStart = 0
vEnd = img.shape[1]

# get row and column maxes for each row and column
hMax = img.max(1)
vMax = img.max(0)

hDone_flag = False
vDone_flag = False

# go through the list of max and begin where the pixel value is greater 
# than the threshold
for i in range(hMax.size):
    if not hDone_flag:
        if hMax[i] > threshold:
            hStart = i
            hDone_flag = True

    if hDone_flag:
        if hMax[i] < threshold:
            hEnd = i
            break

for i in range(vMax.size):
    if not vDone_flag:
        if vMax[i] > threshold:
            vStart = i
            vDone_flag = True

    if vDone_flag:
        if vMax[i] < threshold:
            vEnd = i
            break

# load the color image and choose only the useful area from it
img2 = (cv2.imread('D:\\img.jpg'))[hStart:hEnd, vStart:vEnd,:]

# write the cropped image
cv2.imwrite("D:\\clipped.jpg", img2)

It may not be most elegant or the most efficient, but it does the job and you can get started with this. Maybe you can look up documentation and improve this.

How to remove black bounding box from the image using openCV or Python

1 Answers