Remove black background from image

Question

I'm trying to remove the black background from this image:

in order to keep only the text and help Tesseract to recognize the image at its best. I'm doing it as follows

    img = cv2.imread(path_to_image)
    hh, ww = img.shape[:2]
    gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
    _, thresh = cv2.threshold(gray, 1, 255, cv2.THRESH_BINARY)

    contours = cv2.findContours(thresh, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
    contours = contours[0] if len(contours) == 2 else contours[1]
    cnt = max(contours, key=cv2.contourArea)
    # draw white contour on black background as mask
    mask = np.zeros((hh, ww), dtype=np.uint8)
    cv2.drawContours(mask, [cnt], 0, (255, 255, 255), cv2.FILLED)

    # invert mask so shapes are white on black background
    mask_inv = 255 - mask

    # create new (white) background
    bckgnd = np.full_like(img, (255, 255, 255))

    # apply mask to image
    image_masked = cv2.bitwise_and(img, img, mask=mask)

    # apply inverse mask to background
    bckgnd_masked = cv2.bitwise_and(bckgnd, bckgnd, mask=mask_inv)

    # add together
    result = cv2.add(image_masked, bckgnd_masked)

and this is what I get:

I still get a piece of the boundary of the popup that add noise to the reading of the image with Tesseract. I thought I would have removed also that piece of the image with the code that I posted and I don't know why it still remains after the process

Answer 1

I found out that the disconnected parts were causing me troubles so I added an erosion layer after thresholding like that:

_, thresh = cv2.threshold(gray, 1, 255, cv2.THRESH_BINARY)
thresh = cv2.erode(thresh, np.ones((3, 3), np.uint8))

and this is what I get:

MASK

FILTERED