-1

I want to select a rectangular box from a image where all the content is placed. In other words I want remove the background of the image with unimportant(pixels which are not related to content) pixels. How ever the output image should be a rectangle.

What is the easiest way to do this in Python OpenCV? (I will import files from a folder... Better if I could automate the process)

Christoph Rackwitz
  • 11,317
  • 4
  • 27
  • 36
Oladiano Meriko Onaiker
  • 33
  • 4
  • Post and example image. What have you tried? Have you searched this forum or Google for possible solutions? Threshold your image to isolate your region. Get the contour of that region. Get the bounding box of that contour. Crop the image according to the bounding box. – fmw42 Oct 16 '22 at 18:24

1 Answers1

2

I hope what you asking is "Crop out the white spaces of image automatically". Here we assume a binary image : pixels have high and low values.

def focusToContent(img):
    img_ = 255*(img < 128).astype(np.uint8) 
    coords = cv.findNonZero(img_) # Find all non-zero points (text)
    x, y, w, h = cv.boundingRect(coords) # Find minimum spanning bounding box
    
    rect = img[y:y+h, x:x+w] # Crop the image - note we do this on the original image
    rect_originalSized = cv.resize(rect,(img.shape))
    return rect_originalSized

img should be a opencv image (numpy array with proper data type)

Test the code

#testing
img = cv.imread(r"D:/ENTC/SEM_4/EN2550 - Fundamentals of Image Processing and Machine Vision/~images/int-rec/test/1650009171.5083215.png",0)
assert img is not None
focused = focusToContent(img)

fig,ax = plt.subplots(1,2)
ax[0].imshow(img,cmap="gray")
ax[1].imshow(focused,cmap="gray")
plt.show()

enter image description here

R3KI3G
  • 96
  • 3