Converting an RGBA image to Grayscale and Binary

[Joni_Engr]

Hi, I have a RGBA image which I need to convert to Grayscale and then to binary with a threshold to get binary histogram but I have 2D Grayscale image in my code which I don't understand how to convert to binary.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

# Import libraries from PIL import Image import matplotlib.pyplot as plt import numpy as np import cv2   # Reading an Image image = Image.open('Board1_1.png')   # Properties of an Image print(image.size) print(image.format) print(image.mode)   # Grayscale conversion image_gray = image.convert('LA')   image_gray_array = 255 - np.asarray(image_gray)   image_gray_array_1 = image_gray_array[:,:,0] image_gray_array_2 = image_gray_array[:,:,1]   image_gray.show()

[Pedroski55]

I believe Python's module cv2 can do just about anything with images, and a lot more besides!

This code below gives me a threshold image. It's an excerpt from my own little amateurish OMR programme.

I save the steps on the way just so I can see what is going on.

This is all a bit on the "threshold" of my understanding! But it works!

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41

import cv2   destination = '/home/pedro/omr_steps/'   f = jpg_files[0] original = cv2.imread(f) name = 'step0_original.png' cv2.imwrite(destination + name, original)   # crop the bottom part of the image, the part with the choices, the part to mark # f is just a file path: savename = savepathMarked + 'temp_cropped_image.png' cropped_image = cropAll(f, destination, tup) # im is the image to write the stroked contours on im = cv2.imread(cropped_image) name = 'step1_cropped_image.png' cv2.imwrite(destination + name, im)   # make a grey image grey = cv2.cvtColor(im, cv2.COLOR_BGR2GRAY) name = 'step2_grey.png' cv2.imwrite(destination + name, grey)   # blur the grey image this helps the next step: Canny blurred = cv2.GaussianBlur(grey, (5, 5), 0) name = 'step3_blurred.png' cv2.imwrite(destination + name, blurred)   # get the edges of the blurred image edged = cv2.Canny(blurred, 75, 200) name = 'step4_edged.png' cv2.imwrite(destination + name, edged)   # find contours in the edge map, then use the contours on the thresh       edged2 = edged.copy() contours, hierarchy = cv2.findContours(edged2, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE) print('reading edged2, there are ', len(contours), ' contours')   thresh = cv2.threshold(grey, 0, 255, cv2.THRESH_BINARY_INV | cv2.THRESH_OTSU)[1] name = 'step5_thresh.png' cv2.imwrite(destination + name, thresh)

[Joni_Engr]

Hi, I get an error "ModuleNotFoundError: No module named 'cv2'"

I have already installed "pip install opencv-python" but still get an error

[Pedroski55]

In your post you had it imported! Why, if you are not using it?

I only use Linux, can't help if you use Windoze!

Normally, I install a module with pip, then it is available.