histogram_equalization.py

#------------------------------------#
# Author: Yueh-Lin Tsou              #
# Update: 7/20/2019                  #
# E-mail: hank630280888@gmail.com    #
#------------------------------------#

"""---------------------------------
Implement Histogram Equalization
---------------------------------"""

import pylab as pl	# matplotlib's subpackage as pl for graph
import numpy as np	# use numpy library as np for array object
import argparse
import cv2			# opencv-python

# ------------------ Do image histogram equalization ------------------ #
def Hist_Equal(img):

	image = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)# convert image to grayscal

	# count values for each pixels(image histogram)
	x, y = image.shape 				# get image size x*y, for a image with x rows and y columns
	histo = [0.0] * 256 			# Initializes a 256 array to hold the number of
									# occurrences of each gray level
	for i in range(0, x): 			# count each pixel value(gray level) from 0 to x
		for j in range(0, y): 		# count each pixel value(gray level) from 0 to y
			histo[image[i, j]]+=1	# count the number of occurrences of each gray level

	# cdf and new pixels values
	cdf = np.cumsum(np.array(histo))						# cumulative distribution function
	trans_val = np.uint8((cdf-cdf[0])/((x*y)-cdf[0])*255)	# calculate transfer value

	# applying transfered values for each pixels
	new_img = np.zeros_like(image) 	# Return an empty array new_img with shape and type of input.
	for i in range(0, x):			# count each pixel value(gray level) from 0 to x
		for j in range(0, y):		# count each pixel value(gray level) from 0 to y
			new_img[i, j] = trans_val[image[i, j]]	#fill the equalization result in the array

	# show original and equalized image
	pl.subplot(121)					# image position 1 row, 2 column, first position
	pl.imshow(image)				# show image "grayimg"
	pl.title('original image')		# graph title "original image"
	pl.set_cmap('gray')				# show in gray scale

	pl.subplot(122)					# image position 1 row, 2 column, second position
	pl.imshow(new_img)				# show image "grayimg"
	pl.title('equalized image')		# graph title "original image"
	pl.set_cmap('gray')				# show in gray scale

	pl.show()						# output image

# -------------------------- main -------------------------- #
if __name__ == '__main__':
	# read one input from terminal
    # command line >> python histogram_equalization.py -i input_image.jpg

	ap = argparse.ArgumentParser()
	ap.add_argument("-i", "--image", required = True, help="Path to the input image")
	args = vars(ap.parse_args())

	# read image
	image = cv2.imread(args["image"])

	# function
	Hist_Equal(image) # do histogram equalization