python image auto gamma correction
About Gamma
Gamma correction is a nonlinear store/reduce algorithm for luminance in the dynamic range, i.e., a nonlinear operation performed on the input value so that the output value is exponential with respect to the input value;
Gamma correction adjusts the overall brightness of an image in terms of its effect. An uncorrected image may look too bright or too dark, so Gamma correction is important if you want the image to display more perfectly.
The Gamma correction is calculated as follows:
output=〖input〗^(1/Gamma)
Transform the RGB values of each pixel using the exponential function above.
Specifically perform the following conversion formulas (assuming pixel values in the range of 0 to 255):
R=〖255X(R/255)〗^((1/gamma)) G=〖255X(G/255)〗^((1/gamma)) B=〖255X(B/255)〗^((1/gamma))
Generally dealing with gamma correction is accomplished by manually adjusting the gamma value of
But if there are a lot of images, setting the gamma value manually seems too cumbersome.
This is where automatic Gamma correction is employed to convert the RGB image to grayscale, the
Calculate the mean value of the data for the grayscale plot and calculate the gamma value by using the following formula.
gamma_val = math.log10(0.5) / math.log10(mean / 255)
python implementation
import cv2
import numpy as np
import math
import os
def gamma_trans(img, gamma): # gamma function handling
gamma_table = [(x / 255.0, gamma) * 255.0 for x in range(256)] # Establishment of mapping tables
gamma_table = ((gamma_table)).astype(np.uint8) # Color values are integers
return (img, gamma_table) # Image color lookup table. Alternatively adaptive algorithms can be designed based on the principle of light intensity (color) homogenization.
def nothing(x):
pass
data_base_dir = r'./1' # Enter the path to the folder
outfile_dir = r'./2' # Path to the output folder
list = (data_base_dir)
()
list2 = (outfile_dir)
()
for file in list: # Iterate over target folder images
read_img_name = data_base_dir + '/' + () # Fetch the full path of the image
image = (read_img_name) # Read in the picture
img_gray = (read_img_name, 0) # grayscale map reads for gamma calculation
mean = (img_gray)
gamma_val = math.log10(0.5) / math.log10(mean / 255) # The formula calculates gamma
image_gamma_correct = gamma_trans(image, gamma_val) # gamma transform
out_img_name = outfile_dir + '/' + ()
(out_img_name, image_gamma_correct)
print("The photo which is processed is {}".format(file))summarize
The above is a personal experience, I hope it can give you a reference, and I hope you can support me more.