histogram stretch Algorithm
The histogram stretch algorithm, also known as contrast stretching or normalization, is a digital image processing technique that aims to improve the contrast and overall quality of an image by redistributing the range of pixel intensity values. This method is particularly useful in cases where the image has poor contrast due to the narrow distribution of pixel values. The algorithm works by identifying the highest and lowest pixel intensities in the input image and then stretching the range of these values to fill the entire possible intensity range, such as 0 to 255 for an 8-bit grayscale image. This results in a more balanced distribution of pixel intensities, which enhances the contrast and can make the image easier to analyze or perceive visually.
To implement the histogram stretch algorithm, the minimum and maximum pixel intensity values are first identified in the original image. Then, a linear transformation is applied to each pixel in the image, rescaling its intensity value according to the minimum and maximum values found. The transformation function typically takes the form of a simple linear equation with the slope and intercept determined by the desired output range and the original intensity range. Once each pixel has been transformed, the resulting image will have a more evenly distributed histogram, with pixel intensities spread more uniformly across the available range. This enhanced contrast can reveal subtle details and features in the image that were previously obscured by the limited dynamic range.
"""
Created on Fri Sep 28 15:22:29 2018
@author: Binish125
"""
import copy
import os
import numpy as np
import cv2
import matplotlib.pyplot as plt
class contrastStretch:
def __init__(self):
self.img = ""
self.original_image = ""
self.last_list = []
self.rem = 0
self.L = 256
self.sk = 0
self.k = 0
self.number_of_rows = 0
self.number_of_cols = 0
def stretch(self, input_image):
self.img = cv2.imread(input_image, 0)
self.original_image = copy.deepcopy(self.img)
x, _, _ = plt.hist(self.img.ravel(), 256, [0, 256], label="x")
self.k = np.sum(x)
for i in range(len(x)):
prk = x[i] / self.k
self.sk += prk
last = (self.L - 1) * self.sk
if self.rem != 0:
self.rem = int(last % last)
last = int(last + 1 if self.rem >= 0.5 else last)
self.last_list.append(last)
self.number_of_rows = int(np.ma.count(self.img) / self.img[1].size)
self.number_of_cols = self.img[1].size
for i in range(self.number_of_cols):
for j in range(self.number_of_rows):
num = self.img[j][i]
if num != self.last_list[num]:
self.img[j][i] = self.last_list[num]
cv2.imwrite("output_data/output.jpg", self.img)
def plotHistogram(self):
plt.hist(self.img.ravel(), 256, [0, 256])
def showImage(self):
cv2.imshow("Output-Image", self.img)
cv2.imshow("Input-Image", self.original_image)
cv2.waitKey(5000)
cv2.destroyAllWindows()
if __name__ == "__main__":
file_path = os.path.join(os.path.basename(__file__), "image_data/input.jpg")
stretcher = contrastStretch()
stretcher.stretch(file_path)
stretcher.plotHistogram()
stretcher.showImage()
