Histogram matching with OpenCV, scikit-image, and Python

Last Updated : 03 Jan, 2023

Histogram matching is used for normalizing the representation of images, it can be used for feature matching, especially when the pictures are from diverse sources or under varied conditions (depending on the light, etc). each image has a number of channels, each channel is matched individually. Histogram matching is possible only if the number of channels matches in the input and reference images.

The main target of histogram matching is:

For each image, we need to create histograms.
Take a look at the histogram of the reference image.
Using the reference histogram, update the pixel intensity values in the input picture such that they match.

match_histograms() method:

This method is used to modify the cumulative histogram of one picture to match the histogram of another. For each channel, the modification is made independently.

Syntax: skimage.exposure.match_histograms(image, reference, *, channel_axis=None, multichannel=False)

Parameters:

image: ndarray. This parameter is our input image it can be grayscale or a colour image.
reference: reference image. reference image to match histograms.
channel_axis: optional parameter. int or None. If None is specified, the picture will be presumed to be grayscale (single channel). if not, this argument specifies the array axis that corresponds to channels.
multichannel: optional parameter. boolean value. Matching should be done independently for each channel. This option has been deprecated; instead, use the channel axis.

Example 1: Using OpenCV and scikit-image.

The code begins with importing the necessary packages, reading images using the OpenCV imread() method, and then we check the number of channels of the input image and reference image, if they don't match we cannot perform histogram matching. match_histograms is used to find the matched image. Histograms are plotted for each channel.

Python3

# import packages import matplotlib.pyplot as plt from skimage import exposure from skimage.exposure import match_histograms import cv2  # reading main image img1 = cv2.imread("img.jpeg")  # checking the number of channels print('No of Channel is: ' + str(img1.ndim))  # reading reference image img2 = cv2.imread("2Fw13.jpeg")  # checking the number of channels print('No of Channel is: ' + str(img2.ndim))  image = img1 reference = img2  matched = match_histograms(image, reference ,                            multichannel=True)   fig, (ax1, ax2, ax3) = plt.subplots(nrows=1, ncols=3,                                      figsize=(8, 3),                                     sharex=True, sharey=True)  for aa in (ax1, ax2, ax3):     aa.set_axis_off()  ax1.imshow(image) ax1.set_title('Source') ax2.imshow(reference) ax2.set_title('Reference') ax3.imshow(matched) ax3.set_title('Matched')  plt.tight_layout() plt.show()  fig, axes = plt.subplots(nrows=3, ncols=3, figsize=(8, 8))  for i, img in enumerate((image, reference, matched)):     for c, c_color in enumerate(('red', 'green', 'blue')):         img_hist, bins = exposure.histogram(img[..., c],                                              source_range='dtype')         axes[c, i].plot(bins, img_hist / img_hist.max())         img_cdf, bins = exposure.cumulative_distribution(img[..., c])         axes[c, i].plot(bins, img_cdf)         axes[c, 0].set_ylabel(c_color)  axes[0, 0].set_title('Source') axes[0, 1].set_title('Reference') axes[0, 2].set_title('Matched')  plt.tight_layout() plt.show()

Output:

Example 2: Using only scikit-image:

This example is similar to the previous, except that we load images from the skimage.data package. Then we match histograms, display images, and plot histograms.

Python3

# importing packages import matplotlib.pyplot as plt from skimage import data from skimage import exposure from skimage.exposure import match_histograms  # loading data reference = data.moon() image = data.camera()  # matching histograms matched = match_histograms(image, reference,                             multichannel=True,)  fig, (ax1, ax2, ax3) = plt.subplots(nrows=1,                                      ncols=3,                                      figsize=(8, 3),                                     sharex=True,                                      sharey=True) for aa in (ax1, ax2, ax3):     aa.set_axis_off()  # displaying images ax1.imshow(image) ax1.set_title('Source image') ax2.imshow(reference) ax2.set_title('Reference image') ax3.imshow(matched) ax3.set_title('Matched image')  plt.tight_layout() plt.show()

Output"

Displaying histogram of the above-used images.

Python3

# displaying histograms. fig, axes = plt.subplots(nrows=3, ncols=3, figsize=(8, 8))  for i, img in enumerate((image, reference, matched)):     for c, c_color in enumerate(('red', 'green', 'blue')):         img_hist, bins = exposure.histogram(img[..., c],                                             source_range='dtype')         axes[c, i].plot(bins, img_hist / img_hist.max())         img_cdf, bins = exposure.cumulative_distribution(img[..., c])         axes[c, i].plot(bins, img_cdf)         axes[c, 0].set_ylabel(c_color)  axes[0, 0].set_title('Source image') axes[0, 1].set_title('Reference image') axes[0, 2].set_title('Matched image')  plt.tight_layout() plt.show()