Python OpenCV Histogram of Color Image

In this post, you will learn how to compute the histogram of a colour image using the OpenCV and the cv2.calcHist() function.

A histogram is a graphical representation of data. A histogram of an image represents how frequently various color values occur in the image or the distribution of pixel intensities in a colour or gray-scale image. The histogram plot shows the intensity distribution of an image. It is a plot with pixel values (ranging from 0 to 255, not always) in the X-axis and a corresponding number of pixels in the image on the Y-axis. It means the x-axis represents the different color values, which lie between 0 and 255, and the y-axis represents the number of times a particular intensity value occurs in the image.

Python OpenCV provides the cv2.calcHist() function to calculate the histogram of one or more arrays. We can use this function to calculate the histogram of both single channel images and multi-channel images. In this article, we have used a multi-channel image.

Syntax of calcHist()

cv2.calcHist(images, channels, mask, histSize, ranges)

images: It is the image source of type uint8 or float32 and given in a square bracket.
channels: It represents the index of the channel. It is also given in a square bracket. For grayscale images, its value is [0].
mask: It represents a mask image. To find the histogram of the full image, it is given as "None".
histSize: It represents the number of bins provided as a list.
ranges: It represents the range of intensity values.

Required Modules

These are the modules that we have used in this article to create histograms of a color image or multi-channel image.

• OpenCV (cv2)
• Numpy
• Matplotlib

Python color image histogram without Mask

Here is the code for calculating the histogram of a full multi-channel image. In this case, hist is a (256,1) array. Each value of the array corresponds to the number of pixels with the corresponding tone value.

In this example, the matplotlib library is used to plot the histograms. It provides the hist() function for plotting. The channel values [0], [1] and [2] represent blue, green, and red channels, respectively.

import cv2
import numpy as np
from matplotlib import pyplot as plt
  
img = cv2.imread('nature.jpg')

# Calculate histogram without mask
hist1 = cv2.calcHist([img],[0],None,[256],[0,256])
hist2 = cv2.calcHist([img],[1],None,[256],[0,256])
hist3 = cv2.calcHist([img],[2],None,[256],[0,256])

plt.subplot(221), plt.imshow(img)
plt.subplot(222), plt.plot(hist1), plt.plot(hist2),plt.plot(hist3)
plt.xlim([0,256])

plt.show()

The above code returns output something like this-

Python color image histogram with mask

First, you should know, "What is the mask?" The mask consists of a black image with the same dimensions as the loaded image and some white regions corresponding to the image where we want to calculate the histogram. Here is the complete code to draw a histogram of the image with a mask. In this example, it will be very clear for you to draw a histogram of a multi-channel image with and without the mask.

import cv2
import numpy as np
from matplotlib import pyplot as plt
  
img = cv2.imread('nature.jpg')

# create a mask
mask = np.zeros(img.shape[:2], np.uint8)
mask[100:300, 100:400] = 255
masked_img = cv2.bitwise_and(img,img,mask = mask)

# Calculate histogram without mask
hist1 = cv2.calcHist([masked_img],[0],mask,[256],[0,256])
hist2 = cv2.calcHist([masked_img],[1],mask,[256],[0,256])
hist3 = cv2.calcHist([masked_img],[2],mask,[256],[0,256])

plt.subplot(221), plt.imshow(img)
plt.subplot(222), plt.imshow(mask)
plt.subplot(223), plt.imshow(masked_img)
plt.subplot(224), plt.plot(hist1), plt.plot(hist2),plt.plot(hist3)
plt.xlim([0,256])

plt.show()

The above code returns the following output-

Related Articles