Grayscale and brightness smoothing
Execution environment
Preparing to load images with Google Colaboratory
python
from google.colab import files
from google.colab import drive
drive.mount('/content/drive')
Loading the required libraries
python
import cv2 #opencv
import matplotlib.pyplot as plt
%matplotlib inline
img = plt.imread("/content/drive/My Drive/Colab Notebooks/img/Lenna.bmp")
#↑ plt from this article.I decided to read it with imread.
Various conversions
python
plt.figure(figsize=(9, 6), dpi=100,
facecolor='w', linewidth=0, edgecolor='w')
#Original image
plt.subplot(3,3,1)
plt.imshow(img)
plt.subplot(3,3,4)
color = ('b','g','r')
for i,col in enumerate(color):
histr = cv2.calcHist([img],[i],None,[256],[0,256])
plt.plot(histr,color = col)
plt.xlim([0,256])
#grayscale
plt.subplot(3,3,2)
gray = cv2.cvtColor(img, cv2.COLOR_RGB2GRAY)
plt.imshow(gray)
plt.subplot(3,3,5)
plt.hist(gray.ravel(),256,[0,256])
#Brightness smoothing
plt.subplot(3,3,3)
dst = cv2.equalizeHist(gray)
plt.imshow(dst)
plt.subplot(3,3,6)
plt.hist(dst.ravel(),256,[0,256])
plt.show()
result
From the left Original / Grayscale / Brightness smoothing
grammar grayscale
python
cv2.cvtColor(src, cv2.COLOR_RGB2GRAY)
Brightness smoothing
python
cv2.equalizeHist(src)
By smoothing the brightness, the histogram spreads evenly, It's easier to understand the light and dark. It seems better to do this to detect the feature.