Apply Threshold Filter to Image Using OpenCV 2.4

In this python exercise we convert a color image to grayscale and then apply threshold filter to it. The threshold filter converts the grayscale image to a binary image. We also save the final image to the working folder.

Note that a function called “shape” is used from the numpy library. The shape function gives the number of rows and columns of the source image. I’ve merely used the column information for position different windows side by side.

 1. Read a color image, convert it to grayscale and then use a thresholding
 filter to produce a binary image.
 2. Display the change produced by the two stages in different windows.
 3. Save the grayscale and binary images.

 Usage: needs a single argument: imageName

 Give the name of a color image file and include its path if the image file is
 not in the same location as

 Command line use:
 python /pathToFile/colorPic.jpg

 See cv2.nameWindow, cv2.imshow, cv2.imread, cv2.moveWindow, cv2.cvtColor and
 cv2.threshold in "OpenCV2 Reference Manual" for more information.

 Author: Sameer Khan (
 July 30th 2012
 No copyrights, warranty or guaranties of any kind are associated with this file.

## use sys, cv2 and numpy packages
import sys, cv2
import numpy as np

## main function
 def startExercise(imageName):
 ## read source image and get its shape
 sourceImage = cv2.imread(imageName, -1) # -1 is used to read the image as is
 imgRows, imgCols, imgChannels = np.shape(sourceImage)

 ## color to gray scale
 grayImage = cv2.cvtColor(sourceImage, cv2.COLOR_RGB2GRAY) # cv2.COLOR_RGB2GRAY = 7
 grayImgName = 'gray_' + imageName

 ## gray to binary: threshold = 100 (arbitrary); maxValue = 255; type = cv2.THRESH_BINARY
 flag, binaryImage = cv2.threshold(grayImage, 100, 255, cv2.THRESH_BINARY) # cv2.THRESH_BINARY = 0
 binaryImgName = 'binary_' + imageName

 ## display images in seprate windows and arrange the windows
 cv2.namedWindow('Color Image')
 cv2.moveWindow('Color Image', 0, 0) # position window at 0,0
 cv2.imshow('Color Image',sourceImage)

 cv2.namedWindow('Gray Scale Image')
 cv2.moveWindow('Gray Scale Image', imgCols, 0) # move window to the right
 cv2.imshow('Gray Scale Image',grayImage)

 cv2.namedWindow('Binary Image')
 cv2.moveWindow('Binary Image', 2*imgCols, 0) # move window further to the right
 cv2.imshow('Binary Image',binaryImage)

 ## wait for user input
 keyPress = cv2.waitKey(0) # 0 means wait indefinitely

 if keyPress == ord('s'): # save images and quit
  cv2.imwrite(grayImgName, grayImage)
  cv2.imwrite(binaryImgName, binaryImage)

  print "\nImages were saved in the working directory.\nPress any key to quit.\n"


 elif keyPress == ord('q'): # quit without saving images

## for using from the command line
if __name__ == '__main__':
 if len(sys.argv)!=2:
  print __doc__
  print "\nInstructions:\n\tClick on the display windows to bring them into focus.\n"
  print "Options:\n\tPress s to save images.\n\tPress q to quit without saving.\n"

  imageName = sys.argv[1]

## end of

One thought on “Apply Threshold Filter to Image Using OpenCV 2.4

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s