Translation transformation in OpenCV

Translation is an affine transformation, that moves an object or image by a fixed distance along the x and y axes. Transformation matrix for this operation is defined like below.

 

[1 0 tx]

[0 1 ty]

 

Here, (tx, ty) represent the translation distances in the x and y directions. Using cv2.wrapAffine method and a translation matrix, we can shift the image toward x and y axis in OpenCV.

 

Example

def translate(img, x, y):
    translation_matrix = np.float32([[1, 0, x], [0, 1, y]])
    width = img.shape[1]
    height = img.shape[0]

    dimensions = (width, height)
    return cv.warpAffine(img, translation_matrix, dimensions)

 

a.   If x is positive, then it moves the image to Right

b.   If x is negative, then it moves the image to Left

c.    If y is positive, then it moves the image to Down

d.   If y is negative, then it moves the image to Up

 

Find the below working application.

 

translation.py

import cv2 as cv
import numpy as np

# x : Right
# -x : Left
# y : Down
# -y : Up
def translate(img, x, y):
    translation_matrix = np.float32([[1, 0, x], [0, 1, y]])
    width = img.shape[1]
    height = img.shape[0]

    dimensions = (width, height)
    return cv.warpAffine(img, translation_matrix, dimensions)

# Read the image as matrix of pixels
image = cv.imread('nature.png')
image = cv.resize(image, (1000, 800))

# Display the image in new window
cv.imshow('Nature', image)
cv.imshow('Image shifted 150 pixels right and 100 pixels down', translate(image, 150, 100))
cv.imshow('Image shifted 150 pixels left and 200 pixels up', translate(image, -150, -200))

# Wait for Infinite amount of time for a keyboard key to be pressed
cv.waitKey(0)

# Close the OpenCV windows
cv.destroyAllWindows()

 

Output

 

Original image

 

 

Image shifted 150 pixels right and 100 pixels down

 

 

Image shifted 150 pixels left and 200 pixels up.

 

Previous                                                    Next                                                    Home