Perform Distance Transformation on an Image in OpenCV Python

We can perform the distance transform using the method cv2.distanceTransform(). Following is the syntax of this method.

Syntax

cv2.distanceTransform(src, distanceType, maskSize)

This method accepts the following parameters ?

• src ? 8-bit, single-channel (binary) source image.

• distanceType ? Type of the distance.

• maskSize ? Size of the distance transform mask.

Steps

To perform distance transform on the image, we could follow the below steps-

• Import the required library. In all the following examples, the required Python library is OpenCV. Make sure you have already installed it.

• Read an input image using cv2.imread(). The RGB image read using this method is in BGR format. Optionally assign the read BGR image to img.

• Now convert this BGR image to grayscale image as below using cv2.cvtColor() function. Optionally assign the converted grayscale image to gray.

• Now apply thresholding on the grayscale image to convert it to a binary image. Adjust the second parameter (threshValue) for better binary image.

• Apply distance transform on the binary image using cv2.distanceTransform(). It returns a distance transformed image. Normalize this image for range [0,1].

• Display the distance transformed image.

Let's have a look at some examples for more clear understanding.

Input Image

We will use this image as the input file in the examples below.

Example

In this example, we find the distance transform of the input image. We apply cv2.DIST_L2 as distanceType and the maskSize of 3.

# Load image in grayscale
import cv2
import numpy as np
import matplotlib.pyplot as plt

# Read the input image
img = cv2.imread('sketch.jpg')

# convert the image to grayscale
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)

# apply thresholding to convert the grayscale image to a binary image
ret,thresh = cv2.threshold(gray,150,255,cv2.THRESH_BINARY)

# Apply distance transform on the binary image
dist = cv2.distanceTransform(thresh, cv2.DIST_L2, 3)

# Normalize the distance image for range = {0.0, 1.0}

# so we can visualize and threshold it
cv2.normalize(dist, dist, 0, 1.0, cv2.NORM_MINMAX)
cv2.imshow('Distance Transform Image', dist)
cv2.waitKey(0)
cv2.destroyAllWindows()

Output

When you run the above Python program, it will produce the following output window ?

Example

In this example, we find the distance transform of the input image. We apply five different types of distanceType and the maskSize of 3.

# import required libraries
import cv2
import matplotlib.pyplot as plt

# Read the input image
img = cv2.imread('sketch.jpg')

# convert the image to grayscale
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)

# Apply thresholding to convert the grayscale image to a binary image
ret,thresh = cv2.threshold(gray,150,255,cv2.THRESH_BINARY)

# Apply distance transform on the binary image
dist_C = cv2.distanceTransform(thresh, cv2.DIST_C, 3)
dist_L1 = cv2.distanceTransform(thresh, cv2.DIST_L1, 3)
dist_L2 = cv2.distanceTransform(thresh, cv2.DIST_L2, 3)
dist_LP = cv2.distanceTransform(thresh, cv2.DIST_LABEL_PIXEL, 3)
dist_M = cv2.distanceTransform(thresh, cv2.DIST_MASK_3, 3)

# Normalize the distance image for range = {0.0, 1.0}

# so we can visualize and threshold it
cv2.normalize(dist_C, dist_C, 0, 1.0, cv2.NORM_MINMAX)
cv2.normalize(dist_L1, dist_L1, 0, 1.0, cv2.NORM_MINMAX)
cv2.normalize(dist_L2, dist_L2, 0, 1.0, cv2.NORM_MINMAX)
cv2.normalize(dist_LP, dist_LP, 0, 1.0, cv2.NORM_MINMAX)
cv2.normalize(dist_M, dist_M, 0, 1.0, cv2.NORM_MINMAX)

# visualize the distance images
plt.subplot(231),plt.imshow(dist_C, cmap = 'gray')
plt.title('DIST_C'), plt.xticks([]), plt.yticks([])
plt.subplot(232),plt.imshow(dist_L1, cmap = 'gray')
plt.title('DIST_L1'), plt.xticks([]), plt.yticks([])
plt.subplot(233),plt.imshow(dist_L2, cmap = 'gray')
plt.title('DIST_L2'), plt.xticks([]), plt.yticks([])
plt.subplot(234),plt.imshow(dist_LP, cmap = 'gray')
plt.title('DIST_LABEL_PIXEL'), plt.xticks([]), plt.yticks([])
plt.subplot(235),plt.imshow(dist_M, cmap = 'gray')
plt.title('DIST_MASK_3'), plt.xticks([]), plt.yticks([])
plt.show()

Output

When you run the above Python program, it will produce the following output window showing different transforms obtained after applying different distanceType.

Notice the difference among these five distanceTypes.