Python Pillow - ImageMath.eval() Function

---

---

The Pillow library provides a function called eval() within its ImageMath module, to evaluate image expressions. It allows you to perform operations, such as arithmetic, bitwise, and logical operations, on images. The module supports standard Python expression syntax with additional functions provided by the Pillow library.

The PIL.ImageMath.eval() function evaluates an expression in a given environment. This function supports standard Python expression syntax along with additional functions for image manipulation.

It is important to note that, the ImageMath module currently gives support for single-layer images only. For processing multi-band images, you should use the split() method or merge() function.

Syntax

Following is the syntax of the function −

PIL.ImageMath.eval(expression, environment)

Parameters

Here are the details of this function parameters −

• expression − A string representing a Python expression.

• environment − A dictionary or keyword arguments mapping image names to Image instances.

Return Value

The function returns an image, an integer value, a floating-point value, or a pixel tuple, depending on the expression.

Examples

Example 1

This example demonstrates how to evaluate an expression with a standard arithmetical operator using the eval() function of the ImageMath module.

from PIL import Image, ImageMath

# Open an image and convert it to grayscale
image1 = Image.open("Images/black rose.jpg").convert('L')

# Evaluate the expression 'a + 100' to invert pixel values
resultant_image = ImageMath.eval('a + 100', a=image1)

# Display the original and resultant images
image1.show()
resultant_image.show()
print('Evaluated the expression with standard arithmetical operator successfully...')

Output

Input Image

Resultant Image

Evaluated the expression with standard arithmetical operator successfully...

Example 2

Here is an example that demonstrates how to evaluate an expression with a Bitwise operator using the eval() function.

It is important to note that Bitwise operators are not applicable to floating-point images.

from PIL import Image, ImageMath

# Open and convert the first grayscale image
image1 = Image.open("Images/ColorDots.png").convert('L')

# Open and convert the second grayscale image
image2 = Image.open("Images/TP-W.png").convert('L')

# Apply bitwise AND operation on the two images
resultant_image = ImageMath.eval('a & b', a=image1, b=image2)

# Display the original images and the result
image1.show()
image2.show()
resultant_image.show()
print('Evaluated the expression of the Bitwise operator successfully...')

Output

Input Image 1

Input Image 2

Output Image

Example 3

The following example applies the logical or operator on the entire images.

from PIL import Image, ImageMath

# Open and convert the first grayscale image
image1 = Image.open("Images/ColorDots.png").convert('L')

# Open and convert the second grayscale image
image2 = Image.open("Images/TP-W.png").convert('L')

# Use the logical "or" operation on the entire images
resultant_image = ImageMath.eval('(a or b)', a=image1, b=image2)

# Display the original images and the result
image1.show()
image2.show()
resultant_image.show()

print('Evaluated the expression using the logical OR successfully...')

Output

Input Image 1

Input Image 2

Output Image

Example 3

The following example evaluates an expression with a Built-In function using the ImageMath.eval() function.

from PIL import Image, ImageMath

# Open and convert the first grayscale image
image1 = Image.open("Images/ColorDots.png").convert('L')

# Open and convert the second grayscale image
image2 = Image.open("Images/TP-W.png").convert('L')

# Evaluate the expression "min(a, b)" using ImageMath
resultant_image = ImageMath.eval('min(a, b)', a=image1, b=image2)

# Display the original images and the result
image1.show()
image2.show()
resultant_image.show()

print('Evaluated the expression using the Built-in Functions successfully...')

Output

Input Image 1

Input Image 2

Output Image