Mahotas – Haralick Feature

Mahotas is a computer vision and image processing library in Python that provides a lot of functions. One of those functions is Haralick feature computation on images.

What is Haralick feature?

Haralick features are texture features that are used in image analysis and pattern recognition. These features capture the texture pattern of an image by calculating the gray-level co-occurrence matrix of an image.

How Mahotas can be used to compute Haralick feature?

Mahotas provides a function mahotas.features.haralick to compute Haralick features of an image. This function takes an input image as a numpy array and returns a 13D Haralick feature vector for each pixel in the input image.

import mahotas
import numpy as np

# Load image as a numpy array
image = np.array(Image.open("example.jpg"))

# Calculate Haralick feature
haralick = mahotas.features.haralick(image)

# Output Haralick feature vector
print(haralick)

What are the different types of Haralick Feature?

There are 13 different types of Haralick features that are commonly used, and each feature captures a different aspect of texture pattern. Some of these features are:

1. Angular Second Moment
2. Contrast
3. Correlation
4. Inverse Difference Moment
5. Sum Average
6. Sum Variance
7. Sum Entropy
8. Entropy
9. Difference Variance
10. Difference Entropy
11. Information Measure of Correlation
12. Information Measure of Correlation 2
13. Max Correlation Coefficient

Conclusion

Mahotas is a powerful library for computer vision and image processing in Python. Its mahotas.features.haralick function allows us to compute Haralick feature of an image, which can be very useful in texture analysis and pattern recognition.