A basic Python image filter project

Coded in my CMPEN 454: Computer Vision class at Penn State during the Summer of 2021.

This project implements a simple 2D image filtering pipeline using NumPy. It includes custom padding (zero and replication) and convolution logic (with manual 180° filter rotation) for grayscale images. The code demonstrates how image filters like blur, sharpen, and edge detectors can be applied without relying on external image processing libraries.

Pads a grayscale image using either:

• 'zero': Pads with zeros (default)
• 'replication': Replicates border pixels

Applies a custom filter to an image using:

• Manual 180° filter rotation
• Sliding window convolution
• Uses replication padding

import numpy as np
from filter import myImageFilter

# Sample 5x5 grayscale image
image = np.array([
    [10, 20, 30, 40, 50],
    [15, 25, 35, 45, 55],
    [20, 30, 40, 50, 60],
    [25, 35, 45, 55, 65],
    [30, 40, 50, 60, 70]
])

# 3x3 averaging filter
kernel = np.array([
    [1/9, 1/9, 1/9],
    [1/9, 1/9, 1/9],
    [1/9, 1/9, 1/9]
])

# Apply the image filter
output = myImageFilter(image, kernel)
print(output)