Unit 12.1A · Term 1

PIL Image Filters

Filters transform images by modifying pixel values based on their neighbors. PIL provides built-in filters (blur, sharpen, edge detect) and tools to build custom filters pixel by pixel.

Learning Objectives

12.6.2.4 Create filters for image processing

Lesson Presentation

12.1A-pil-filters.pdf · Slides for classroom use

View Download

Conceptual Anchor

The Instagram Filter Analogy

Every photo filter on your phone is code that reads pixel values, applies a mathematical formula, and writes new pixel values. PIL gives you the same power — blur, sharpen, edge detection, color adjustments — all from Python.

Rules & Theory

Built-in Filters (ImageFilter)

Filter	Effect	Usage
`BLUR`	Smooths image, reduces noise	`img.filter(ImageFilter.BLUR)`
`GaussianBlur(r)`	Gaussian blur with radius r	`img.filter(ImageFilter.GaussianBlur(5))`
`SHARPEN`	Enhances edges, makes image crisp	`img.filter(ImageFilter.SHARPEN)`
`CONTOUR`	Extracts contour lines	`img.filter(ImageFilter.CONTOUR)`
`DETAIL`	Enhances fine details	`img.filter(ImageFilter.DETAIL)`
`EDGE_ENHANCE`	Subtly enhances edges	`img.filter(ImageFilter.EDGE_ENHANCE)`
`FIND_EDGES`	Finds all edges (like a sketch)	`img.filter(ImageFilter.FIND_EDGES)`
`EMBOSS`	3D embossed effect	`img.filter(ImageFilter.EMBOSS)`
`SMOOTH`	Light smoothing	`img.filter(ImageFilter.SMOOTH)`

Pixel-Level Adjustments (ImageEnhance)

Enhancer	What It Adjusts	Factor
`ImageEnhance.Brightness`	Light/dark	0.0 = black, 1.0 = original, 2.0 = 2× bright
`ImageEnhance.Contrast`	Difference between darks and lights	0.0 = gray, 1.0 = original, 2.0 = 2× contrast
`ImageEnhance.Color`	Color saturation	0.0 = grayscale, 1.0 = original
`ImageEnhance.Sharpness`	Edge definition	0.0 = blurred, 2.0 = extra sharp

from PIL import Image, ImageFilter, ImageEnhance

img = Image.open("photo.jpg")

# Apply built-in filter
blurred = img.filter(ImageFilter.GaussianBlur(3))

# Adjust brightness
enhancer = ImageEnhance.Brightness(img)
bright = enhancer.enhance(1.5)  # 50% brighter

Worked Examples

1 Applying Multiple Filters

from PIL import Image, ImageFilter

img = Image.open("photo.jpg")

# Chain filters: sharpen then find edges
result = img.filter(ImageFilter.SHARPEN)
result = result.filter(ImageFilter.FIND_EDGES)
result.save("sketch_effect.png")

2 Custom Grayscale Filter (Pixel by Pixel)

from PIL import Image

img = Image.open("photo.jpg")
w, h = img.size

# Convert to grayscale manually
for x in range(w):
    for y in range(h):
        r, g, b = img.getpixel((x, y))
        # Weighted average (human eye sensitivity)
        gray = int(0.299 * r + 0.587 * g + 0.114 * b)
        img.putpixel((x, y), (gray, gray, gray))

img.save("manual_gray.png")

3 Custom Sepia Tone Filter

from PIL import Image

img = Image.open("photo.jpg")
w, h = img.size

for x in range(w):
    for y in range(h):
        r, g, b = img.getpixel((x, y))
        # Sepia formula
        tr = min(255, int(0.393*r + 0.769*g + 0.189*b))
        tg = min(255, int(0.349*r + 0.686*g + 0.168*b))
        tb = min(255, int(0.272*r + 0.534*g + 0.131*b))
        img.putpixel((x, y), (tr, tg, tb))

img.save("sepia.png")

4 Threshold (Black & White) Filter

from PIL import Image

img = Image.open("photo.jpg")
w, h = img.size
threshold = 128

for x in range(w):
    for y in range(h):
        r, g, b = img.getpixel((x, y))
        avg = (r + g + b) // 3
        if avg > threshold:
            img.putpixel((x, y), (255, 255, 255))  # white
        else:
            img.putpixel((x, y), (0, 0, 0))        # black

img.save("bw_threshold.png")

5 Custom Color Channel Filter

from PIL import Image

img = Image.open("photo.jpg")
w, h = img.size

# Keep only the red channel
for x in range(w):
    for y in range(h):
        r, g, b = img.getpixel((x, y))
        img.putpixel((x, y), (r, 0, 0))

img.save("red_only.png")

Common Pitfalls

filter() Returns a New Image

img.filter() does NOT modify the original. Always assign the result: blurred = img.filter(ImageFilter.BLUR).

Pixel Values Must Be 0–255

When computing new pixel values, always clamp them: min(255, max(0, value)). Going above 255 or below 0 causes errors.

Tasks

Apply

Write a program that applies Gaussian Blur, then saves the blurred version of any image.

Apply

Write a custom "negative" filter that inverts all pixel colors (255 – value).

Create

Create your own "warm" filter: increase red by 20, decrease blue by 20, keep green the same. Apply it pixel-by-pixel.

Self-Check Quiz

Q1: What module provides BLUR, SHARPEN, FIND_EDGES?

ImageFilter from PIL — used with img.filter(ImageFilter.BLUR).

Q2: How would you make an image 50% brighter?

enhancer = ImageEnhance.Brightness(img)
bright = enhancer.enhance(1.5)

Q3: What does a threshold filter do?

Converts each pixel to black or white based on whether its average intensity is above or below a threshold value (e.g., 128).