Digital Image Processing — Complete BSCS Notes
Introduction & Human Visual System
Definition: Digital Image Processing (DIP) is the manipulation of digital images using computer algorithms.
Human Visual System: The eye detects light and converts it into signals for the brain.
Electromagnetic Spectrum: Visible light is a small part of EM spectrum.
Example: Image enhancement in mobile camera.
Camera & Image Representation
Digital Camera: Converts light into digital signals using sensors.
Pixels: Smallest unit of image.
Image Representation: Image is stored as matrix of pixel values.
Example: 256×256 image = 65536 pixels.
Sampling & Quantization
Sampling: Converting continuous image into discrete pixels.
Quantization: Assigning intensity values.
Example: 8-bit image → 256 gray levels.
Mathematics of Image Formation
Convolution: Applying filter using kernel.
Camera Projection: 3D → 2D mapping.
Example: Blurring image using averaging filter.
Fourier & Frequency Domain
Fourier Transform: Converts image into frequency components.
Use: Noise removal, filtering.
Example: Removing high-frequency noise.
Image Filtering
Spatial Domain: Direct pixel processing.
Frequency Domain: Processing in Fourier space.
Example: Sharpening and smoothing filters.
Morphological Operations
Definition: Shape-based image processing.
Operations: Dilation, Erosion, Opening, Closing.
Example: Removing noise from binary images.
Color Models
Models: RGB, HSV, CMY.
Use: Color representation in images.
Example: RGB used in screens.
Feature Detection
Definition: Identifying important parts of image.
Examples: Edges, corners.
Use: Face detection, object recognition.
Image Segmentation & Pattern Recognition
Segmentation: Dividing image into regions.
Pattern Recognition: Identifying objects in image.
Example: Detecting tumors in medical images.
0 Comments