Contents of this chapter
- Principles of image display and storage
- Image Enhancement Goals
- image processing method
- Point Operations: Histogram-Based Contrast Enhancement
- Morphological processing
- Spatial Domain Processing: Convolution
- Application of convolution (smoothing, edge detection, sharpening, etc.)
- Frequency domain processing: Fourier transform, wavelet transform
- Application cases: smooth, edge, CLAHE, etc.
Principles of image display and storage
color space
RGB color space
- Additive Color Mixing: Color Displays
- Value range【0,255】【0.0,1.0】
CMY(K) color space
- Subtractive color mixing, printing
- 4-channel CMYK
- Value range [0,255], [0.0, 1.0]
HSV color space
- Concept of human vision, painter color matching
- 3 elements: HSV
CIE-XYZ color space
- International Association of Illumination, 1931
- Direct measurement based on human vision
- Human visual system - visual cells
- shortwave medium wavelength
- Tristimulus value channel (red, green, blue)
Image storage principle
mainstream color space
- three color map
- Single channel grayscale image
Image Enhancement Goals
- Improve image visuals
- Includes image sharpening, smoothing, denoising, gamma adjustment (contrast enhancement)
image processing method
Point Operations: Histogram-Based Contrast Enhancement
Histogram: quantizes the data space (bin).
Histogram equalization
Adaptive Histogram Equalization
CLASH
Morphological processing
Spatial Domain Processing: Convolution
Spatial Domain Processing and Transformation
border padding strategy
Airspace Analysis and Transformation
sum must be 0
Frequency Domain Analysis and Transformation
Application of convolution (smoothing, edge detection, sharpening, etc.)
Frequency domain processing: Fourier transform, wavelet transform