Definition

: Relationship between input and output pixel values.

Depends on intensity, not spatial coordinates.

1. Introduction to Intensity Transformations (Page 1) Characteristics
Referred to as intensity (gray-level) transformations.

Well-Defined Mathematical Functions.

Types
Arbitrary User-Defined (Piecewise Linear) Functions.

Reverses intensity levels.

Negative Transformation Highlights hidden details in dark regions with dominant black areas.

Linear Transformations (Pages 1-2) Enhances white/gray details in dark images.

Identity Transformation Maintains original intensity levels.

Compresses dynamic range of high-intensity values.

Expands low-intensity values for better visibility.

Logarithmic Transformations (Pages 2-3)
Enhances darker regions while compressing brighter ones.
2. Well-Defined Mathematical Function-Based Transformations
Comprehensive methods for enhancing image quality. (Pages 1-7) Example: Reveals details in dark parts of an image.
9. Conclusion (Page 35) Enhances brighter regions, compresses darker regions.
Selection depends on image characteristics and enhancement goals. Exponential (Inverse Logarithmic) Transformations (Pages 3-4)
Example: Emphasizes bright regions in photographic images.

Adjusts contrast dynamically using power-law functions.

Medical Imaging: Enhances MRI scans for fine details.
Enhances low or high-intensity values based on parameters.
Aerial Photography: Adjusts contrast for terrain visibility. Power-Law (Gamma) Transformations (Pages 4-6)
8. Applications (Pages 33-34) Applications: Medical imaging, aerial photography.
Image Compression: Reduces storage requirements using bit-plane slicing.
Example: Improves visibility in MRI scans and adjusts contrast in photos.
Localized Adjustments: AHE/CLAHE enhance specific regions.

Equalization: Redistributes intensities for uniform contrast. Chapter 2 part II Definition

Histogram Equalization vs. Matching : User-defined transformations controlled manually.
Matching: Matches tones to a reference histogram.
7. Comparison of Methods (Pages 31-32) 3. Arbitrary User-Defined Transformations (Pages 7-9)
Contrast Stretching Expands gray-level range in low-contrast images.
Equalization: Simple, automatic.
Strengths and Limitations Gray-Level Slicing Highlights specific gray-level ranges.
Matching: Provides controlled output but needs a reference. Types
Analyzes individual bits in pixel representation.
Bit-Plane Slicing
Useful for compression and noise reduction.
Negative Transformation: Highlights white details in dark images.

Logarithmic Transformation: Compresses high values, reveals dark details.

Dark images: Intensities clustered at lower values.
Power-Law Transformation: Dynamically adjusts brightness and contrast. 6. Examples of Techniques (Pages 24-30)
Histogram Basics (Pages 10-11) Bright images: Intensities clustered at higher values.
Histogram Equalization: Enhances visibility in underexposed/overexposed images.
Well-contrasted images: Intensities spread over the full range.
AHE/CLAHE: Enhances fine details in localized areas.
Expands intensity range for better contrast.
Histogram Stretching (Pages 12-13)
Linear or nonlinear.

Redistributes intensities to achieve uniform contrast.

4. Histogram-Based Contrast Enhancement (Pages 10-20) Histogram Equalization (Pages 14-16)
Enhances underexposed images.

Applies localized adjustments within image tiles.

Adaptive Histogram Equalization (AHE) (Pages 16-18)
CLAHE: Limits contrast amplification, reduces noise.

Matches input histogram to a reference histogram.

Histogram Matching (Pages 19-20)
Used for tone matching and artistic effects.

Definition

5. Lookup Table (LUT) Transformations (Pages 21-23) : Pre-computed table mapping input intensity values to output values.

Applications Enhances contrast and adjusts brightness efficiently.