Digital-Image-Processing

Textbook followed: Digital Image Processing, 3rd edition Rafael C. Gonzales

Professor: Fred Fontaine

Independent study on digital image processing with a focus on computer vision and super-resolution. I mainly followed Gonzales' Digital Image Processing, reading through 1-2 sections a week, and towards the end switched to reading research papers in my areas of interest. Throughout the semester, I proceeded to implement a number of the algorithms I read about, which can be seen in the notebooks provided in the repository.

Primary tools: OpenCV, Pytorch

Topic Breakdown:

1. Digital Image Fundementals

Sampling
Pixel relationships
Linear algebra
Probability
Color spaces

2. Image Transformations

Intensity transformations
Histogram processing
Histogram matching
Spatial filters
Smoothing/Sharpening filters

3. Frequency domain filtering & Color

Sampling
Aliasing issues
DFT
2D FTs
Frequency based smoothing/sharpening
Selective filters
Color spaces
Color transformations
Colored Smoothing/sharpening
Noise in color images
Color-based segmentation

4. Noisy Images

Noise models:
- Gaussian
- Rayleigh
- Gamma
- Exponential
- Uniform
- Impulse
- Periodic noise
- Estimating noise
Spatial filtering for denoising
Frequency domain filtering for denoising
Estimating noise
Adaprive filters for noise reduction
MSE filtering
Least square filtering

5. Morphological Image Processing

Erosion
Dilation
Opening
Closing
Hit-or-Miss
Boundary extraction
Hole filling
Convex hull
Thickening
Thinning
Skeletons
Grey-scale morphology

6. Image Segmentations

Point detection
Line detection
Gradients
Edge detection
- Marr-Hildreth
- Canny
Edge models
Thresholding
Dam construction

7. Image Formation

Quaternions
Velocity fields
Perspective geometry
Photonic image formation
Stereo

8. Compression

Info theory
Lossless compression
Wavelets
Transform coding

9. Image Super-Resolution

From papers:

yuvalofek / digital-image-processing Goto Github PK