MATH3360
Mathematical imaging
Spectral and variational image restoration, wavelets, image geometry and scale space, feature detection, motion, morphology and segmentation.
Your progress
0/ 24 sections marked complete
Next section: Fourier analysis and stable image restoration
Completion marks are saved in this browser.
This course in the web
In the MATH3360 notes
- Theorem cards
- 47
- Definitions
- 34
- Counterexamples
- 25
- Concept lenses
- 17
- Histories of ideas
- 3
- Proof X-rays
- 24
Sections with a history of an idea
- Total variation, variational restoration, and diffusion2.1
- Structure tensors, scale-space features, and image pyramids5.1
- Region energies, moving curves, and level sets10.1
None of these is on the timeline yet.
No concept lens from this course is recorded in the web yet.
Counts cover the 24 public sections of this course and what is written in them, not recorded connections. 0 of these sections are indexed into the web in detail so far.
In the webTheorem cards 47 · Definitions 34 · Counterexamples 25 · Concept lenses 17 · Histories of ideas 3 · Proof X-rays 24
None of these is on the timeline yet.
No concept lens from this course is recorded in the web yet.
Counts cover the 24 public sections of this course and what is written in them, not recorded connections. 0 of these sections are indexed into the web in detail so far.
Chapter 1 · 5 sections
Spectral restoration
Convolution, unstable inversion, and transform regularization.
Theorems 10Counterexamples 5Lenses 4Proof X-rays 5
Chapter outline →See this chapter on the web →Not yet indexed in the web
- 1.1
Fourier analysis and stable image restoration
Prove the periodic Fourier model and compare Tikhonov filtering with Haar shrinkage.
- 1.3
Poisson counting and multiplicative image restoration
Derive a Poisson likelihood, uniqueness condition, and descent step.
- 1.4
Generalized quadratic restoration and vanishing noise
Understand derivative penalties, common nullspaces, and a rigorous small-noise limit in finite-dimensional imaging.
- 1.5
Inpainting as a graph boundary-value problem
Derive harmonic interpolation of missing pixels, establish when it is unique, and examine its limitations.
- 1.6
Linear minimum mean-square restoration
Derive a statistical restoration filter and distinguish risk from residual size.
Chapter 2 · 4 sections
Variational imaging
TV denoising, dual certificates, and diffusion.
Theorems 8 · Definitions 7Counterexamples 4Lenses 1History 1Proof X-rays 4
Chapter outline →See this chapter on the web →Not yet indexed in the web
- 2.1
Total variation, variational restoration, and diffusion
Establish finite-dimensional TV guarantees and distinguish smooth flows from discrete schemes.
- 2.2
Nonlinear edge diffusion on a finite graph
Prove well-posedness, nonexpansiveness and stability of a smoothed graph-TV flow.
- 2.3
An implicit step for singular graph total variation
Prove existence, conservation and stability of a nonsmooth graph-TV proximal step.
- 2.4
Graph coarea: from TV denoising to segmentation
Prove finite coarea and threshold optimality for binary cuts.
Chapter 3 · 3 sections
Image geometry and scale
Translation, region labels, directional information, and scale space.
Theorems 5Counterexamples 3Lenses 3Proof X-rays 3
Chapter outline →See this chapter on the web →Not yet indexed in the web
- 3.1
Registration, segmentation, and multiscale features
Analyze local registration identifiability, segmentation thresholds, and Gaussian scales.
- 3.2
Landmark registration by a similarity transform
Fit translation, rotation and scale to corresponding landmarks, and understand when the solution is identifiable.
- 3.3
Image resampling and aliasing
Follow an inverse warp to pixel values, and derive aliasing in a finite image pyramid.
Chapter 4 · 2 sections
Wavelets and approximation
Finite Haar analysis, image pyramids, and noise
Theorems 4 · Definitions 3Counterexamples 3Lenses 1Proof X-rays 2
Chapter outline →See this chapter on the web →Not yet indexed in the web
- 4.1
Finite Haar analysis, image pyramids, and noise
Finite Haar analysis, image pyramids, and noise
- 4.2
Wavelet denoising risk and threshold selection
Derive an unbiased fixed-threshold risk estimate and examine the effect of choosing a threshold from the same data.
Chapter 5 · 2 sections
Features and scale space
Structure tensors, scale-space features, and image pyramids
Theorems 4 · Definitions 8Counterexamples 2Lenses 2History 1Proof X-rays 2
Chapter outline →See this chapter on the web →Not yet indexed in the web
- 5.1
Structure tensors, scale-space features, and image pyramids
Structure tensors, scale-space features, and image pyramids
- 5.2
Descriptor matching and robust registration
Turn local appearance into tentative correspondences and test geometric agreement.
Chapter 6 · 1 section
Convergent restoration algorithms
A convergent projected-dual solver for TV restoration
Theorems 2 · Definitions 1Counterexamples 1Lenses 1Proof X-rays 1
Chapter outline →See this chapter on the web →Not yet indexed in the web
- 6.1
A convergent projected-dual solver for TV restoration
A convergent projected-dual solver for TV restoration
Chapter 7 · 2 sections
Motion and discrete segmentation
Optical flow, affine registration, and binary graph cuts
Theorems 4 · Definitions 4Counterexamples 2Lenses 2Proof X-rays 2
Chapter outline →See this chapter on the web →Not yet indexed in the web
- 7.1
Optical flow, affine registration, and binary graph cuts
Optical flow, affine registration, and binary graph cuts
- 7.2
Windowed optical flow: evidence, uncertainty, and ambiguity
Derive weighted local motion estimation and distinguish a unique estimate from reliable evidence.
Chapter 8 · 1 section
8. Binary morphology
Study erosion, dilation and boundary-aware compositions.
Theorems 2 · Definitions 1Counterexamples 1Proof X-rays 1
Chapter outline →See this chapter on the web →Not yet indexed in the web
- 8.1
Binary morphology and boundary conventions
Prove the dilation–erosion adjunction and analyze opening, closing and finite-image boundaries.
Chapter 9 · 2 sections
Continuum regularization
Fourier function spaces, minimizers, diffusion, and boundary models.
Theorems 4 · Definitions 4Counterexamples 2Lenses 2Proof X-rays 2
Chapter outline →See this chapter on the web →Not yet indexed in the web
- 9.1
Boundary conditions for image diffusion
Construct finite nonperiodic heat flows, prove their energy laws, and see how boundary choices affect mass and numerical stability.
- 9.2
Heat flow on an interval: boundary values and Fourier modes
Construct and compare continuum Neumann and Dirichlet heat flows using complete sine and cosine bases.
Chapter 10 · 1 section
Continuous region segmentation
Shape derivatives, means, and conditional curve evolution.
Theorems 2 · Definitions 4Counterexamples 1Lenses 1History 1Proof X-rays 1
Chapter outline →See this chapter on the web →Not yet indexed in the web
- 10.1
Region energies, moving curves, and level sets
Derive a continuum segmentation first variation and its conditional descent law, with explicit means, circle solutions, and singularity limits.
Chapter 11 · 1 section
Intensity segmentation
Partition a finite image using its intensity histogram.
Theorems 2 · Definitions 2Counterexamples 1Proof X-rays 1
Chapter outline →See this chapter on the web →Not yet indexed in the web
- 11.1
Global Histogram Thresholding and Otsu's Rule
A finite-image treatment of global thresholding. The note proves the variance decomposition, shows why minimizing within-class variance is equivalent to maximizing between-class variance, works out exact examples, and explains what the rule cannot see.