MATH3360 - Mathematical Imaging - 2023/24
Announcement
- Assignment 5 has been posted and updated on Nov 21, and the deadline is postponed to Nov 29 before 1159PM. Please submit your homework through the Blackboard system.
- Assignment 4 has been posted. It will be due on November 10 before 1159PM. Please submit your homework through the Blackboard system.
- Midterm Examination will be held on October 26 (Thursday) from 12:30pm-2:15pm during class (combining tutorial and lecture). The venue is Wu Ho Man Yuen Bldg 408 (same classroom as the tutorial and lecture). The coverage of the exam will be up to p.5 of Lecture Class Note 11.
- Assignment 3 has been posted. It will be due on October 20 before 1159PM. Please submit your homework through the Blackboard system.
- Assignment 2 has been posted. It will be due on October 6 before 1159PM. Please submit your homework through the Blackboard system.
Submission of homework assignments
- Log onto https://blackboard.cuhk.edu.hk/ and click on our course 2023R1 Mathematical Imaging (MATH3360). Click on "course contents" and click on "Assignment X (Due...)". Follow the instructions therein to upload your solution. An illustration can be downloaded here.
- Please scan your written solution into a single pdf file and save it with the name like: YourStudentID_HW1.pdf. There are several useful apps for you to take a picture of your solution and scan your document (such as CamScanner HD and Microsoft Lens).
- Assignment 1 has been posted. It will be due on September 22 before 1159PM. Please submit it via Blackboard.
- There will be no tutorial in the first week.
General Information
Lecturer
-
Prof. Ronald Lok Ming LUI
- Office: LSB 207
- Tel: 3943-7975
- Email:
Teaching Assistant
-
Lai Ka Ho
- Office: LSB 222A
- Tel: 3943 3575
- Email:
-
Lin Chenran
- Office: LSB 222A
- Tel: 3943 3575
- Email:
Time and Venue
- Lecture: Tu 4:30PM - 6:15PM (Lady Shaw Bldg LT3); Th 1:30PM - 2:15PM Wu Ho Man Yuen Bldg 408)
- Tutorial: Th 12:30PM - 1:15PM (Wu Ho Man Yuen Bldg 408)
Course Description
This course gives an introducion on mathematical models and techniques for various image processing tasks. A wide array of topics will be covered, including image restoration (denoising, deblurring), image segmentation, image compression and so on. Students will become familiar with essential mathematical techniques for imaging tasks, such as image processing in the spatial domain (using gradient, Laplacian, convolution) and frequency domain (using discrete Fourier transform).
Our goal of this course is to help students appreciate the importance of mathematics in imaging sciences. Students will have a chance to learn how existing image processing techniques are built based on mathematical theories. Upon successful completion of the course, interested students are also welcome to approach the lecturer to ask for opportunities to work on some research projects related to mathematical image processing.
Lecture Notes
- Chapter 1: Basic concepts in Digital Image Processing
- Chapter 2: Image decomposition
- Chapter 3: Image Enhancement in the Frequency Domain (by modifying DFT)
- Chapter 4: Image Enhancement in the Spatial Domain
Class Notes
- Lecture 1 (Class note): Image transformation and point spread function
- Lecture 1 (Powerpoint): Basic background about image processing
- Lecture 2 (Class note): Separable image transformation & Convolution
- Lecture 3 (Class note): More about convolution
- Lecture 3 (Powerpoint): Examples of convolution on real images
- Lecture 4: (Class note): Similarities between images, transformation matrix
- Lecture 5: (Class note): Image decomposition and SVD
- Lecture 6: (Class note): Error analysis of SVD decomposition
- Lecture 6: (Powerpoint): Example of image compression by SVD
- Lecture 7: (Class note): Discrete Haar transform, intro to discrete Fourier transform
- Lecture 7: (Powerpoint): Examples of image compression by Haar wavelet transform
- Lecture 8: (Class note): Interesting properties of discrete Fourier transform for image processing
- Lecture 9: (Class note): Image decomposition using DFT, mathematics of JPEG
- Lecture 9: (Powerpoint): Examples of image decomposition using DFT and JPEG compression
- Lecture 10: (Class note): DFT of convolution & rotated/translated images
- Lecture 11: (Class note): Low pass filtering using DFT
- Lecture 11: (Powerpoint): Examples of low pass filtering
- Lecture 12: (Class note): High pass filtering using DFT, mathematics of image blur
- Lecture 12: (Powerpoint): Examples of high pass filtering
- Lecture 13: (Class note): Image deblurring (1)
- Lecture 13: (Powerpoint): Examples of image deblurring by modifying DFT
- Lecture 14: (Class note): Image deblurring (2): Wiener's filtering
- Lecture 14: (Powerpoint): Examples of Wiener's filtering
- Lecture 15: (Class note): Image deblurring (3): Constrained least square filtering (1)
- Lecture 15: (Powerpoint): Examples of Constrained Least Square filtering
- Lecture 16: (Class note): Constrained least square filtering (2), image sharpening, linear filtering in the spatial domain
- Lecture 16: (Powerpoint): Examples of image sharpening, linear filtering
- Lecture 17: (Class note): Non-linear filtering for image denoising
- Lecture 17: (Powerpoint): Examples of nonlinear filtering
- Lecture 18: (Class note): Non-local mean filtering, Anisotropic diffusion for image denoising
- Lecture 18: (Powerpoint): Examples of non-local mean filtering, Anisotropic difussion for image denoising
- Lecture 19: (Class note): Image demonising by considering PDEs
- Lecture 19: (Powerpoint): Examples of image denoising by considering PDE
- Lecture 20: (Class note): Image processing by minimization
- Lecture 21: (Class note): Image processing by variational approach
- Lecture 22: (Class note): TV (ROF) image denoising model
- Lecture 22: (Powerpoint): Examples of TV (ROF) image denoising
- Lecture 23: (Class note): Active contour model for image segmentation
- Lecture 23: (Powerpoint): Image segmentation by active contour (snake) model
- Lecture 24: (Class note): Chan-Vese Segmentation model
- Lecture 24: (Powerpoint): Examples of Chan-Vese Segmentation
Tutorial Notes
- Tutorial 1
- Tutorial 1 code
- Tutorial 2
- Tutorial 2 code
- Tutorial 3
- Tutorial 3 code
- Tutorial 4
- Tutorial 5
- Tutorial 5 code
- Tutorial 6
- Tutorial 7
- Tutorial 8
- Tutorial 9
- Tutorial 10
Assignments
- Assignment 1 (Due on September 22 before 1159PM)
- Assignment 2 (Due on October 6 before 1159PM)
- Assignment 3 (Due on October 20 before 1159PM) (revised 2)
- Assignment 3 code
- Midterm practice (Revised on 25/10)
- Assignment 4 (Due on Noverber 10 before 1159PM)(Revised on 31/10)
- Assignment 5 (Due on Noverber 29 before 1159PM) (Updated on Nov 21)
- Final practice
Quizzes and Exams
Solutions
- Solution for assignment 1
- Solution for assignment 2
- Code for assignment 2
- Solution for midterm practice (Revised on 25/10)
- Solution for assignment 3
- Code for assignment 3
- Solution for assignment 4 (Revised on 28/11)
- Solution for final practice
- Solution for assignment 5
Assessment Scheme
| Homework | 15% | |
| Midterm (October 26, 12:30pm-2:15pm during class) | 35% | |
| Final exam (TBA) | 50% |
Honesty in Academic Work
The Chinese University of Hong Kong places very high importance on honesty in academic work submitted by students, and adopts a policy of zero tolerance on cheating and plagiarism. Any related offence will lead to disciplinary action including termination of studies at the University. Although cases of cheating or plagiarism are rare at the University, everyone should make himself / herself familiar with the content of the following website:
http://www.cuhk.edu.hk/policy/academichonesty/and thereby help avoid any practice that would not be acceptable.
Assessment Policy Last updated: December 02, 2023 20:31:47