All lecture content sans the assignment(s) and important announcmenets will be hosted on the lectures Github Page which you can find it here.
Welcome to the lecture materials for use in B.Sc - Digital Image Processing where our focus will be on the topics of:
- Fundamentals on discrete mathematics,
- Convolution
- Discrete Fourier Transform
- Information Theory
- Display technologies and Cameras,
- Camera Types
- Lenses used in Industry
- Image processing techniques,
- Morphological Operations
- Historgram Operations
- An example of using ML in image recognition techniques.
The details of the lecture are given below.
| DESCRIPTION | VALUE |
|---|---|
| Program Name | B.Sc "Mechatronics Design and Innovation" |
| Module Name | Image Processing |
| Semester | 5 |
| Room | Lecture Room |
| Assessment(s) | Midterm Assignment (40 %) Group Assignment (60 %) |
| Lecturer | Daniel McGuiness |
| Software | Python |
| Hardware | - |
| SWS Total | 4 |
| Total Units | 60 |
| ECTS | 5 |
| Lecture Type | ILV |
There will be two (2) assignments for this course.
The grade breakdown is as follows:
| DEFINITION | GRADE (%) |
|---|---|
| Individual Assignment | 40 |
| Group Assignment | 60 |
| Sum | 100 |
An individual assignment will be given to you to work on. This assignment will consist of questions pertaining to concepts and image processing techniques.
The grade breakdown is as follows:
| DEFINITION | GRADE (%) |
|---|---|
| Report Style | 15 |
| Q1 - Blurring Filters | 15 |
| Q2 - Image Channel Analysis | 10 |
| Q3 - RNG Map Generation | 10 |
| Q4 - Image Cleaning | 10 |
| Q5 - Shape Recognition | 30 |
| Q6 - Image Quality Comparison | 10 |
| Sum | 100 |
NOTE: The assignment is individual and is not meant to be worked as a group. Once the code and the work is submitted it will be vetted against a software to determine if any collusion has occured.
The group assignment focuses on a student defined project which its presentation will be done in the last 3 sessions of the course. You are to come up with a group and a project within the first 3 weeks of the lecture otherwise one will be given to you.
The grade breakdown is as follows:
| DEFINITION | GRADE (%) |
|---|---|
| Report Style | 15 |
| Content | 55 |
| Q & A | 30 |
| Sum | 100 |
In report writing students must declare their contribution to the work and they will be asked regarding their field of work during the Q&A (i.e., if Student A has worked with blurring filter he may be asked on why a specific one is chosen and/or the concepts and maths behind the said filter).
NOTE: Students will be graded based on their contribution to the project and answers during the Q&A, therefore will be graded individually.
As it currently is, the lecture covers topic from vision technologies (i.e., camera, display) to methods in improving/analysing gathered images. The structure of the lecture is shown below.
| ORDER | TOPIC | DESCRIPTION | SESSION |
|---|---|---|---|
| 1 | Introduction | Discussion of the lecture structure and what will be covered | 1 |
| 2 | Mathematical Fundamentals | Convolution, sampling theorem and Fourier analysis | 1 |
| 3 | Perception | Colour spaces and industry standards (i.e., colour science) | 2 |
| 4 | Camera | Camera operation principles and lenses | 2 - 3 |
| 5 | Display | Display technologies and standards | 4 |
| 6 | Noise | Types of noise encountered and how to mode them | 4 - 5 |
| 7 | Histogram Operations | Analysis of histogram, both in grey and colour, along with masking and stretching | 6 |
| 8 | Morphological Operations | Morphological operators (i.e., dilation, gradient, …) | 7 |
| 9 | Blurring Filters | Types of blurring filters used for noise reduction and smoothing applications | 8 |
| 10 | Feature Analysis | Algorithms used to extract features from images | 9 |
| 11 | Edge Detection | Methods and alhorithms used in detecting edges for computer vision | 10 |
| 12 | Neural Networks for Image Processing | A Brief introduction to ANNs for use in image recognition | 11 - 12 |
| 13 | Group Assignment Presentations | Presentations of your group assingments and the following Q & A | 13 - 15 |
The Code supplement is a Github webpage dedicated to hosting all the relevant code used in the lecture as it is not feasible to fit all the content of the code to the slides and it is easier to share this way.
Visit the Code Supplement Website
The following materials are recommend reading for the coure but by no means are they mandatory.
| TITLE | AUTHOR | PUBLISHER |
|---|---|---|
| Fundamentals of Image Processing | Young, I. | Delft |
| Computer Vision: Algorithms and Applications | Szeliskti R. | Springer |
| Feature Extraction and Image Processing for Computer Vision | Nixon M., et. al | Academic Press |
| Digital Image Processing | Gonzalez, R. | Pearson |
–DTMc