Type of Credit: Elective
Credit(s)
Number of Students
With the popularity of smartphones, where camera technology has advanced tremendously in recent years, everyone can take photos anywhere, anytime. According to the statistics, 350 million photos are uploaded on Facebook every day in 2018. All the images/videos presented to you have gone through the Image Signal Processing pipeline, which involves various image/video processing technologies, such as noise removal, image sharpening, gamma correction, image/video compression, image/video restoration, etc. We will also introduce image processing techniques based on deep learning.
能力項目說明
Students will learn about various image/video processing technologies in the course so as to understand more about imaging technologies. In addition, students will have hands-on experience in using image processing techniques for various problems.
| 教學週次Course Week | 彈性補充教學週次Flexible Supplemental Instruction Week | 彈性補充教學類別Flexible Supplemental Instruction Type |
|---|---|---|
Week 1
Subject： Introduction & syllabus
Covering topics: Introduction to digital image processing.
Reading: Chapter 1 in the textbook
Teaching/HW: Explain the syllabus and introduce image processing
Hours spent for preview and review: 1 hour each
--
Week 2
Subject： Background and math tools in digital image processing
Covering topics: Mathematical concepts
Reading: Chapter 2.6
Teaching/HW: Teach students related math concepts used in digital image processing
Hours spent for preview and review: 2 hours each
--
Week 3: Holiday
--
Week 4:
Subject： Human visual system, color models
Covering topics: Elements of visual perception, color models
Reading: Chapters 2, 6.1, 6.2
Teaching/HW: How the human visual system works? Illustrate different color models. Homework 1 will be released. Turn in HW within two weeks after it is released. No late submission is allowed.
Hours spent for preview and review: 2 hours each
--
Week 5
Subject： Processing of binary/gray images
Covering topics: Morphological Image Processing
Reading: Chapter 9
Teaching/HW: Illustrate morphological operators and their applications.
Hours spent for preview and review: 2 hours each
--
Week 6, 7
Subject： Contrast enhancement
Covering topics: Histogram equalization, specification, and stretching.
Reading: Chapter 3
Teaching/HW: Illustrate various techniques for contrast enhancement. Homework 2 will be released. Turn in HW within two weeks after it is released. No late submission is allowed.
Hours spent for preview and review: 2 hours each.
--
Week 9 Midterm
--
Week 10 & 11
Subject： Edge detection and sharpening
Covering topics: Zero-crossing, various edge-detection operators
Reading: Chapters 6.6, 10.2
Teaching/HW: Illustrate different edge detectors and how to sharpen images based on edges. Homework 3 will be released. Turn in HW within two weeks after it is released. No late submission is allowed.
Hours spent for preview and review: 2 hours each
--
Week 12 & 13
Subject： Noise removal
Covering topics: Noise models, noise reduction, image smoothing
Reading: Chapter 5
Teaching/HW: Introduce noise models, smoothing filters, and how to use them. Students need to turn in the final project proposal by the end of Week 12.
Hours spent for preview and review: 2 hours each
--
Week 14 校慶
Week 15
Subject： Image/video compression, Image restoration
Covering topics: Introduction to image and video compression
Reading: Chapters 5 & 8
Teaching/HW: Compression concepts and designs
Hours spent for preview and review: 2 hours each
--
Week 16
Teaching/HW: Introduce deep-learning-based image-processing techniques
Hours spent for preview and review: 2 hours each
Week 17
Subject: Prepare final report and presentation
Week18
Subject: Final presentation
1. Homework (30%) – two homework assignments
2. Midterm (30%)
3. Final Presentation (40%) – Choose one recently published paper (from IEEE/ACM journals or Top-tier Conferences) and study it. At the end of the semester, present the paper you chose. A bonus (up to 5%) will be given if experimental results are tested by running the code (obtained from the authors or implemented by yourselves) and/or a "better" method is proposed.
4. Class Participation (10%) - Quiz
Textbook: Digital Image Processing by Gonzalez and Woods. 4th edition, Pearson, 2017
References:
1. Digital Image Processing Using Matlab, 2nd edition, 2009
2. Introduction to Digital Image Processing, William K. Pratt, 1st edition, CRC Press, 2013
3. Computer Vision: Algorithms and Applications, Springer-Verlag, 2011