Course Syllabus

Course Information：

Year-Semester	Course Number	Course Name	Phase	Credits	Hours	Required/Elective	Instructor	Class	Student NO.	Withdraw	Note
Spring Semester Year 2024	331607	Computer Vision and Image Measurement	1	3.0	3	★	Yuan-Sen Yang	GCDPE	6	0	土木208

Course Plan：

Instructor	Yuan-Sen Yang　Office Hours
Email	ysyang@ntut.edu.tw
Last Updated	2024-05-23 07:54:02
Course Objective	Compared to the course last year, the course this year includes more materials about recent development of artificial intelligence (AI). The rapid development and population of digital imaging hardware and algorithms bring high potential of computer vision and image measurement on practical applications. These applications can include but not limited to measurement, structural health monitoring, and safety identification. This course introduces the theory of computer vision and image measurement, guides students to operate and implement structural experiments and measurement projects. Students are encouraged to bring up the concepts and theories, programming skills, experimental design and implementation, oral presentations, as well as experimental documentation and technical writing ability.
Course Schedule	1. Overview. Python installation. OpenCV installation. Tensorflow+Keras installation. Hands-on: Basic Python coding and rapid typing 2. Fundamental 2D single/multi-channel array of images. Image file I/O. Color conversion. Geometric transformation. Edge detection. Hands-on: a basic picture-in-picture generator. 3. General data types of images, matrices, and vectors: Row-wise format of matrix/vector. Channels of data. Memory operations in programming. Hands-on: an advanced picture-in-picture generator. 4. Basic vector and matrix operations. Matrix multiplication/inversion. Fourier transform. Normalization. Hands-on: Fourier transform GUI tool 5. Homogeneous coordinate transformation: Rotation. Stretch. Warping. Hands-on: Interactive mouse-control photo warping tool 6. Image tracking. Template match. Optical flow. Enhanced correlation. Hands-on: Interactive mouse-control photo warping tool 7. Camera calibration. Intrinsic and extrinsic parameters. Hands-on: Calibration of your mobile phone cameras. 8. Image warping and enhance correlation coefficient tracking 9. Stereo triangulation 10. Midterm 11. Fundamentals of deep learning, tensors, gradient. Hands-on: a nonlinear regression problem. 12. Convolutional based neural network (I) Hands-on: MNIST exercise 13. Convolutional neural network and MNIST exercise 14. Multilayer convolutional 2D and model adjustment 15. Transfer learning. Segmentation. Generative Adversarial Network (GAN) 16. Presentations of final project proposals 17. Current trends of computer vision and image measurement 18. Final project presentation.
Evaluation and grading policy	Midterm 25% +/- 5% Assignments 25% +/- 5% Final Presentation 25%+/- 5% Final Report 25%+/- 5% Failure Due to Absences (FDA): Absences of more than 6 hours may result in a failing grade.
Materials	All right reserved. Do not reproduce without permission. Foreign-language textbooks:Yes 1. Kaehler, A.; Bradski, G. (2016). Learning OpenCV 3 Computer Vision in C++ with the OpenCV Library, O'Reilly. ISBN: 978-1-491-93799-0. 2. Chollet, K. (2021). Deep Learning with Python, Second Edition. (https://livebook.manning.com/book/deep-learning-with-python/)
The access to curricular consultation	8:10-9:00 AM. Monday to Friday.
Expanding teaching and resources
The course corresponds to the SDGs
Does the course incorporate AI
Note	The lecture is regularly video recorded and is available at the NTUT iStudy (北科 i 學園). In case that the lecture has to be carried out online, it will be at https://meet.google.com/yiu-jwqa-ckm See the schedule updates about the 16th, 17th, and the 18th weeks in the Schedule above.