Curricular Unit:			Code:
Computer Vision			1126VCOP
Year:	Level:	Course:	Credits:
2	Master	Computer Systems Engineering (Mobile Computation)	5 ects
Learning Period:		Language of Instruction:	Total Hours:
Winter Semester		Portuguese/English	65
Learning Outcomes of the Curricular Unit:
Upon successful completion of this course unit students should be able to (learning outcomes - LO): LO1-Understand the fundamentals of image and of visual information representation LO2-discuss the human visual system LO3-apply morphological operators in computer vision LO4-Apply Spatial Domain Filtering LO5-apply frequency domain filtering LO6-explain color spaces and color-based image segmentation LO7-apply discontinuity and contour detection methods LO8-apply camera calibration and obtain depth information LO9-Experiment Visual Object Segmentation LO10-Identify, test and apply software libraries, development tools and hardware to implement computer vision applications LO11-Use and combine tools, involving software and hardware, in the specification, development and testing of computer vision projects
Syllabus:
1. Introduction to Computer Vision 1.1. Intro to Computational Vision (VC) 1.2. VC Application Domains 2. Fundamentals of image 2.1. Introduction 2.2. Visual perception 2.3. Image formation 3. Morphological image processing 3.1. Introduction 3.2. Mathematical morphology 3.3. Morphological algorithms 4. Image filtering and enhancement 4.1. Filtering in the spatial domain 4.2. Filtering in the frequency domain 5. Color Image Processing 5.1. Introduction 5.2. Color spaces 5.3. Color-based targeting 6. Image segmentation 6.1. Intro do image segmentation 6.2. Corner detection 6.3. Region detection 7. Representation and description 7.1. Description and representation of images 7.2. Description and representation of regions 8. Cameras and stereo vision 8.1. Cameras 8.2. Epipolar geometry and depth 9. Object and image recognition 9.1. Introduction 9.2. Pattern Recognition
Demonstration of the Syllabus Coherence with the Curricular Unit's Objectives:
The syllabus presented are consistent with the learning objectives of the curricular unit since there is a large convergence between the table of contents and the knowledge that the student is supposed to acquire in each of the program topics. The fundamental concepts about computer vision, image fundamentals and image formation are presented in the introductory chapters, in the following chapters are presented various concepts, techniques and algorithms related to image processing in gray scale and color. Are still covered algorithms and segmentation techniques, image description and recognition of visual objects. The learning objectives are achieved by supplementing the theoretical concepts with concrete examples and exercises run in lab environment using appropriate software
Teaching Methodologies (Including Evaluation):
This Course Unit (UC) is classified as a Project and contains core competences that are not subject to examination. There are elements of continuous assessment whose weighted average is required to be positive, the Practical Score of Continuous Assessment (NPAC) Assessment results: a) Student achieves minimum goals (NPAC >= 9.5 values) and a positive Final Grade (NF1 >=9.5 values) in continuous assessment. Approves the UC with NF1 b) Student achieves minimum goals (NPAC >= 9.5 values) and (NF1 < 9.5 values). Can be assessed on examination. Exam assessment is independent of continuous assessment. UC Final Grade is NF2 c) Student does not achieve minimum goals (NPAC < 9.5 values). Does not pass the UC and will not be able to access the exam Expected assessment elements: 1. Test 1 2. Test 2 3. Practical project 4. Exam Continuous Assessment Model: NPAC = (3) NF1 = ((1) + (2) + 2*NPAC)/4, NPAC >= 9.5 Exam Assessment Model (NPAC >= 9.5): NF2 = (4)
Demonstration of the Coherence between the Teaching Methodologies and the Learning Outcomes:
The teaching/learning methodology applied in this curricular unit as well as its evaluation system is perfectly aligned with the objectives to be attained by the students at the end of the term. The theoretical concepts are presented, discussed, applied and evaluated in the context of lectures, which guarantees students a solid foundation to understand the challenges facing this area of knowledge. On the other hand, so that the study is not restricted to conceptual models, in the practical lessons are presented concrete case studies and implemented solutions for real problems using appropriate software. This combination guarantees training for students that allows them to meet the scientific goals, essential to a good understanding of the theme, as well as the ability to adapt to technological changes. The evaluation process consists of theoretical tests and practical work also guarantees a correct balance between the efforts dedicated to both components. The objective is to train professionals’ specialized in state-of-the-art techniques and tools but also ensure its ability to follow future developments. In this curriculum unit the major computer vision concepts and techniques are presented and evaluated in theoretical component. These concepts are then applied in the resolution of the worksheets and practical work in the context of practical classes
Reading:
[1] Linda Shapiro, George Stockman, "Computer Vision", Prentice Hall, 2001 [2] Rafael Gonzalez, Richard Woods, "Digital Image Processing (4rd Edition)", Prentice Hall, 2018 [3] Richard Szeliski, "Computer Vision: Algorithms and Applications", Springer, 2010 (http://szeliski.org/Book/) [4] Jain, Kasturi, Schunck, "Machine Vision", McGraw-Hill, 1995 [5] Kaehler, Bradski, "Learning OpenCV 3: Computer Vision in C++ with the OpenCV Library", O'Reilly Media, 2017 [6] Jan Erik Solem, “Programming Computer Vision with Python: Tools And Algorithms For Analyzing Images”, 1st Edition, O'Reilly Media, 2012 [7] Reinhard Klette, "Concise Computer Vision: An Introduction into Theory and Algorithms", 2014, Springer
Lecturer (* Responsible):
José Manuel Torres (jtorres@ufp.edu.pt)