PDFs - Image Processing and Pattern Recognition Outline PDF

Preview

Summary

Image Processing and Pattern Recognition _ Outline
Image Processing and Pattern Recognition _ Outline...

Description

SUBJECT OUTLINE 31256 Image Processing and Pattern Recognition Course area

UTS: Information Technology

Delivery

Spring 2020; standard mode; City

Credit points 6cp Requisite(s)

48024 Applications Programming These requisites may not apply to students in certain courses. See access conditions.

Result type

Grade and marks

Attendance: Forms of attendance and mode of delivery in this subject have changed to enable social distancing and reduce the risks of spreading COVID-19 in our community.

Subject coordinator A/Prof. Stuart Perry School of Electrical and Data Engineering, Room CB11.08.217. Email: [email protected]. Phone: (+61 2) 9514 7605. The Subject Co-ordinator may be contacted by email or phone if you have matters of a personal nature to discuss, e.g., illness, study problems, or a request for an appointment outside the given consultation hours. All email must bear a meaningful description in the ‘Subject’ box at the top of the email, beginning with the Subject number: e.g. 31256 request for late submission due to illness, etc. Generally, questions regarding assessment and the subject should be raised in public forums such as lectures and tutorials. This ensures that all students will get the benefit of the information given. Emails that are considered better answered in public may not receive a private response.

Teaching staff Prof. Massimo Piccardi (lecturer, tutor) Email: [email protected]; Phone: (+61 2) 9514 7942. A/Prof. Stuart Perry (lecturer, tutor) Email: [email protected]; Phone: (+61 2) 9514 7605. Dr Firas Al-Doghman (tutor) [email protected] Ms Helia Farhood (tutor) [email protected]

Subject description Images and videos contain enormous amounts of information that can be extracted automatically by means of image processing and pattern recognition techniques. The extracted information is at the basis of many innovative applications such as video surveillance, diagnosis from medical images, automatic indexing and retrieval of multimedia data, human-computer interaction. This subject gives students the ability to understand the principles of image processing and pattern recognition and develop software for the automatic analysis and interpretation of images and videos. The goal of this subject is to teach skills used by professional engineers working at developing image processing and computer vision products, services and solutions. During the project students apply the knowledge they have gained to scope, solve, test and communicate a solution to a real-world image processing problem in a collaborative team-based environment. Examples include detection of people and objects in video surveillance, automated diagnosis from medical images, and detection and recognition of faces in imagery. 10/07/2020 (Spring 2020)

© University of Technology Sydney

Page 1 of 8

Subject learning objectives (SLOs) Upon successful completion of this subject students should be able to: 1. Use foundational techniques of image processing and analysis such as filtering, segmentation and local features to solve image processing problems of real world application 2. Build a statistical classifier and know how to use other classifiers 3. Apply image processing and pattern recognition techniques to detect objects and activities in images and video 4. Collaborate with team members to successfully complete a project 5. Utilise Matlab to develop scripts in these areas

Course intended learning outcomes (CILOs) This subject also contributes specifically to the development of the following Course Intended Learning Outcomes (CILOs): Socially Responsible: FEIT graduates identify, engage, interpret and analyse stakeholder needs and cultural perspectives, establish priorities and goals, and identify constraints, uncertainties and risks (social, ethical, cultural, legislative, environmental, economics etc.) to define the system requirements. (B.1) Design Oriented: FEIT graduates apply problem solving, design and decision-making methodologies to develop components, systems and processes to meet specified requirements. (C.1) Collaborative and Communicative: FEIT graduates work as an effective member or leader of diverse teams, communicating effectively and operating within cross-disciplinary and cross-cultural contexts in the workplace. (E.1)

Teaching and learning strategies Formal and informal, in-class and out-of-class engagement by students constitutes a large part of the learning activities that are covered in this subject. Students should prepare before every class by accessing the weekly preparation material in UTSOnline that will generally consist of short videos and guided questions on topics related to the subject matter presented in that class. In each session students will discuss the preparation material and then more in-depth content will be presented. Together with the lecture material, students will do exercises using MATLAB to practice their understanding of the concepts with verbal feedback from the tutors. These exercises can be done individually or in small groups. For the next 5 weeks of the subject, the students will be working on a group project. The students will group together in small teams and apply the knowledge they have gained in the course of the previous classes to solve a real world image processing problem. Students are expected to apply learnt content, problem solving and research skills to write a specification of the problem they are attempting to solve, implement the solution and present the solution to the class. To complete this task successfully, students will need to work effectively in a team as well as research the problem individually and in a group outside of the class and bring that knowledge to the sessions. A formative feedback quiz will be given in class during week 4 with feedback on correct answers and a worked solution given by the tutor to help students assess their understanding of the subject content. Feedback from the lecturer will also be given at various points during the group project. Firstly, written feedback will be given on the submitted design specification for the project, this feedback can help students with the second assessment task. Secondly, verbal feedback from both the lecturer and tutors during the project sessions to help students gauge how they are progressing. Additionally, there will be a point mid-way through the project where students will provide formative peer feedback on group contribution.

Content (topics) 1. Image and video representation 2. Image processing, segmentation and analysis 3. Matching in 2D 4. Feature detection 5. Statistical pattern recognition 10/07/2020 (Spring 2020)

© University of Technology Sydney

Page 2 of 8

6. Video analysis

Program Week/Session

Dates

Description

1

Week Beginning 27 Jul

Introduction to subject 31256. An overview of image processing and pattern recognition. Tutorial on Matlab and linear algebra. Notes: This subject does not have an orientation week, so prior to Week 1 students can check the introduction slides available on subject website on UTS Online to get an idea of what the course is about.

2

Week Beginning 3 Aug

Image and video formats. Image processing. Matlab tutorial.

3

Week Beginning 10 Aug

Image analysis. Matlab tutorial.

4

Week Beginning 17 Aug

Feature detection. Matlab tutorial. Notes: At the end of the class session a short formative quiz will be given to students. To allow them to gauge their understanding of the material, verbal feedback on correct answers to the quiz will be supplied by the lecturer.

5

Week Beginning 24 Aug

Probability and statistics fundamentals, part I Matlab tutorial.

6

Week Beginning 31 Aug

Probability and statistics fundamentals, part II Matlab tutorial.

7

Week Beginning 7 Sep

Classification. Matlab tutorial. Weka demonstration.

-

Week Beginning 14 Sep

10/07/2020 (Spring 2020)

Mid-Session STUVAC: No lectures or Project Sessions will be held this week.

© University of Technology Sydney

Page 3 of 8

8

Week Beginning 21 Sep

Introduction to the project. Project session.

9

Week Beginning 28 Sep

Project session. Notes: Due 02/10/2020, 11:59pm: Assignment 1 (Project Requirements and Specifications)

10

Week Beginning 05 Oct

Project session.

11

Week Beginning 12 Oct

Project session.

12

Week Beginning 19 Oct

Project Session Notes: Due 23/10/2020, 11:59pm: Assignment 2 (Project Implementation) Due 23/10/2020, 11:59pm: Assignment 3 (Project Presentation)

Assessment The assessment of this course will involve the students forming small teams and applying the knowledge they have gained in the course of the classes to solve a real world image processing problem. To complete this task successfully, students will need to work effectively in a team as well as research the problem individually and in a group outside of the classes and bring that knowledge to the classes. Assessment of Assessment Task 1 “Project Requirements and Specifications” will be shared by the entire group of students, however for Assessment Task 2 students will be required to clearly indicate their individual contributions to the project implementation and will be assessed based on how successful the project implementation as a whole is and how much of that success is due to the student’s individual contribution. For Assessment Task 3, students will present an aspect of the group project and will be graded individually based on their presentations. Assessment components due in this subject should be submitted via the subject's webpage on UTSOnline. There is also information on the assessment tasks in the subject introduction slides which can be found in the "Subject Documents" section of the subject's page on UTSOnline under the heading "Introduction to the Course". There is also information about the project assessment which can be found in the "Assignments" section of the subject's page on UTSOnline under the heading "Project Guidelines". Reassessment You may feel that you have received a mark that does not accurately reflect the quality of your work in the subject or the level of knowledge that you have demonstrated. If so, you should initially ask the person who marked the assessment task to re-evaluate it for you, and explain the reason for their decision. If you are still unhappy with the result, then you should discuss this with the subject coordinator. Assignment Extensions Where you feel that an extension is warranted, you should see the subject lecturer or coordinator a minimum of three days prior to the normal due date. Any request must be appropriately supported. Any requests for extensions after this time will be refused except in exceptional circumstances.

10/07/2020 (Spring 2020)

© University of Technology Sydney

Page 4 of 8

Assessment task 1: Project Requirements and Specifications Objective(s): This assessment task addresses the following subject learning objectives (SLOs): 1, 2 and 3 This assessment task contributes to the development of the following Course Intended Learning Outcomes (CILOs): B.1, C.1 and E.1 Type:

Essay

Groupwork: Group, group assessed Weight:

25%

Task:

This assignment will require each group of 5-6 students to produce a written document containing requirements and specifications for an image processing and pattern recognition project. The project topic should be granted approval by the lecturer. The specifications must cover: • analysis of the problem, including the availability of one or more suitable, public datasets; • selection of the techniques that the students expect to solve the problem. This will include various, plausible alternatives; • motivation for the approach based on the knowledge acquired during lectures and independent readings; • a performance evaluation methodology which will be used to assess the approach’s performance. This will require conducting experiments on the chosen dataset(s) with the various alternative techniques and collate their performance; • a list of references.

Length:

4,500 words

Due:

11.59pm Friday 2 October 2020 Submission via the Assignment menu on UTSOnline

Assessment task 2: Project Implementation Objective(s): This assessment task addresses the following subject learning objectives (SLOs): 1, 2, 3, 4 and 5 This assessment task contributes to the development of the following Course Intended Learning Outcomes (CILOs): C.1 and E.1 Type:

Project

Groupwork: Group, individually assessed Weight:

50%

10/07/2020 (Spring 2020)

© University of Technology Sydney

Page 5 of 8

Task:

This assignment will require each group of students to implement the project described in their Assignment 1. The project will be developed as a Matlab program. Whereas the project implementation submission will be shared by the entire group, each student will submit a short document detailing their contribution to the group and the project implementation submission.

Due:

11.59pm Friday 23 October 2020 Submission via the Assignment menu on UTSOnline

Assessment task 3: Project Presentation Objective(s): This assessment task addresses the following subject learning objectives (SLOs): 3 and 4 This assessment task contributes to the development of the following Course Intended Learning Outcomes (CILOs): E.1 Type:

Presentation

Groupwork: Group, individually assessed Weight:

25%

Task:

This assignment will require each group of students to present their project as a video presentation. The presentation will be based on PowerPoint slides describing the project and a demo of the program. Each member of the group will deliver a part of the presentation and the PowerPoint slides and one video for the entire team will be submitted by the students using UTS Online.

Due:

11.59pm Friday 23 October 2020 The presentations will be submitted to UTS Online as a video presentation. The video presentation should last no less than 15 minutes and no more than 20.

Moderation of marks Where assessment items are marked by more than one marker, moderation will occur in line with UTS policy.

Assessment feedback Consistent with UTS policy, students will receive feedback in a timely manner that assists them to understand the learning objectives achieved and how they could improve the quality of their work.

Minimum requirements In order to pass the subject, a student must achieve an overall mark of 50% or more.

Required texts There is no single textbook in this subject. The book that is most heavily referenced is: Linda G. Shapiro, George C. Stockman, Computer Vision, Prentice Hall.

References For the various topics covered by this subject, students can make reference to a number of excellent books in the broader areas of image processing, image analysis, computer vision, and pattern recognition: Christopher M. Bishop, Pattern Recognition and Machine Learning, Springer Rafael C. Gonzalez, Richard E. Woods, Digital Image Processing, 3/e, Prentice Hall Richard Hartley, Andrew Zisserman, Multiple View Geometry in Computer Vision, Cambridge University Press David A. Forsyth, Jean Ponce, Computer Vision: A Modern Approach, Prentice Hall

10/07/2020 (Spring 2020)

© University of Technology Sydney

Page 6 of 8

Other resources Online support is on the subject's website on UTSOnline: http://online.uts.edu.au/ (this subject runs at 'Level 2': the website makes available subject information, materials, gradebook and an unmoderated discussion board).

Graduate attribute development For a full list of the faculty's graduate attributes refer to the FEIT Graduate Attributes webpage. For the contribution of subjects taken in the Bachelor of Engineering (Honours) or Master of Professional Engineering to the Engineers Australia Stage 1 Competencies, see the faculty's Graduate Attributes and the Engineers Australia Stage 1 Competencies webpage.

Assessment: faculty procedures and advice Marking criteria Marking criteria for each assessment task will be available on the Learning Management System: UTS Online. Extensions When, due to extenuating circumstances, you are unable to submit or present an assessment task on time, please contact your subject coordinator before the assessment task is due to discuss an extension. Extensions may be granted up to a maximum of 5 days (120 hours). In all cases you should have extensions confirmed in writing. Special consideration If you believe your performance in an assessment item or exam has been adversely affected by circumstances beyond your control, such as a serious illness, loss or bereavement, hardship, trauma, or exceptional employment demands, you may be eligible to apply for Special Consideration. Late penalty Work submitted late without an approved extension is subject to a late penalty of 10 per cent of the total available marks deducted per calendar day that the assessment is overdue (e.g. if an assignment is out of 40 marks, and is submitted (up to) 24 hours after the deadline without an extension, the student will have four marks deducted from their awarded mark). Work submitted after five calendar days is not accepted and a mark of zero is awarded. For some assessment tasks a late penalty may not be appropriate – these are clearly indicated in the subject outline. Such assessments receive a mark of zero if not completed by/on the specified date. Examples include: a. weekly online tests or laboratory work worth a small proportion of the subject mark, or b. online quizzes where answers are released to students on completion, or c. professional assessment tasks, where the intention is to create an authentic assessment that has an absolute submission date, or d. take-home papers that are assessed during a defined time period, or e. pass/fail assessment tasks. Querying results If you believe an error may have been made in the calculation of your result in an assessment task or the final result for the subject, it is possible to query the result with the Subject Coordinator within five (5) working days of the date of release of the result.

Academic liaison officer Academic liaison officers (ALOs) are academic staff in each faculty who assist students experiencing difficulties in their studies due to: disability and/or an ongoing health condition; carer responsibilities (e.g. being a primary carer for small children or a family member with a disability); and pregnancy. ALOs are responsible for approving adjustments to assessment arrangements for students in these categories. Students who require adjustments due to disability and/or an ongoing health condition are requested to discuss their situation with an accessibility consultant at the Accessibility Service before speaking to the relevant ALO.

Statement about assessment procedures and advice This subject outline must be read in conjunction with the Coursework Assessments policy and procedures. 10/07/2020 (Spring 2020)

© University of Technology Sydney

Page 7 of 8

Statement on copyright Teaching materials and resources provided to you at UTS are protected by copyright. You are not perm...