ICP3038 Assignment 3 – Computer Vision Page 1 of 4
Module: ICP3038 – Computer Vision
Department: School of Computer Science
Module credit: 20
Organiser: Dr Franck P. Vidal
Assignment weight: 15%
Assignment deadline: see Blackboard
Assessment 3 – OpenCV Portfolio
Description
In this assignment, you will demonstrate your ability to develop computer vision applications using OpenCV. It
is a portfolio in which you will show what you did in the labs and explain the corresponding concepts involved.
Contribution of this assessment
This assessment contributes to 15% of the overall module mark.
Tasks & Requirements
The assignment is based on the implementation of all the labs of the second semester. Your code should be able
to:
Lab 1 of semester 2: Linear and spatial filtering using OpenCV
1. Convert an RGB image into a Greyscale Image;
2. Convert an image in UCHAR into FLOAT32;
3. Display an image in the Log Scale;
4. Apply a mean Filter;
5. Apply a median Filter;
6. Apply a Gaussian Filter.
Lab 2 of semester 2: Pyramid blending
1. Compute Gaussian pyramids;
2. Compute Laplacian pyramids;
3. Blend images using pyramids.
Lab 3 of semester 2: Motion detection and tracking
1. Detect motion using your own background subtraction
• Load a video from a file;
• Save a video into a file;
• Choose an appropriate background;
• Calculate the absolute difference and threshold to compute the foreground mask;
• Clean a binary mask using mathematical morphology;
• Locate moving objects
2. Ignore ”objects” that are too small
3. Update the background
4. Estimate the motion of the objects
Lab 4 of semester 2: Stitch images together to create panoramas
1. Carry out a feature detection;
2. Perform a feature description;
3. Match features;
ICP3038 Assignment 3 – Computer Vision Page 2 of 4
4. Apply a geometric transformation;
5. Stitch images.
Note: The functionalities of the code have to be tested and validated. Evidences of testing
and validation have to be included in your report.
Task 1: Complete all the labs
You should have completed all the labs.
Task 2: Testing
The code has to be tested. Evidence of the results of the test will be assessed. If appropriate,
submit your output videos on Blackboard as additional material. Alternatively, you may wish to
upload them onto YouTube (this is your decision, I do not mind). See https://www.youtube.
com/playlist?list=PLR_5kWb9r72gcaJuaOUOWXH5ZX19HK8ey for videos from past students.
Task 3: Write a short report
Your final submissions should include the code and a report. For each of the labs mentioned
above, your report should describe your computer vision algorithms, what you have implemented,
what works, what does not, and how you tested it. Whenever possible, use pictures to illustrate
your text. For example, you can use a flow chart to present your algorithm, show the input
images as well as the output images. At the end of your report, provide a critical analysis
of your performance. Include your source code in the appendix. Note that your code should
include comments. You can structure your report as follows:
• Title page
• Introduction (small section describing the aim of the assignment and the content of the
report)
• One section for each lab, each section containing:
– A paragraph stating the aim of the lab (inc. description of the imaging pipeline if
applicable).
– What you have implemented, what works, what does not
– Command line arguments
– Input images
– Output images
– Source code snapshots (if applicable) to explain what you did
• Conclusion (inc. a personal statement with a critical analysis of your performance)
• The source code of all the labs as a single ZIP file
Submission procedure
Write your report in a Word document or in LATEX. You must submit:
1. a .doc or .pdf file of your short report in the following format: “username-ass3.pdf”, and
2. a .zip file of your source code in the following format: “username-ass3.zip”.
The assignment is due on: see Blackboard. Submissions must be made via Blackboard.
Please take into account upload times and internet connections when considering how much
time you have remaining.
ICP3038 Assignment 3 – Computer Vision Page 3 of 4
Assessment method
Your report and your code will be marked. In addition, you might be interviewed to demonstrate
your understanding of the material you submitted.
Note that care will be given to details. An appropriate coding standard should be used
consistently. Comments should be included in the code. The code should compile without
error. The code should compile without warning if possible.
Plagiarism and Unfair Practice
Plagiarised work will be given a mark of zero. Remember when you submit you agree to the
standard agreement:
This piece of work is a result of my own work except where it is a group assignment
for which approved collaboration has been granted. Material from the work of others
(from a book, a journal or the Web) used in this assignment has been acknowledged
and quotations and paraphrasing suitably indicated. I appreciate that to imply that
such work is mine, could lead to a nil mark, failing the module or being excluded from
the University. I also testify that no substantial part of this work has been previously
submitted for assessment.
Late Submission
Work submitted within one week of the stated deadline will be marked but the mark will be
capped at 40%. A mark of 0% will be awarded for any work submitted 1 week after the deadline.
Acceptable reasons for submitting work late include: serious personal illness with a doctor’s
certificate (a self-certified medical note should not be accepted); the death of a relative or
close friend; serious family problems such as divorce, separation and eviction. Examples of
unacceptable reasons for failing to submit work on time include: having exams; having other
work to do; not having access to a computer; having computer related problems; being on
holiday; not being able to find information about a subject.
> 80, exceptional use of OpenCV. Clear demonstrable understanding of computer vision techniques. Outstanding
development of the algorithm. Superb use of coding standards and comments. Overall an exceptional and
well-designed program that demonstrates the validity of proposed algorithm. The report is exceptional and provides
a comprehensive and clear critical analysis of the work performed. An exemplar solution that could be used to
demonstrate good practice of motion detection and OpenCV programming to colleagues.
> 70, good implementation that demonstrates a good understanding of OpenCV programming. The code works
effectively and the test programs are effective to check its validity. Most of the time the pipeline can handle noisy
data, changes in lighting condition, and it shows where motion occurs. The code is written using coding standards.
The code is well commented. The report is comprehensive and makes a good critical analysis of the work. Overall
a very good solution to this assignment. A well-structured report is provided (including a good critical analysis of
the work provided).
> 60, a good implementation that demonstrates a suitable understanding of problems and of OpenCV
programming. The code works effectively. Coding standards are relatively well applied and there are some
comments. Some limitations may exist in the work, however a good attempt made, and although there may be
some limitations with the program a comprehensive report (with an excellent and well critiqued section) is included.
> 50, appropriate demonstration of the challenge. Maybe full functionality is not provided, however the student
demonstrates that they understand the processes required to use OpenCV and understand some of the challenges
of computer vision. Comments are sparse and the use of coding standards is approximate. Even with these
limitations the report is well presented, and their achievements are well criticised and discussed. Limitations to
the work are clearly presented in the report and the student clearly understands what they have achieved and the
limitations thereof. Overall, maybe some flaws, but a reasonable submission.
> 40, threshold performance. Demonstrates some understanding of the problems in computer vision and can use
OpenCV. Some attempt has been made over creating the code. The code is not well commented and the coding
standards are not used appropriately. The report discusses some of the issues, and provides a basic critique of the
work submitted.
< 40, below threshold performance, with little demonstration of how motion could be detected. Little thought has
been made over this assessment, and understanding is confused.
ICP3038 Assignment 3 – Computer Vision Page 4 of 4
Feedback details
Description Timeframe
Formative
(On-going)
Verbal Feedback – Verbal feedback will be available by request. It is
suggested that you keep a written note of this feedback to aid in your
personal development.
Instant
Summative (Post
Assessment)
Written Feedback – Written feedback will be made available through
blackboard after an assignment is submitted. To access your written
feedback see the comments section of your assignment submission.
1-2 weeks

欢迎咨询51作业君