SCHOOL OF ARCHITECTURE, COMPUTING, &
ENGINEERING
CN7023 Coursework Guidelines:
Include a Cover Page to the front of the assignment with your Student ID, Name and
Topic Title.
All pages to be numbered sequentially
Module code CN7023
Module title Artificial Intelligence & Machine Vision
Module leader Dr Shaheen Khatoon
Assignment tutor Lectures: Shaheen Khatoon, Azhar Mahmood, Mehdi
Karami
Lab Instructors: Mehdi Karami, Afroza RAHMAN,
Zulfiqar ALI, Aakriti Maharjan, Mohammed Noman,
Salam Soofiyan, Rohboam Amin
Assignment title AI & Machine Vision coursework
Assignment number 1/1
Weighting Individual Assignment 100%
Handout date W4
Submission date Wednesday August 27th, 2025, 6pm
Learning outcomes
assessed by this
assignment Learning Outcomes: 1-8
(see course
handbook)
Turnitin submission Yes
requirement
Additional information ASSESSMENT FEEDBACK - Feedback on your
assessment will be available in four working weeks from
the submission date. Please refer to the module pages
on UEL+ for assessment specific details.
CN7023 Assessment: Complete all the Tasks (100 Marks)
Read the following tasks carefully to complete your coursework:
Task 1: Data Science Skills (20 marks)
Complete MATLAB Online Courses to learn Data Science skills. Earn a certificate for
each course, acquire the following FIVE certificates to complete this part of the
assessment.
Note: Please append your certificates at the end of the coursework report to be
counted in the final marks
1
1. Course 1: MATLAB Onramp - Get started quickly with the basics of MATLAB. (4
marks)
2. Course 2: Machine Learning Onramp - Learn the basics of practical machine
learning methods for classification problems. (4 marks)
3. Course 3: Deep Learning Onramp - Get started quickly using deep learning
methods to perform image recognition. (4 marks)
4. Course 4: Image Processing Onramp - Learn the basics of practical image
processing techniques in MATLAB. (4 marks)
5. Course 5: Computer Vision Onramp - Learn the basics of computer vision by
applying a typical workflow—tracking-by-detection (4 marks)
Task 2 – Continuous Assessments (10 marks)
During the practical session, complete the given task and show them to your lab
instructor to get the marks. Make sure your lab instructor is recording your marks for
the graded labs.
Please speak to your lab instructor and seek feedback.
Task 3: Design, implement and report on neural network-based
techniques for Image processing applications. (70 marks)
Complete an image processing project applying a typical workflow of image
processing application such as image classification or object detection/segmentation.
The report should include the following sections, the marks for each section are
highlighted:
Introduction (5 Marks)
Objective of the coursework (Research questions(s) you are exploring) (2)
Identification of Real-World Problem where image processing can be applied and
potential impact (e.g., medical imaging, fruit classification, skin cancer detection
etc.). (2)
An overview of the report content (1)
Creative and innovative approaches to the selected problem (10 Marks)
Innovative and original approaches to designing and implementing deep learning-
based techniques for image dataset. (4)
Clear description of the methods or strategies proposed (e.g., choice of neural
network architecture, data preprocessing techniques) (4).
Justification for why the chosen approach is suitable for the specific dataset and
problem (2).
Simulations (25 Marks)
Provide a description of the dataset (for example dataset source, dataset size,
number of classes, number of images in each class etc.) including sample images
(5)
How did you encode the dataset so that you could use the images as input to the
neural network? (include discussion of any preprocessing steps and image encoding
techniques used. (5)
Explain the network architecture that you used, how you trained, validated and tested
the network, explain the learning algorithm used. (15)
2
Results Obtained (15 marks)
Provide appropriate explanation with justifications of your results in the three
different ways:
1. Clear reporting of the test set accuracy as a percentage (%). E.g., accuracy for the
test set, i.e. the test set achieved 80% accuracy. (5)
2. Include an accuracy curve figure for the training, validation and testing results.
The x-axis will represent the number of epochs and the y-axis will represent the
percentage accuracy. (5)
3. Include a confusion matrix figure with explanation as a visual representation of the
accuracy you achieve on the test set. (5)
Critical Analysis of Results (10 Marks)
Provide information on how you have achieved the results, and how the results can
be improved by considering alternatives approaches i.e. by making changes to the
simulations. (5)
Provide detailed analysis and discussion of the results you achieved by considering
factors affecting performance and potential for further optimization (5).
Conclusions (5 Marks)
Restate the research problem addressed in the report and summarize your overall
arguments or findings. (3)
Suggest the key takeaways from your report. (2)
NOTE: You can choose to complete the coursework using any one of the following
approaches:
1. Mixture of image processing with ANN, Deep learning and computer vision (with
MATLAB or Python)
2. Image processing with ANN (with MATLAB or Python)
3. Image processing with Deep Learning only (with MATLAB or Python)
4. Image processing with Computer vision only (with MATLAB or Python)
Task 1 & Task 3 Submission instructions
1. Upload completed report to Turnitin before deadline.
2. Append MATLAB certificates in PDF form at the end of the report and create a single
PDF
3. Name your report with file name [Your Student Id-Name].
4. The Turnitin link can be found on the module’s Moodle site.
Plagiarism Policy
50% and Above Similarity Index: Assignments with 50% or more similarity will
be considered as a straight fail. Students will receive a zero grade for the
assignment and may face disciplinary actions as per the university's academic
integrity policies.
20% to 49% Similarity Index: Penalized based on the severity of the overlap.
Students may be asked to revise and resubmit the assignment with a deduction
of marks (20% marks deduction).
Below 20% Similarity Index: Minor overlaps with properly cited references are
permissible.
3
Additional Resources:
Datasets
You can choose any image datasets for your coursework. The dataset you use
should have reasonable number of classes according to the selected application and
reasonable number of images appropriate for training DL network.
You can use Kaggle to look for a dataset of your interest. Go to Kaggle site, select
dataset and filter using computer vision category.
https://www.kaggle.com/datasets?tags=13207-Computer+Vision
Some examples of the dataset are the following, you can choose any one of them or
choose from any other source according to you interest.
1. Plant disease recognition dataset, referring to three classes Healthy",
"Powdery", "Rust",
https://www.kaggle.com/datasets/rashikrahmanpritom/plant-disease-
recognition-dataset
2. Chest CT-Scan images Dataset,
https://www.kaggle.com/datasets/mohamedhanyyy/chest-ctscan-images
3. The SLICE-3D dataset - skin lesion image crops extracted from 3D TBP for
skin cancer detection: https://www.kaggle.com/competitions/isic-2024-
challenge/data
4. PlantVillage Dataset: https://www.kaggle.com/datasets/emmarex/plantdisease
5. Animals-10 - Animal pictures of 10 different categories taken from google
images, https://www.kaggle.com/alessiocorrado99/animals10
6. Natural Images - A compiled dataset of 6899 images from 8 distinct classes,
https://www.kaggle.com/prasunroy/natural-images#motorbike_0000.jpg
7. Fruit Classification - 22495 Images of Fruit!,
https://www.kaggle.com/datasets/sshikamaru/fruit-recognition
8. More
https://www.kaggle.com/datasets?tags=13207-Computer+Vision
MATLAB Example Data Sets
MATLAB® has many data sets spanning a variety of file formats and sizes. Use any
of the Image Data datasets after assuring appropriate size and classes.
https://uk.mathworks.com/help/matlab/import_export/matlab-example-data-sets.html
UCI Data Repository:
You can also explore UCI dataset repository and select image dataset which
interests you.
some examples of the dataset are following:
1. NASA Flood Extent Detection, 50000 images various flood events,
https://archive.ics.uci.edu/dataset/770/nasa+flood+extent+detection
2. More
https://archive.ics.uci.edu/datasets?
skip=0&take=10&sort=desc&orderBy=NumHits&search=&Types=Image
4
Learning Resources:
MATLAB has great resources to learn how to implement an end-to-end deep
learning/image processing project. Here are few examples of videos/webinars and
projects. Please locate appropriate to learn more:
Deep Learning Toolbox- Videos
https://uk.mathworks.com/support/search.html?fq%5B
%5D=asset_type_name:video&fq%5B%5D=category:deeplearning/
index&page=1&s_tid=CRUX_topnav
Get Started with Deep Network Designer
Get Started with Deep Network Designer
https://uk.mathworks.com/help/deeplearning/gs/get-started-with-deep-
network-designer.html?searchHighlight=deep%20network
%20designers&s_tid=srchtitle_support_results_4_deep%2520network
%2520designers
Multilabel Image Classification Using Deep Learning
https://uk.mathworks.com/help/deeplearning/ug/multilabel-image-
classification-using-deep-learning.html?searchHighlight=multilabel%20image
%20classification%20using%20deep
%20learning&s_tid=srchtitle_support_results_1_multilabel%2520image
%2520classification%2520using%2520deep%2520learning
Object Detection Using YOLO v2 Deep Learning
https://uk.mathworks.com/help/deeplearning/ug/object-detection-using-yolo-
v2.html
Pedestrian and Bicyclist Classification Using Deep Learning
https://uk.mathworks.com/help/radar/ug/pedestrian-and-bicyclist-
classification-using-deep-learning.html
Classify Tumors in Multiresolution Blocked Images
https://uk.mathworks.com/help/deeplearning/ug/classify-tumors-in-
multiresolution-out-of-memory-images.html
More
https://uk.mathworks.com/help/deeplearning/examples.html?
category=applications&s_tid=CRUX_topnav
