AI PROJECT LOGBOOK

Resource for Students

(Adapted from “IBM EdTech Youth Challenge – Project Logbook” developed by IBM
in collaboration with Macquarie University, Australia and Australian Museum)

KEY PARTNERS

INDIA IMPLEMENTATION PARTNERS

GLOBAL PARTNERS

1
 AI Project Logbook

PROJECT NAME: OBJECT DETECTION

STUDENT NAME : KUSHAGRA AGARWAL

SCHOOL NAME: SETH ANANDRAM JAIPURIA SCHOOL

YEAR/CLASS: 2024-25 Class 12th

TEACHER NAME: Mr. Hrithik Gautam

TEACHER EMAIL: [email protected]

TEAM MEMBER NAMES AND GRADES:

1. A d i t y a U p a d h y a y

2. P r a k h a r P a n d e y

3. Kushagra Agarwal

4.

5.

6.

Note: Add more rows if there are more members in your team

2
1. Introduction
This document is your Project Logbook, and it will be where you record your
ideas, thoughts and answers as you work to solve a local problem using AI.

Make a copy of the document in your shared drive and work through it digitally
with your team. You can also print a copy of the document and submit a scanned
copy once you have completed the Project Logbook. Feel free to add pages and
any other supporting material to this document.

Refer to the AI Project Guide for more details about what to do at each step of your
project.

2. Team Roles
2.1 Who is in your team and what are their roles?

Role Role description Team Member Name

Researcher To research about the required modules , ADITYA UPADHYAY
frameworks and tools in the project
Ex. YoloV5,Pytorch

CODER Responsible for integrating the Kushagra Agarwal

frameworks and creating the dataset.

TESTER Responsible for the correct working and Prakhar Pandey

checking for refinements.

3
 2.2 Project plan

The following table is a guide for your project plan. You may use this or create
your own version using a spreadsheet which you can paste into this section. You
can expand the ‘Notes’ section to add reminders, things that you need to follow
up on, problems that need to be fixed urgently, etc.

Phase Task Planned Planned Planned Actual Actual Actual Who is Notes/Remarks
start end duration start end date duration responsible
date date (hours, date (hours,
minutes) minutes)
Preparing for Coursework, 01/08/24 29/09/24 26 hours 02/08/24 17/09/24 26 hours All team
the project readings members
Set up a 02/08/24 02/08/24 1 hour 02/08/24 02/08/24 1 hour All team
team members
folder
on a shared
drive
Defining the Background 03/08/24 04/08/24 2 hours 03/08/24 04/08/24 2 hours All team
problem reading members
Research 04/08/24 09/08/24 6:30 hours 03/08/24 04/08/24 6:30 hours Aditya Upadhyay
issues in our
community
Team 09/08/24 11/08/24 2:30 hours 09/08/24 11/08/24 2:30 hours All team
meeting to members
discuss
issues and
select an
issue for
the
project
Complete 12/08/24 17/08/24 5 days 12/08/24 17/08/24 5 days All team member
section 3
of the
Project
Logbook
Rate 17/08/24 17/08/24 30mins 17/08/24 17/08/24 30mins All team
yourselves members
Understanding Identify 18/08/24 18/08/24 1 hour 18/08/24 18/08/24 1 hour All team
the users users members
Meeting 20/08/24 20/08/24 6 hours 20/08/24 20/08/24 6 hours Prakhar Pandey
with users
to
observ
e them
Interview 20/08/24 20/08/24 6 hours 20/08/24 20/08/24 3 hours Aditya Upadhyay
with user (1)
Interview 20/08/24 20/08/24 6 hours 20/08/24 20/08/24 3 hours Kushagra
with user Agarwal
(2),
etc…
Complete 21/08/24 21/08/24 1 hour 21/08/24 21/08/24 hour 1 Hour All team
section 4 members
of the
Project
Logbook
Rate 21/08/24 21/08/24 30 min 21/08/24 21/08/24 30 min All team
yourselves members
Brainstorming Team 22/08/24 24/08/24 3 hour 22/08/24 24/08/24 3 hour All team
meeting members
to
generate
ideas for
a
solution
Complete 25/08/24 25/08/24 30 min 25/08/24 25/08/24 30 min All team
section 5 members
of the
Project
Logbook
Rate 25/08/24 25/08/24 5 min 25/08/24 25/08/24 5 min All team
yourselves members

4
Designing Team 27/08/24 30/08/24 5 hour 27/08/24 30/08/24 5 hour All team
your solution meeting members
to design
the
solution
Complete 31/08/24 31/08/24 30 min 31/08/24 31/08/24 30 min All team
section 6 of members
the logbook
Rate 31/08/24 31/08/24 5 min 31/08/24 31/08/24 5 min All team
yourselves members
Collecting and Team 1/09/24 4/09/24 1 hour 1/09/24 4/09/24 1 hour All team
preparing data meeting to members
discuss data
requirements
Collecting and Data 5/09/24 9/09/24 16 hour 5/09/24 9/09/24 16 hour All team
preparing data collection members
Prototyping

Data 10/09/24 11/09/24 5 hour 10/09/24 11/09/24 5 hour All team

preparatio members
n and
labelling
Complete 11/09/24 11/09/24 30 min 11/09/24 11/09/24 30 min All team
Section 6 members
of the
Project
Logbook
Team 12/09/24 15/09/24 4 hour 12/09/24 15/09/24 4 hour All team
meeting to members
plan
prototyping
phase
Prototyping Train your 15/09/24 15/09/24 3.4 hour 15/09/24 15/09/24 3.4 hour All team
Testing model with members
input dataset
Test your 16/09/24 19/09/24 8 hour 16/09/24 19/09/24 8 hour All team
model and members
keep
training
with more
data until
you think
your
model is
accurate
Write a 20/09/24 21/09/24 2 hour 20/09/24 21/09/24 2 hour All team
program to members
initiate
actions
based on
the result
of your
model
Complete 21/09/24 21/09/24 30 min 21/09/24 21/09/24 30 min All team
section 8 members
of the
Project
Logbook
Rate 21/09/24 21/09/24 5 min 21/09/24 21/09/24 5 min All team
yourselves members
Team 21/09/24 23/09/24 2 hour 21/09/24 23/09/24 2 hour All team
meeting members
to
discuss
testing
plan
Testing Invite 24/09/24 24/09/24 1 hour 24/09/24 24/09/24 1 hour All team
Creating the users to members
video test your
prototype
Conduct 24/09/24 24/09/24 1 hour 24/09/24 24/09/24 1 hour All team
testing members
with
users
Complete 24/09/24 24/09/24 30 min 24/09/24 24/09/24 30 min All team
section 9 members

5
 of the
Project
Logbook
Rate 24/09/24 24/09/24 5 min 24/09/24 24/09/24 5 min All team
yourselves members
Team 25/09/24 26/09/24 3 hour 25/09/24 26/09/24 3 hour All team
meeting members
to
discuss
video
creatio
n
Write your 26/09/24 26/09/24 30 min 26/09/24 26/09/24 30 min All team
script members
Film your 26/09/24 26/09/24 1:40 hour 26/09/24 26/09/24 1:40 hour All team
video members
Edit 26/09/24 26/09/24 1 hour 26/09/24 26/09/24 1 hour All team
your members
video
Completing Reflect 27/09/24 27/09/24 2 hour 27/09/24 27/09/24 2 hour All team
the logbook on the members
project
with
your
team

Complete 27/09/24 27/09/24 30 min 27/09/24 27/09/24 30 min All team

sections 10 members
and 11 of
the Project
Logbook
Review 28/09/24 28/09/24 5 hour 28/09/24 28/09/24 5 hour All team
your members
Project
logbook
and
video
Submission Submit 29/09/24 29/09/24 1 hour 29/09/24 29/09/24 1 hour All team
your members
entries on
the IBM

2.3 Communications plan

Will you meet face-to-face, online or a mixture of each to

communicate? How often will you come together to share your

progress?

Who will set up online documents and ensure that everyone is

contributing? What tools will you use for communication?

2.4 Team meeting minutes (create one for each meeting held)

Date of meeting:
Who attended: All team members
Who wasn’t able to attend: -
6
Purpose of meeting: To

discuss about the

7
 3. Problem Definition
3.1 List important local issues faced by your school or community

3.2 Many applications rely on object detection, such as security surveillance, autonomous
vehicles, and smart monitoring. However, implementing an efficient and real-time
detection system remains challenging.

Which issues matter to you and why?

 Enhances security and automation

 Can be used for real-time surveillance and monitoring
 Aids visually impaired individuals

3.3 Which issue will you focus on?

Our project will focus on real-time object detection using a camera to

identify multiple objects in a given environment efficiently.

8
3.4 Write your team’s problem statement in the format below.

How can we help automate object detection using a camera so that it can efficiently
recognize multiple objects in real-time for various applications?

Rate yourself 3

Problem Definition

1 point - A local problem is described

2 points - A local problem which has not been fully solved before is described.
3 points - A local problem which has not been fully solved before is explained in detail
with supporting research.

9
4. The Users
4.1 Who are the users and how are they affected by the problem?

 Security personnel for monitoring

 Individuals requiring smart assistance (e.g., visually
impaired)
 Researchers in AI & computer vision

4.2 What have you actually observed about the users and how the problem affects
them?

 Need for accurate real-time detection

 Importance of a fast and efficient model
 Requirement for scalability across multiple applications

10
4.3 Record your interview questions here as well as responses from users.

1. What challenges do you face in object

detection systems?
Response: Slow detection speed,
inaccurate results in poor lighting.
2. What features would you expect from
an AI detection system?
Response: High accuracy, real-time
detection, multi-object tracking.

11
4.4 Empathy Map

Map what the users say, think, do and feel about the problem in this table

What our users are saying What our users thinking

Detection is fast Need for real time analysis

Accurate Should be highly accurate
System lags at times AI should be more responsive
AI must recognize objects precisely

What our users are doing How our users feel

Use existing tools Frustrated by slow speed/lag

Manually review footage
Report issues to admin Concerned about security
Try different angles for better detection
Hopeful for improvements

12
4.5 What are the usual steps that users currently take related to the problem and
where are the difficulties?

Capturing video feeds manually - Requires human effort and constant monitoring.

Identifying objects visually - Leads to errors due to human limitations and distractions.

Recording and analyzing video manually - Time-consuming and inefficient.

Reviewing security footage - Tedious process prone to missing critical events.

Using traditional detection methods - Often results in false positives/negatives.

Dependence on multiple tools for detection - Leads to integration issues.

Difficulty in detecting objects in low-light conditions - Affects accuracy and

performance.

Handling occlusions and overlapping objects - Reduces detection precision.

Lack of real-time alerts - Slower response to critical situations.

Storage and processing of large video data - Requires significant computing resources.

4.6 Write your team’s problem statement in the format below.

All members

are experiencing issues with time management today

because of Mid-Term examinations.

Rate yourself 3
The Users

1 point - The user group is described but it is unclear how they are affected by the
problem.
2 points - Understanding of the user group is evidenced by completion of most of the
steps in this section.
3 points - Understanding of the user group is evidenced by completion of most of the
steps in this section and thorough investigation

13
5. Brainstorming
5.1 Ideas

How might you use the power of AI/machine learning to solve the users’ problem by
increasing their knowledge or improving their skills?

AI Idea #1 AI-powered object detection for security cameras.

AI Idea #2 AI-based real-time speech alert system for the visually impaired.

AI Idea #3 AI-driven automation for product sorting in industries.

AI Idea #4

AI Idea #5

14
 5.2 Priority Grid

Evaluate your five AI ideas based on value to users and ease of creation and
implementation.
Hig
h
High value to users, easy to create High value to users, hard to create

AI-powered object detection for security

AI-based real-time speech alert system
cameras
for the visually impaired
VALUE TO USERS

Low value to users, easy to create Low value to users, hard to create

AI-based object detection for non-security purposes

AI-driven automation for product sorting in
industries

Lo
w Easy Hard
EASE OF DEVELOPMENT

15
 5.3 Based on the priority grid, which AI solution is the best fit for your users and for
your team to create and implement?

Briefly summarize the idea for your solution in a few sentences and be sure to
identify the tool that you will use.

Based on the priority grid, the best AI solution is using YOLOv5 for real-time object
detection through AI-powered cameras to enhance security and
accessibility. This solution provides high value to users, is relatively feasible
for the team to develop, and leverages existing open-source datasets for faster
implementation. The model will process video feeds in real-time, detecting
objects with high accuracy, improving security surveillance, and assisting visually impaired
individuals with object identification.

Rate yourself 3

Brainstorming

1 point – A brainstorming session was conducted. A solution was selected.

2 points - A brainstorming session was conducted using creative and critical thinking. A
solution was selected with supporting arguments in this section
3 points - A brainstorming session was conducted using creative and critical thinking.
A compelling solution was selected with supporting arguments in this section.

16
6. Design
6.1 What are the steps that users will now do using your AI solution to address the
problem?

1. Camera captures real-time footage

2. YOLOv5 processes frames for object detection

3. Identified objects are labeled with bounding boxes

4. Detected objects are logged and recorded

5. Data is analyzed for further insights

Rate yourself 3

Design

1 point – The use of AI is a good fit for the solution.

2 points - The use of AI is a good fit for the solution and there is some documentation
about how it meets the needs of users
3 points - The use of AI is a good fit for the solution. The new user experience is clearly
documented showing how users will be better served than they are today.

17
7. Data

7.1 What data will you need to train your AI solution?

We need labeled image datasets containing

multiple objects to train the model effectively.

7.2 Where or how will you source your data?

Where will the Do you have Ethical

data come Who owns the permission to considerations
Data needed from? data? use the data?

Have
Open-source Public YES Data privacy
datasets maintained
(COCO,
ImageNet)

- - - -
Want/Need

- - - ---
Nice to have

Rate yourself 3

Data

1 point – Relevant data to train the AI model have been identified as well as how the
data will be sourced or collected.
2 points - Relevant data to train the AI model have been identified as well as how the
data will be sourced or collected. There is evidence that the dataset is balanced.
3 points - Relevant data to train the AI model have been identified as well as how the
data will be sourced or collected. There is evidence that the dataset is balanced, and
that safety and privacy have been considered.

18
8. Prototype

8.1 Which AI tool(s) will you use to build your prototype?

 YOLOv5 (for object detection)

 OpenCV (for video processing)

8.2 Which AI tool(s) will you use to build your solution?

 Detects objects and labels them

 Alerts users if certain objects appear (e.g., suspicious items
in surveillance)

8.3 What decisions or outputs will your tool generate and what further action needs to
be taken after a decision is made?
Decision/Output Further Action Needed
Object detected Display bounding box and label on-screen
Suspicious object detected Trigger an alert notification
Multiple objects detected Classify and log detected objects for review
Low confidence in detection Retrain model with additional data
Object occluded or partially Improve dataset diversity for better
visible accuracy
Enhance preprocessing techniques to adjust
Poor lighting affects detection
brightness

Rate yourself
3
Prototype
19
1 point – A concept for a prototype shows how the AI model will work.
2 points - A prototype for the solution has been created and trained.
3 points - A prototype for the solution has been created and successfully trained to
meet users’ requirements.
9. Testing
9.1 Who are the users who tested the prototype?

Friends, Relatives, 1 Visually Impaired Friend and some society people

9.2 List your observations of your users as they tested your solution.
What Needs Ideas for
What Works
Improvement Refinement
Improve dataset
Fast detection Accuracy in low light
quality
Multi-object Train on diverse
Handling occlusions
tracking data

20
9.3 Complete the user feedback grid

What works What needs to change

Real time object detection Detection accuracy in complex
Low latency backgrounds
Easy integration with security False positives in certain
systems conditions
User-friendly interface Limited detection in poor
lighting
Difficulty detecting small
objects

Questions? Ideas
Improve background filtering Enhance preprocessing
Adjust confidence thresholds techniques
Use infrared-enhanced datasets Fine-tune YOLOv5 parameters
Implement multi-scale detection Train model with low-light
images
Optimize feature extraction
layers

21
9.4 Refining the prototype: Based on user testing, what needs to be acted on now so
that the prototype can be used?

Based on user testing, the following actions need to be taken to

improve the prototype:
1. Enhance Background Filtering – Implement advanced
preprocessing techniques to reduce noise and improve
detection accuracy in complex backgrounds.
2. Optimize Confidence Thresholds – Adjust the detection
threshold to minimize false positives without missing actual
objects.

9.5 What improvements can be made later?

1. Improve Low-Light Detection – Train the model with infrared-

enhanced datasets and apply image enhancement techniques for
better detection in low-light environments.
2. Improve Small Object Detection – Use multi-scale detection
strategies and optimize feature extraction layers to enhance
detection of small objects.
3. Enhance Occlusion Handling – Train the model with diverse
datasets containing partially visible objects to improve occlusion
robustness.
4. Fine-Tune YOLOv5 Parameters – Experiment with different
anchor boxes, learning rates, and augmentation techniques to
improve overall model performance.
5. User Interface Enhancements – Improve UI responsiveness and
add real-time detection logs to enhance user experience.

Rate yourself 3

Testing

1 point – A concept for a prototype shows how it will be tested.

2 points - A prototype has been tested with users and improvements have been
identified to meet user requirements.
3 points - A prototype has been tested with a fair representation of users and all tasks
in this section have been completed.

22
10.Team collaboration
10.1 How did you actively work with others in your team and with stakeholders?

 Weekly meetings (online + in-person)

 Task assignments via shared docs
 Using GitHub for code collaboration

Rate yourself 3

Team collaboration

1 point – There is some evidence of team interactions among peers and stakeholders.
2 points - Team collaboration among peers and stakeholders is clearly documented in this
section.
3 points - Effective team collaboration and communication among peers and
stakeholders is clearly documented in this section.

23
11.Individual learning reflection
11.1. Team Reflections

A good way to identify what you have learned is to ask yourself what surprised
you during the project. List the things that surprised you and any other thoughts
you might have on issues in your local community.

Team member name: Aditya Upadhyay

I learned about the different types of modules and frameworks

that were used in the project and also got fascinated by the
yolov5 model

Team member name: Prakhar Pandey

I learned about the complexities of working with a model

and the skill of eliminating errors.

Team member name: Kushagra Agarwal

I learned integrating different modules into python and their

working and processing

24
Team member name:

Team member name:

Team member name:

Note: Add more boxes if there are more members in your team

Rate yourself 3

Individual Learning Reflection

1 point – Some team members present an account of their learning during the project.
2 points - Each team presents an account of their learning during the project.
3 points - Each team member presents a reflective and insightful account of their
learning during the project.

25
12.Video link

Enter the URL of your team video:

https://www.youtube.com/watch?

v=e6E-SC2sPPc

Enter the password (if any):

26
Appendix
Recommended Assessment Rubric (for Teachers)

LOGBOOK AND VIDEO CONTENT

Steps 3 points 2 points 1 point Points
Given
3
Problem A local problem which has A local problem which has A local problem is
definition not been fully solved before not been fully solved before described
is explained in detail with is described.
supporting research.

The Users Understanding of the user Understanding of the user The user group is 3
group is evidenced by group is evidenced by described but it is
completion of all of the completion of most of the unclear how they are
steps in Section 4 The Users steps in Section 4 The affected by the
and thorough investigation. Users. problem.

Brainstorming A brainstorming session was A brainstorming session A brainstorming 3

conducted using creative and was conducted using session was
critical thinking. A compelling creative and critical conducted. A solution
solution was selected with thinking. A solution was was selected.
supporting arguments from selected with supporting
Section 5 Brainstorming. arguments in Section 5
Brainstorming.
Design The use of AI is a good fit for The use of AI is a good fit The use of AI is a good 3
the solution. The new user for the solution and there fit for the solution.
experience is clearly is some documentation
documented showing how about how it meets the
users needs of users.
will be better served than
they are today.
Data Relevant data to train the AI Relevant data to train the AI Relevant data to train 3
model have been identified model have been identified the AI model have
as well as how the data will as well as how the data will been identified as well
be sourced or collected. be sourced or collected. as how the data will be
There is evidence that the There is evidence that the sourced or collected.
dataset is balanced, and dataset is balanced.
that safety and
privacy have been
considered.
Prototype A prototype for the solution A prototype for the A concept for a 3
has been created and solution has been created prototype shows how
successfully and trained. the AI model will work
trained to meet users’
requirements.
Testing A prototype has been A prototype has been A concept for a 3
tested with a fair tested with users and prototype shows how
representation of users and improvements have been it will be tested.
all tasks in Section 9 identified to meet
Testing have been completed. user requirements.
Team Effective team collaboration Team collaboration among There is some evidence 3
collaboration and communication among peers and stakeholders is of team interactions
peers and stakeholders is clearly documented in among peers and
clearly documented in Section 10 Team stakeholders.
Section 10 Team collaboration.
collaboration.
Individual Each team member Each team presents an Some team members 3
learning presents a reflective and account of their learning present an account of
insightful account of their during the project. their learning during the
learning during the project. project.

27
Total points

27
VIDEO PRESENTATION
Points
Given
Criteria 3–
excellent
2 – very
good
1 – satisfactory
3

Communicatio The video is well-paced and communicated, following

n a clear and logical sequence.

3
Demonstrations and/or visuals are used to illustrate
Illustrative
examples, where appropriate.

3
Accurat The video presents accurate science and technology
e and uses appropriate language.
languag
e
3
The video demonstrates passion from team members
Passion
about their chosen topic/idea.

3
Sound and
The video demonstrates good sound and image
image
quality quality.
3
The content is presented in the video within a 3-
Length
minute timeframe.

18
Total points

28