Professional Documents
Culture Documents
Corrected
Corrected
PROJECT REPORT
Submitted to the University of Calicut in partial fulfilment of requirement
for the award of degree in
BACHELOR OF COMMERCE (FINANCE)
SUBMITTED BY
AKSHAYA R
Register number
ZNAVBCM002
Under the guidance of
Mrs. SARASWATHY R
(Assistant Professor, Department of commerce)
CERTIFICATE
Mrs. BINDU G
HOD-Department of commerce
Govt. Arts and Science College, Tholanur
Signature of the HOD
External Examiner
DECLARATION
We must appreciate the government and its Motor Vehicle department for
introducing the same.
1
1.2 STATEMENT OF THE PROBLEM
The AI camera in transportation enhances safety and efficiency by leveraging
machine learning algorithms to analyse traffic patterns, detect anomalies, and
facilitate smart traffic management. This technology can optimize traffic flow,
monitor road conditions, and contribute to a more secure and streamlined
transportation system. This study would stand as a sincere attempt to evaluate a
customer’s attitude towards AI camera. Special reference is taken from peoples
of Palakkad district.
Day by day accidents and violation of traffic rules are happening and are
increasing, so after installation of these AI cameras the people may become
vigilante, will follow the traffic rules and also, they will drive with care in order
to avoid the fines that the MVD may charge if they violate the rules, this will
happen when the AI camera captures it. Another thing is that the no. of vehicles
is increasing every day, but the length of the road is not so, to some extent we can
control by using this device that is by using AI cameras.
2
1.5 SIGNIFICANCE OF STUDY
AI cameras manage traffic, ensure pedestrian safety, and monitor public spaces
efficiently. Their predictive capabilities also enable proactive maintenance,
reducing downtime and costs. This study brings out the awareness and attitude
among people towards AI cameras. This would help to gain better knowledge
about AI cameras in transportation.
The research design of the study is descriptive in nature. Research design is the
framework about the method and technique to be used. Research design is a plan
that specifies the source and type of information relevant to the research problem.
In this study descriptive technique is used.
3
Secondary data: Secondary data is the data which is already available in public
records. This is the data by past researchers. Secondary data are those which have
been collected by some other person for his purpose and published. Secondary
data is collected from several sources like websites, research reports, and books.
Data collected from respondents was analysed through percentage analysis. The
interpretation of data was done with the help of tables and diagrams.
A sample design is a definite plan for obtaining a sample from a given population.
It refers to the technique or the procedure the researcher would adopt in selecting
items for the sample. The study is mainly based on convenience sampling
method.
4
1.7 LIMITATIONS OF THE STUDY
2. Since this initiative has just started, there will be fewer people who have
received penalties, which will make it more difficult for us to get information
about those individuals.
3. Some AI cameras had also made fines by sharing false details about the vehicle
no. and the names of those registered vehicles. It also has some limitations.
5
CHAPTER 2
REVIEW OF LITERATURE
REVIEW OF LITERATURE
Kim J's (2021) study titled ‘AI – based surveillance for pedestrian safety in urban
transportation’ published in "Transportation Research Records," focuses on the
application of AI-based surveillance for pedestrian safety within urban
transportation systems. The literature review likely begins by examining the
historical context of pedestrian safety measures, discussing traditional
surveillance methods, and the gaps that AI technologies aim to fill.
Garcia M's (2021) study titled ‘Deep learning for vehicle recognition in
transportation’ published in the "Transportation Research" journal, focus likely
on employing deep learning techniques for vehicle recognition within the
transportation domain. The literature review may commence by examining the
historical landscape of vehicle recognition methods, providing insights into the
evolution of traditional approaches and the gaps that deep learning aims to
address. Garcia M's work might delve into the fundamental principles of deep
learning, elucidating how neural networks and convolutional architectures have
been applied to enhance the accuracy and efficiency of vehicle recognition
systems.
Patel A's (2020) study titled ‘Real time monitoring with AI cameras for public
transportation ‘published in the "Journal of Urban Mobility," focus lies on real-
time monitoring through AI cameras specifically tailored for public transportation
systems. Patel A likely investigates how artificial intelligence technologies,
particularly AI cameras, can play a pivotal role in enhancing the efficiency and
safety of public transit operations. The literature review might start by examining
the historical evolution of public transportation monitoring methods and the
challenges associated with traditional surveillance systems.
6
Park Y's (2020) study titled ‘Integration of AI cameras with autonomous
vehicles for safe transportation’ published in the "Journal of Intelligent
Transportation Systems," focus centres on the integration of AI cameras with
autonomous vehicles to enhance safety in transportation. The literature review is
likely to commence by exploring the landscape of autonomous vehicle
technology, highlighting advancements and challenges in achieving safe and
efficient autonomous transportation systems. Park Y may delve into the specific
role of AI cameras, elucidating how these technologies contribute to perception,
decision-making, and overall situational awareness for autonomous vehicles.
Hernandez C's (2020) study titled ‘Enhancing Road safety through AI – enabled
Traffic surveillance’ published in "Transportation Research Part F: Traffic
Psychology and Behaviour" focuses on the augmentation of road safety through
AI-enabled traffic surveillance. The literature review may commence by
examining historical approaches to road safety and the evolution of surveillance
methods. Hernandez C's work likely delves into the specific AI technologies
employed for traffic surveillance, emphasizing advancements in real-time
monitoring, anomaly detection, and predictive analytics.
7
Gupta R's (2019) study titled ‘AI-powered surveillance for urban mobility’
published in the "Urban Transportation Technology" journal, focuses on the
integration of AI-powered surveillance for urban mobility. The literature review
likely begins by exploring the historical context of urban mobility challenges and
the evolution of surveillance methods. Gupta R's work may delve into the specific
AI algorithms employed, emphasizing advancements in computer vision, deep
learning, and real-time data analytics for enhancing surveillance capabilities.
Zhang X’s (2019) study titled ‘Traffic flow optimization using AI camera’s: A
case study’ published in the “Journal of Intelligent Transportation Engineering,”
focuses on traffic flow optimization through the utilization of AI cameras, with a
specific emphasis on a case study. The literature review likely begins by
exploring the historical context of traffic flow optimization, highlighting
traditional methodologies and the emergence of AI-based solutions. Zhang X’s
work may delve into the specifics of AI algorithms employed for real-time traffic
monitoring, adaptive signal control, and dynamic route planning, showcasing the
practical implications of these technologies in the presented case study.
Kim S's (2019) study titled ‘AI camera-based Anomaly detection in public
transportation’ published in the "International Journal of Transport Security,
“focuses on AI camera-based anomaly detection in public transportation. The
literature review begins by exploring the historical context of anomaly detection
methods in public transportation and the evolution of surveillance technologies.
Kim S’s work may delve into the specific AI algorithms employed, emphasizing
advancements in computer vision and machine learning for real-time anomaly
detection.
8
Wu H's (2019) study titled ‘Real time Traffic monitoring using Deep learning
Techniques’ published in the "Journal of Intelligent Transportation Systems," the
research focuses on real-time traffic monitoring utilizing deep learning
techniques. The literature review likely begins by exploring the historical
evolution of traffic monitoring methods, emphasizing the transition to deep
learning approaches. Wu H's work may delve into the specifics of deep learning
algorithms employed, highlighting advancements in convolutional neural
networks or recurrent neural networks for accurate and efficient real-time traffic
analysis.
Chen Y's (2018) study titled ‘smart cities: Integration of AI cameras for
intelligent transportation management’ published in the "Journal of Urban
Technology," focuses on the integration of AI cameras for intelligent
transportation management within the context of smart cities. The literature
review may commence by providing an overview of the concept of smart cities,
highlighting the imperative of leveraging AI cameras to enhance the efficiency
of transportation systems. Chen Y's work likely explores key applications, such
as traffic monitoring, congestion management, and adaptive traffic signal control,
demonstrating how AI-driven technologies contribute to more intelligent and
responsive urban transportation networks.
Smith J's (2018) study titled ‘Enhancing transportation safety with AI camera’s
‘published in the "Transportation Technology Review," delves into the realm of
enhancing transportation safety through the utilization of AI cameras. The study
likely explores the multifaceted applications of artificial intelligence in bolstering
safety measures within the transportation sector. Smith J's work may encompass
the integration of advanced camera systems for real-time monitoring, analysis,
and decision-making to mitigate potential risks and enhance overall safety
protocols.
9
Li H's (2018) study titled ‘Smart Traffic management using AI camera’s – A
comprehensive Review’ published in "Transportation Research Part B:
Methodological," delves into smart traffic management through the
comprehensive application of AI cameras. The literature review likely begins by
tracing the historical evolution of traffic management methodologies and the
pivotal role that AI cameras play in modernizing these approaches. Li H's work
may explore the diverse applications of AI cameras in traffic flow optimization,
congestion detection, and real-time decision-making, highlighting the
transformative impact on urban transportation systems.
Wang L's (2017) study titled ‘Machine learning in traffic surveillance: A review’
published in the "Intelligent Transport Systems Journal,” provides a
comprehensive review of machine learning applications in traffic surveillance.
The literature review likely begins by tracing the historical progression of traffic
surveillance methods and the pivotal shift toward machine learning techniques.
Wang L's work may explore the diverse range of machine learning algorithms
employed in traffic surveillance, emphasizing their roles in pattern recognition,
anomaly detection, and predictive modelling.
10
CHAPTER 3
THEORETICAL FRAMEWORK
3.1 INTRODUCTION
11
pedestrian and cyclist detection, as well as contributing to smart city initiatives
aimed at improving overall urban mobility.
Investment and interest in AI boomed in the first decades of the 21stcentury when
machine learning was successfully applied to many problems in academia and
industry due to new methods, the application of powerful computer hardware,
The Artificial Intelligence cameras were installed to monitor traffic violations on
Kerala’s roads under the Safe Kerala project. According to the official data, the
number of road accidents in July 2023 was 67 as compared to 313 recorded last
year in July and the collection of immense data sets.
1. Accidents and traffic violation have dropped after AI camera has been installed
and this has led to road safety.
3. Clear detection and instant alert can be made on unknown people and vehicles
using AI camera.
12
4. AI cameras play a vital role in enhancing security by monitoring and detecting
suspicious activities, ensuring the safety of transportation systems.
In Kerala, these cameras are being used to detect various traffic violations such
as helmet-less riding, triple riding on two-wheelers, failure to use seatbelts,
mobile phone use while driving, jumping red signals, over speeding, dangerous
driving, and illegal parking.
13
1 To reduce the traffic rule violations.
2 To prevents the occurrence of accidents.
3 To follow road safety rules and regulations.
The vehicle owner will receive an SMS alert once the AI camera detects a traffic
violation.
2. Enhanced Road Safety: AI cameras can detect and capture traffic violations
such as over speeding, red light running, and dangerous driving behaviour. This
can act as a deterrent and promote safer driving practices.
14
4. Stolen Vehicle Detection: AI cameras equipped with automated number plate
recognition can help in identifying stolen vehicles, aiding in their recovery and
preventing their misuse.
The Kerala State government had installed AI cameras across the state under the
‘safe Kerala’ project to monitor the traffic violations. The MVD recently installed
726 AI cameras to monitor traffic violations.
Trivandrum – 93
Kollam – 58
Pathanamthitta – 44
Alappuzha – 41
Kottayam – 44
Idukki – 38
Ernakulam – 64
Trissur – 57
Palakkad – 47
Malappuram – 49
Kozhikode – 63
Wayanad – 27
Kannur – 50
Kasargode – 47
15
CHAPTER 4
DATA ANALYSIS AND INTERPRETATION
TABLE 4.1 SHOWING AGE OF RESPONDENTS
0
2
8
18-25
25-35
35-45
45&above
90
INTERPRETATION
The above table shows that 90% of the respondents belong to the age group of
18-25, 8% of the respondents belong to the age of 25-35, 2% of the respondents
belongs to the age 35-45 and no respondents belongs to the age group above 45.
16
TABLE 4.2 SHOWING GENDER OF RESPONDENTS
0%
40%
male
female
others
60%
INTERPRETATION
The above analysis shows that 60% of the respondents are female, 40% of the
respondents are male and there are no other respondents. Hence majority of the
respondents are female.
17
TABLE 4.3 SHOWING OCCUPATION OF THE
RESPONDENTS
4%
6%0% 0%
2%
student
govt employees
pvt employees
self employed
88%
unemployed
others
INTERPRETATION
From the above diagram, 88% of the respondents are students, 6% of the
respondents are private employees, 4% of the respondents are unemployed, 2%
are government employees and no respondents for self-employed and others.
18
TABLE 4.4 SHOWING MONTHLY INCOME OF THE
RESPONDENTS
70
60
50 below 5000
40 5000-25000
30 25000-45000
45000-55000
20
55000 & above
10
0
below 5000 5000-25000 25000-45000 45000-55000 55000 & above
INTERPRETATION
From the above chart, 66% of the respondents belongs to the income group of
below 5000, 26% of the respondents belongs to 5000-25000, 6% of the
respondents belongs to 25000-45000, 2% of the respondents belongs to
55000&above and no respondents belongs to the income group of 45000-55000.
19
TABLE 4.5 SHOWING LOCATION OF THE RESPONDENTS
26%
40% urban
rural
suburban
34%
INTERPRETATION
From the above chart, 40% of the respondents reside in urban area, 34% of the
respondents are in rural area and 26% of the respondents reside in suburban area.
20
TABLE 4.7 SHOWING WHETHER RESPONDENTS ARE
AWARE OF PENALTIES
somewhat
no
yes
0 10 20 30 40 50 60 70
somewhat no yes
INTERPRETATION
From the above analysis, it can be inferred that 64% of the respondents are aware
of penalties, 24% of the respondents are choose somewhat and 12% of the
respondents are not aware of penalties.
21
TABLE 4.8 SHOWING WHETHER RESPONDENTS
VIOLATED ANY RULE
2%
0%
16%
never
16% rarely
sometimes
66%
often
always
INTERPRETATION
From the above chart, 66% of the respondents never violated any rule, 16% of
the respondents choose rarely and sometimes, 2% of the respondents often
violated the rule and no respondents choose always.
22
TABLE 4.9 SHOWING THE RULES VIOLATED BY
RESPONDENTS
not wearing seatbelt not wearing helmet violating speed limit other than the above never
70
60
50
40
30 62
20
10 18
6 10
0 4
not wearing seatbelt not wearing helmet violating speed limit other than the above never
INTERPRETATION
From the above analysis, 62% of the respondents never violated any rule, 18% of
the respondents violated for not wearing helmet, 10% of the respondents violated
speed limit, 6% of the respondents violated for not wearing the seatbelt, 4% of
the respondents violated other rules.
23
TABLE 4.10 SHOWING THE IMPACT ON SAFETY AND
SECURITY
6%
16% 28%
high impact
moderate impact
low impact
no impact
50%
INTERPRETATION
From the above chart, 50% of the respondents finds that AI cameras have
moderate impact on safety and security, 28% of the respondents says that AI
cameras have high impact, 16% of the respondents says that AI cameras have low
impact, and 6% of the respondents says that AI cameras have no impact on safety
and security.
24
TABLE 4.11 SHOWING THAT STRICT GUIDELINES
SHOULD BE FOLLOWED TO PREVENT BIASES AND FAIR
USE OF AI CAMERA
60
50
strongly disagree
40
disagree
30 58
neutral
20 strongly agree
22
10 16 agree
4 0
0
agree strongly neutral disagree strongly
agree disagree
INTERPRETATION
From the above analysis, 58% of the respondents agree that strict guidelines
should be followed to prevent biases and fair use of AI camera, 22% of the
respondents neutrally agreed, 16% of the respondents strongly agree, 4% of the
respondents disagree, and no response for strongly disagree.
25
TABLE 4.12 SHOWING THE USER EXPERIENCE WITH THE
AI CAMERA SYSTEM
8
14
excellent
14
good
fair
poor
64
INTERPRETATION
From the above chart, 64% of the respondents have good opinion to the user
experience with the AI camera system, 14% of the respondents have fair and poor
opinion to the user experience with the AI camera, 8% of the respondents have
excellent opinion.
26
TABLE 4.13 SHOWING THE LEVEL OF SUPPORT ON
INSTALLATION OF AI CAMERA
8 12
strongly support
32
support
neutral
48
oppose
strongly oppose
INTERPRETATION
From the above analysis, 48% of the respondents support to the installation of AI
camera, 32% of the respondents neutrally support, 12% of the respondents
strongly support, 8% of the respondents oppose to it and no response on strongly
oppose.
27
TABLE 4.14 SHOWING THE LEVEL OF SATISFACTION
TOWARDS THE USE OF AI CAMERA
Extremely
Extremely
Satisfied,
Dissatisfied,
2 4
Satisfied, 28
Dissatisfied, 28 Extremely Dissatisfied
Dissatisfied
Neutral
Satisfied
Extremely Satisfied
Neutral, 38
INTERPRETATION
The above diagram shows that 38% of the respondents gives Neutral on the use
of AI cameras, 28% of the respondents gives Dissatisfied and Satisfied, 4% of the
respondents gives Extremely Dissatisfied and 2% of the respondents gives
Extremely Satisfied with the use of AI cameras.
28
TABLE 4.15 SHOWING WHETHER RESPONDENTS
BELIEVE THAT AI CAMERA REDUCES TRAFFIC
CONGESTION
8 2 10
strongly believe
believe
neutral
36
disbelieve
44
strongly disbelieve
INTERPRETATION
From the above chart, 44% of the respondents says neutral on ‘AI camera reduces
traffic congestion’, 36% of the respondents believe, 10% of the respondents
strongly believe, 8% of the respondents disbelieve, 2% of the respondents
strongly disbelieve.
29
TABLE 4.16 SHOWING THE REGULARITY OF AUDITING
THE AI CAMERA SYSTEM
2
never
6
rarely
never
rarely
neutral 34
neutral
occasionally
occasionally
46 regularly
regularly
12
0 5 10 15 20 25 30 35 40 45 50
INTERPRETATION
From the above analysis, 46% of the respondents says occasionally auding should
be done, 34% of the respondents says neutral opinion, 12% of the respondents
says regularly auding should be done, 6% of the respondents says rarely, 2% of
the respondents says auding should not be done regularly.
30
TABLE 4.17 SHOWING THE LEVEL OF AGREEMENT
TOWARDS PROBLEMS OF AI CAMERA
31
CHART 4.17 SHOWING THE LEVEL OF AGREEMENT
TOWARDS PROBLEMS OF AI CAMERA
26 27
23
30 20 21
23 22 23
25 20
8 18
20 6 5
4 4
15
disagree
10
neutral
5
agree
0
presence of AI People may face There are Possibility of The presence of
camera raises challenges significant misinterpreting ai cameras may
potential issues regarding their privacy the footage lead tp a
related to data personal concerns captured by AI decrease in face
security information associated with cameras, to face
being collected the use of AI leading to false interaction
and analyzed by cameras in accusations
AI camera in transportation against
transportation individuals
INTERPRETATION
From the above diagram, out of 50, 23 each of the respondents agree and neutrally
agree to the first statement and 4 of the respondents disagree to it.
32
TABLE 4.18 SHOWING THE LEVEL OF OPINION ABOUT
THE FUTURE OF AI CAMERA
yes, 38
may be , 42
no, 20
INTERPRETATION
From the above chart, 42% of the respondents says future of AI camera may be
promising, 38% of the respondents says it will be promising, 20% of the
respondents says it will not be promising.
33
WEIGHTED RANKING METHOD
TABLE 4.6 SHOWING THE BENEFITS OF AI CAMERA IN
TRANSPORTATION
No. of respondents
Options Total
1 2 3 4 5
Improved 24 15 4 4 3 50
safety
Enhanced 10 21 13 2 4 50
traffic
management
Faster 18 8 15 4 5 50
emergency
response
Better data 10 16 11 8 5 50
for urban
planning
Infrastructure 11 9 14 3 13 50
development
34
TABLE 4.6 SHOWING RANK GIVEN TO THE BENEFITS OF
AI CAMERA IN TRANSPORTATION
No. of respondents
Rank 1 2 3 4 5 Total Rank
Weight 5 4 3 2 1
Improved 120 60 12 8 3 203 I.
safety
Infrastructure 55 36 42 6 13 152 V.
development
INTERPRETATION
Majority of the respondents have given first rank to improved safety, second
rank to enhanced traffic management, third rank to faster emergency response,
fourth rank to better data for urban planning, fifth rank to infrastructure
development.
35
CHAPTER 5
FINDINGS, SUGGESTIONS AND
CONCLUSION
5.1 FINDINGS
36
Majority (52%) of the respondents says neutral to the problem –
‘Personal information being collected and analyzed by AI camera’.
Majority (44%) of the respondents agree to the problem – ‘Significant
privacy concerns associated with the use of AI cameras in
Transportation’.
Majority (46%) of the respondents agree to the problem –
‘Misinterpreting the footage captured by AI cameras.
Majority (54%) of the respondents says neutral opinion to the problem –
‘AI camera may lead to decrease in face-to-face interaction and a
potential loss of empathy.
➢ Majority (42%) of the respondents says future of AI camera may be
promising.
37
5.2 SUGGESTIONS
• Give technical awareness among the people who are not aware of it.
• Improve technical side of AI camera, so that false footage can be reduced.
• Use AI camera to monitor traffic flow and congestion in busy areas.
• Create a system that alerts drivers about traffic jams and accidents in real-
time.
• Install multi-angle cameras to provide a comprehensive view of the
surrounding environment, minimizing blind spots and enhancing situation
awareness.
• Quick paying of penalty can be made, so that future of AI camera can be
promising.
38
5.3 CONCLUSION
39
BIBLIOGRAPHY
BIBLIOGRAPHY
BOOKS
Research Methodology and techniques – C R Kothari
ONLINE REFERENCES
1. https://keralakaumudi.com/en/news/mobile/news.php?id=1081311
&u=road-camera-fine-from-today-children-to-be-exempted-
1081311
2. https://www.infosysbpm.com/blogs/business-
transformation/advantages-of-using-artificial-intelligence-in-video-
surveillance.html
3. https://chat.openai.com/
4. https://en.m.wikipedia.org
5. https://www.manoramaonline.com
6. https://malayalam.indiatoday.in
40
APPENDIX
GOVERNMENT ARTS AND SCIENCE COLLEGE, THOLANUR
Department of commerce
Dear respondent,
SUB: “request to fill the questionnaire” Regarding a research study
I am a final year student pursuing Bachelor of Commerce (finance) at
GOVT. ARTS AND SCIENCE COLLEGE THOLANUR (under Calicut
University). I am conducting “A study on people’s attitude towards AI
camera in transportation with special reference to Palakkad
district”. This research (project) is taken as partial fulfilment of the
completion of my B.com (finance) degree under Calicut University.
I seek your kind assistance in completing the attached questionnaire
which would take approximately 10 minutes of your valuable time.
1. AGE
a) 18-25
b) 25-35
c) 35-45
d) 45 and above
2. GENDER
a) Male
b) Female
c) Others
3. OCCUPATION
a) Student
b) Government sector employees
c) Private sector employees
d) Self employed
e) Unemployed
f) Others
4. MONTHLY INCOME
a) Below 5000
b) 5000-25000
c) 25000-45000
d) 45000-55000
e) 55000 and above
41
5. LOCATION
a) Urban
b) Rural
c) Sub-rural
a) Improved safety
b) Enhanced traffic
management
c) Faster emergency
response
d) Better data for urban
planning
e) Infrastructure
development
a) Yes
b) No
c) Somewhat
42
a) Never
b) Rarely
c) Sometimes
d) Often
e) Always
a) High impact
b) Moderate impact
c) Low impact
d) No impact
a) Agree
b) Strongly agree
c) Neutral
d) Disagree
e) strongly disagree
43
12. HOW WOULD YOU RATE THE USER EXPERIENCE OF
INTERACTING WITH AI CAMERA SYSTEM?
a) Excellent
b) Good
c) Fair
d) Poor
a) Strongly support
b) Support
c) Neutral
d) Oppose
e) Strongly oppose
EXTREMELY DISSATISFIED
a) 1
b) 2
c) 3
d) 4
e) 5
EXTREMELY SATISFIED
44
a) Strongly believe
b) Believe
c) Neutral
d) disbelieve
e) strongly disbelieve
a) Regularly
b) Occasionally
c) Neutral
d) Rarely
e) Never
45
d) Possibility of
misinterpreting the footage
captured by AI camera’s,
leading to false accusations
against individual
e) The presence of AI
cameras may lead to a
decrease in face-to-face
interactions and a potential
loss of empathy
a) Yes
b) No
c) It depends
46