"EcoSort: Advancing Smart Bins Waste Segregation and Management with Image Recognition and Incentive Rewards"

Table 1. Compilation of Technologies. In this section, list the technologies being used. Add more rows to the table, if necessary.
Technology Strength Weakness Areas of Application

IoT - Real-time monitoring, control, and automation of - IoT devices can be the target of cyberattacks and ● Smart homes and home automation
diverse systems are made possible through connectivity and efficient illegal access, necessitating strict security rules and controls. As a ● Industrial automation and monitoring
data interchange, increasing operational effectiveness and result, security flaws and privacy concerns present threats. ● Environmental monitoring and sustainability
productivity. - The creation of standardized protocols and ● Healthcare and remote patient monitoring
- IoT device data-driven insights help with informed frameworks is required because interoperability issues between ● Smart cities and infrastructure management
decision-making, proactive maintenance, and process optimization, various IoT devices and platforms obstruct seamless integration and ● Transportation and logistics
which results in reduced costs and better results. scaling. ● Agriculture and precision farming
● Retail and supply chain management
● Energy management and smart grids
● Safety and surveillance systems

Robotics - Automation of labor-intensive, repetitive processes can - -Robotics system adoption may be constrained by ● Manufacturing and assembly lines
increase productivity and efficiency while also lowering costs. the initial high costs of acquisition, deployment, and maintenance, ● Warehousing and logistics
- The improvement of quality and reliability through especially for smaller firms. ● Healthcare and medical procedures
increased precision and accuracy in activities requiring dexterous or - Robots have limited capacity to adapt to complicated, ● Agriculture and farming
sophisticated actions. chaotic surroundings or jobs that call for human-level dexterity and ● Construction and infrastructure
judgment. ● Automotive industry
● Service industry (e.g., hospitality, retail)
● Research and exploration (e.g., space, deep-sea)
● Home assistance and domestic tasks
● Entertainment and gaming industry

Artificial Intelligence - Strong data analysis capabilities that let businesses - A responsible approach to development and ● Finance and banking
draw out important trends, patterns, and insights from big, complicated deployment is required due to ethical considerations and potential ● Healthcare and medical diagnosis
datasets. biases in AI algorithms that could result in discriminatory outcomes ● Customer service and chatbots
- Routine jobs can be automated, freeing up human and privacy problems. ● Cybersecurity and fraud detection
resources for more innovative and strategic duties, improving overall - Organizations with limited resources may face ● Personalized marketing and recommendation systems
productivity and efficiency. difficulties due to the demand for significant computational power ● Autonomous vehicles and transportation
and data infrastructure to develop and train AI models. ● Natural language processing and voice assistants
● Image and video recognition
● Smart homes and Internet of Things (IoT) integration
● Predictive analytics and decision support systems

Machine Learning - Predictive analytics, picture recognition, and natural - Relying on the availability of broad, high-quality ● Predictive maintenance and fault detection
language processing are just a few examples of difficult jobs that training datasets, which can be difficult and time-consuming to ● Healthcare diagnosis and personalized treatment
might benefit from the ability to learn from massive amounts of data obtain, especially for specialized topics. ● Financial risk assessment and fraud detection
 and improve performance over time. - Interpretability and explainability of machine learning ● Demand forecasting and supply chain optimization
- Data analysis, pattern recognition, and decision- models are important, as outcomes from complicated algorithms ● Recommendation systems and personalized marketing
making are automated, which improves accuracy, productivity, and sometimes aren't transparent or easy to understand. ● Natural language processing and sentiment analysis
scalability in a variety of applications. ● Image and video recognition
● Virtual assistants and chatbots
● Autonomous vehicles and robotics
● Cybersecurity and threat detection

Hyper Automation - Automating end-to-end business processes with a - To achieve a seamless transition and effective use of ● Business process automation and optimization
variety of automation technologies, including robotic process the hyper automation framework, complex implementation and ● Data entry and document processing
automation (RPA), artificial intelligence (AI), and machine learning integration processes are needed, which need knowledge and ● Customer service and support
(ML), will increase productivity and agility. resources. ● Supply chain management and logistics
- Decreasing errors and increasing productivity by - Monitoring and management of automated systems ● Financial and accounting processes
streamlining procedures and eliminating human handoffs through on an ongoing basis to verify their accuracy, dependability, and ● Human resources and employee onboarding
seamless communication and collaboration between various compliance with evolving legal requirements or corporate needs. ● IT operations and infrastructure management
automated systems. ● Sales and marketing operations
● Compliance and regulatory tasks
● Workflow coordination and task allocation

Image Processing - Real-time object recognition, feature extraction, and - Particularly when working with huge image ● Biomedical imaging and diagnostics
pattern detection are made possible by high-speed image analysis collections or real-time processing, computational complexity and ● Security and surveillance systems
and processing, supporting a wide range of applications in numerous resource demands can be taxing, necessitating a strong hardware ● Automated quality control and inspection
sectors. infrastructure. ● Augmented reality and virtual reality
- Noise reduction and picture restoration procedures, for - Variations in lighting conditions, image quality, ● Satellite and aerial imagery analysis
example, enhance the quality and clarity of photographs, allowing for occlusions, and other factors may have an impact on accuracy and ● Autonomous vehicles and robotics
improved visual interpretation and analysis. resilience, necessitating the use of robust algorithms and ● Facial recognition and biometrics
preprocessing techniques. ● Digital image restoration and forensics
● Object detection and tracking
● Virtual tours and interactive media

Computer Vision - Computers can now interpret and comprehend visual - Complex computer vision models that are difficult to ● Autonomous vehicles and drones
data thanks to sophisticated algorithms and machine learning interpret may make it difficult to comprehend the underlying decision- ● Robotics and industrial automation
approaches that automatically extract, analyze, and comprehend making process, thus undermining trust and transparency. ● Medical imaging and diagnostics
visual information from photographs and videos. - Robust preprocessing and data augmentation ● Surveillance and security systems
- Applications including autonomous navigation, techniques are needed because variations in image circumstances, ● Virtual reality and augmented reality
augmented reality, and medical imaging, among others, are supported such as illumination, perspective, and occlusions, can affect the ● Gesture recognition and human-computer interaction
by object identification, image segmentation, and scene understanding precision and dependability of computer vision algorithms. ● Retail and e-commerce (visual search, product
capabilities. recommendations)
● Sports analytics and action recognition
 ● Artificial intelligence in gaming and simulation
● Facial analysis and emotion detection

Table 2. Problem Statement.

1. From among the technologies in Table 1, which do you intend to use? Justify why it is the best choice.
- With its numerous uses and rising demand, the Internet of Things (IoT) stands out among technologies as the best choice. The potential impact increases in importance when IoT is
combined with the idea of Smart Bin Waste Segregation. Utilizing IoT technology, Smart Bin Waste Segregation improves waste management procedures while boosting resource efficiency and environmental sustainability. By lowering landfill waste, increasing recycling, and
practicing resource conservation, it tackles real-world challenges. Additionally, IoT-enabled Smart Bin Waste Segregation modifies conventional waste management processes, enhancing efficiency and effectiveness.
2. What motivates you in pursuing the study? (from “Technology” in Table 1)
- The goal to advance environmental sustainability, maximize resource usage, encourage technical innovation, improve public health and wellbeing, and contribute to policy and governance
frameworks are the driving forces behind the pursuit of the study of smart bin waste segregation. Researchers and individuals can work to create a cleaner, more sustainable future by studying the possibility of IoT-enabled Smart Bin Waste Segregation. This topic of research
provides a forum for cross-disciplinary cooperation and innovation, enabling us to address the world's waste management dilemma and have a good impact on society and the environment.
3. Give the Problem Statement of your research. Use the Justification (1 and 2 above) to formulate this.

- The ineffectiveness of the existing waste management strategies can be attributed to a lack of knowledge on waste segregation, a lack of real-time monitoring tools, and a lack of exact data to allow for informed decision-making. Implementing IoT-powered
Smart Bin Waste Segregation systems will change waste management procedures. The purpose of this research is to address these problems by utilizing IoT technology to improve trash segregation, enable real-time monitoring, and provide insightful data for effective
resource allocation and sustainable waste management. This will enhance garbage management and encourage environmental sustainability.

Table 3. List of Chosen Technology. Add more rows, if necessary.

Technology Author/Year(Cite) Description Strength Weakness Application

IoT Bhutada, S., et.al. Smart Waste Management using IOT Real-time data on the amount of trash is gathered One of the disadvantages of the IoT-based Smart Smart Cities
(2018) by ultrasonic sensors positioned beneath the bin lids Waste Management system is its potential for hacking
and sent to a server via a Wi-Fi modem. Each trash and illegal access. IoT devices that are connected to
can contains a low-cost integrated chip that tracks the internet are at risk of security flaws and privacy
the level of waste and has a distinct ID. The device violations. To reduce the possibility of security
communicates the relevant data and ID when the problems and protect against privacy issues, it is
waste level rises above a preset threshold, allowing crucial to install strict security measures, strong
authorities to remotely examine the information and encryption protocols, and continual monitoring.
take immediate action on the impacted bins.

Robotics Vijay, S., et al. (2019) Smart Waste Management System The use of an embedded IoT system for effective Due to its reliance on sensors for trash identification Smart Cities

using ARDUINO waste management in big cities is one of the and segregation, the given waste management
strengths of the provided waste management system has a potential drawback. Environmental
system. The system eliminates the need for ongoing conditions, sensor calibration, and upkeep
manual monitoring by enabling remote monitoring requirements are just a few of the variables that may
and management of dumpsters through the use of have an impact on how precise and reliable the
IoT technology. By doing this, waste collection and sensors are at detecting and classifying garbage. A
handling procedures become more effective overall waste management system's performance could be
and use fewer resources, including human labor. jeopardized by inaccurate sensor readings or sensor
failures that result in inappropriate waste segregation.
segregation.
Artificial Intelligence Malik, M., et al. (2022) Waste Classification for Sustainable The EfficientNet-B0 model, which provides excellent The reliance on pre-training the EfficientNet-B0 model Resource Optimization

Development Using Image accuracy levels in image classification tasks, is one on the ImageNet dataset presents a potential

Recognition with Deep Learning of the study paper's strong points. The study shortcoming of the study. Despite the fact that the

Neural Network Models illustrates the capacity to tailor the model for ImageNet dataset is frequently used for pretraining
particular demographic regions by utilizing transfer models, it might not exactly match the distinctive
learning and fine-tuning strategies, attaining qualities and variations visible in litter photos. The
equivalent accuracy to bigger models while using biases and constraints of the pretraining dataset may
noticeably fewer parameters. This method makes a have a negative impact on the model's performance
significant contribution to the field of image-based in litter classification tasks, sometimes resulting in
litter classification because it not only increases inaccurate classifications or subpar accuracy.
efficiency in terms of processing resources but also
improves classification performance for litter photos.

Machine Learning Hin, L.C., et al. (2021) An Intelligent Smart Bin for Waste The potential of machine learning to be included into The smart bin system needs adequate and Image recognition

Management the smart bin system to continuously develop and representative training data for machine learning to be
increase performance is one of its benefits. The integrated. Poor performance and incorrect waste
system may adapt and improve garbage classification can result from biased or limited data.
classification and sorting procedures over time by Moreover, the logistical difficulties posed by the
utilizing machine learning algorithms. The smart bin requirement for routine upkeep and updates of
can learn from data, recognize patterns, and make machine learning models. To make the most of the
more accurate decisions thanks to the machine smart bin system's capabilities, access to high-quality
learning component, which boosts recycling rates training data and continual model upkeep are
and waste management effectiveness. essential.

Hyper Automation Sigongan, J., et al. GULP: Solar-Powered Smart Garbage The deployed smart bin system's novel design and The deployed smart bin system's reliance on SMS Sustainability and Energy
(2023) Segregation Bins with SMS Notification automation of solid waste disposal procedures are notifications for overflow prevention presents a Independence
and Machine Learning Image its main advantages. The system offers effective possible problem. While cellular networks may have
Processing trash management and helps minimize bin overflow restrictions in terms of their reach and dependability in
by including cutting-edge features like fill level some locations or during network failures, SMS
monitoring and SMS notifications. The system's notifications can be an efficient way to warn relevant
sustainability and independence are further individuals. This can lead to ignored or delayed
increased by integrating a solar energy source with notifications, which could result in a bin overflow and
the supporting application, which lessens reliance ineffective waste management. Additionally, relying
on outside power sources. This strength enhances on a solar energy source may present difficulties for
the institution under study's waste management the system's operation and energy delivery in regions
procedures while also fostering environmental with little sunlight or during protracted periods of bad
sustainability and lowering operating expenses. weather.

Image Processing Mitra, A., (2020) Detection of Waste Materials Using The provided automated trash detection system The automated waste detection framework's Automation and Productivity
 Deep Learning and Image Processing excels in effectively identifying and classifying waste dependence on a sizable image dataset for deep
products with the use of deep learning algorithms learning algorithm training is one of its drawbacks.
and image processing methods. The framework The dataset's quality, diversity, and
achieves great accuracy in finding patterns and representativeness have a significant impact on the
classifying garbage into six unique categories by framework's efficacy and accuracy. The framework's
training the algorithms with a large dataset of performance may be constrained if the dataset does
photos. The framework's adaptability also enables it not completely account for all potential trash changes
to recognize and categorize waste elements from or if it contains biases, which could result in waste
video feeds or live webcam streams in addition to classification errors or inaccurate waste detection.
photos, broadening its application and enabling real- Therefore, assuring the availability of thorough and
time waste detection capabilities. objective training datasets is essential for improving
the framework's performance and dependability.

Computer Vision Aarthi.R., et al. A Vision Based Approach to Localize The use of the Mask RCNN deep learning The reliance on the Mask RCNN deep learning Robotic arm systems
(2023) Waste Objects and Geometric architecture for real-time autonomous waste architecture for waste identification and classification

Features Exaction for Robotic material recognition and segregation is one of the is a potential drawback of the work as it is currently

Manipulation study's strengths. The performance of this being reported. Despite the architecture's reputation
architecture, which offers excellent accuracy and for cutting-edge speed, it could still have trouble
reliability in identifying trash objects in natural correctly identifying and classifying trash materials,
situations, is known for being state-of-the-art in especially in chaotic or complicated situations. The
terms of object recognition and classification tasks. accuracy and robustness of the detecting system may
be affected by variables in lighting, occlusions, or
overlapping waste materials.

Table 4. Proposed Model/System to Chosen Technology. The proposed modification must be unique and different from those in Table 5.
1. From the technology identified in Table 3 (whichever is applicable), what particular technology are you going to use?

- The use of IoT-based smart waste management in smart cities may be used because it makes it possible to collect data in real-time and monitor the amount of trash in trash bins, improving waste management procedures and
encouraging resource efficiency. The system can reduce possible risks related to hacking and unauthorized access and ensure the privacy and integrity of acquired data in Smart Cities by putting in place strong security measures, such as stringent access controls and
encryption methods.
2. Will the proposed study provide a solution to the research problem as identified in Table 2? How?

- The proposed study may effectively recognize and categorize waste materials by incorporating image recognition technology, addressing the lack of information regarding trash segregation. The lack of real-time monitoring can also be addressed by incorporating
real-time monitoring capabilities into the smart bins, which would provide current data on fill levels and trash composition. The gathered information can be used to make well-informed decisions about waste management tactics, enhancing the efficacy and efficiency of waste
management procedures. Additionally, including incentive prizes might encourage people to appropriately separate waste, raising awareness and encouraging participation in waste management initiatives. Overall, by offering information, real-time monitoring, and data-driven
decision-making skills, the proposed developments in smart bin technology can help solve the inefficiency of current waste management systems.
3. Write down your research statement – that your proposed solution (Table 2 above) will solve the identified problem.

- The study aims to advance smart bin waste segregation and management by incorporating image recognition technology and real-time monitoring capabilities. The study seeks to address the lack of information on waste segregation and the inefficiency of existing
waste management systems by accurately recognizing and categorizing waste materials in real-time. By leveraging the collected data, the research intends to enable data-driven decision-making, enhancing the efficacy and efficiency of waste management procedures.
Additionally, the inclusion of incentive rewards aims to promote proper waste segregation, raising awareness and encouraging active participation in waste management initiatives. The research ultimately seeks to contribute to more effective and sustainable waste management
practices through the advancement of smart bin technology.
4. What Is the proposed application for your technology? What will be the final output (model, system, etc.)? Describe.

- The proposed IoT application is under the Smart Waste Management, aiming to improve waste management practices through the integration of IoT technology. The output will be an advanced system consisting of smart bins equipped with camera-based image
processing, Arduino-based boards for bin functionality, sensors, and an incentive reward mechanism. The system will enable real-time data collection on waste fill levels and send SMS notifications to update bin status to the client or user. By enabling data-driven decision-
making and proactive waste collection, the goal is to enhance efficiency, minimize expenses, and foster sustainability in waste management operations.

"SmartTranscript: Revolutionizing Academic Records with Blockchain-Enabled Smart Contract Issuance and Verification"

Table 1. Compilation of Technologies. In this section, list the technologies being used. Add more rows to the table, if necessary.
Technology Strength Weakness Areas of Application

Blockchain - Decentralization guarantees the safe and unalterable storage of - Scalability may be an issue, as the blockchain's capacity to
academic records manage a high volume of transactions involving academic records ● Educational institutions
- Immutability and transparency facilitate efficient and reliable may become constrained. ● Human resources departments
verification of academic credentials, thereby reducing fraud and ● Credential verification agencies
enhancing confidence. - Institutions with limited resources may struggle to implement and ● Government agencies
- Smart contract functionality automates issuance and verification sustain blockchain technology due to the need for adequate ● Professional licensing bodies

processes, streamlining administrative duties and minimizing technical infrastructure and skilled personnel. ● Immigration and visa services
documentation. ● Online learning platforms
● Job recruitment platforms
● Background check providers
● Scholarship and grant organizations

Smart Contracts - Smart contracts automate and enforce predefined rules for the - Complex smart contract programming may require specialized
issuance and verification of academic records, thereby reducing skills and, if not properly implemented and tested, may result in ● Academic record issuance and verification
human error and enhancing accuracy. errors. ● Supply chain management
- They eliminate the need for intermediaries, thereby streamlining the - If errors or disputes arise, the immutability of smart contracts can ● Insurance claims processing
process and decreasing administrative expenses. be a detriment, as they cannot be readily modified or updated. ● Financial transactions
- Smart contracts provide transparency and auditability, enabling ● Real estate transactions
stakeholders to monitor and validate each step of the record ● Supply chain management
management process. ● Digital identity management

Distributed Ledger Technology (DLT) - DLT guarantees the decentralized and distributed storage of - Proof-of-work and other DLT consensus mechanisms can ● Academic records management
academic records, thereby augmenting security and removing single necessitate substantial computational resources and energy ● Supply chain management
points of failure. consumption. ● Healthcare data management
- It provides mechanisms for transparency and consensus, nurturing - The size of the distributed ledger may grow over time, necessitating ● Voting systems
trust among multiple stakeholders in the academic record ecosystem. a considerable amount of storage space and possibly impacting ● Asset tokenization
- DLT enables efficient data sharing and collaboration while preserving system performance. ● Intellectual property rights management
 data integrity, which is advantageous for academic research and
collaborative projects.

Cryptography - Cryptography protects academic records from unauthorized access - Weak encryption algorithms or improper key management can ● Secure academic record storage and transmission
and interference by ensuring their confidentiality and integrity. expose academic records to vulnerabilities and compromise their ● Identity and access management
- Encryption safeguards sensitive data, allowing only authorized security. ● Secure communication channels
parties to access and decrypt the information. - Non-technical users may have difficulty comprehending and ● Digital rights management
- Digital signatures offer irrefutability, assuring the originality and implementing secure practices due to cryptography's mathematical Secure online transactions
authenticity of academic records. foundations and complexity. ● Secure messaging systems

Cybersecurity - Cybersecurity measures secure academic records against - Cyber threats and attack techniques are constantly evolving, ● Protection of academic records from cyber threats
unauthorized access, data breaches, and cyber threats. necessitating continuous monitoring and updates to maintain robust ● Network security for academic institutions Data privacy and
- Systems for detecting and preventing intrusions protect against cybersecurity. compliance
malicious activities, ensuring the availability and integrity of academic - Academic records can be jeopardized by human error, such as ● User authentication and access control Incident response and
records. insecure passwords or phishing attacks, which can undermine threat intelligence Security awareness and training
- Regular security audits and vulnerability assessments identify and cybersecurity measures.
address vulnerabilities, thereby decreasing the risk of data breach. - The implementation of comprehensive cybersecurity measures
may be hindered by budgetary and resource constraints.

Web Development - Web development technologies facilitate access to and interaction - Rapidly evolving web technologies necessitate ongoing education
with academic record systems via user-friendly interfaces. and adaptation in order to remain current. ● Web-based academic record portals
- Responsive web design assures device compatibility, thereby - When developing web applications, cross-browser compatibility ● Online enrollment and registration systems Interactive learning
enhancing the user experience for students, institutions, and issues may arise, necessitating additional testing and optimization platforms
businesses. efforts. ● E-learning management systems
- The incorporation of web services and application programming - Vulnerabilities in Web security, such as cross-site scripting and ● Career services platforms
interfaces (APIs) facilitates data exchange and interoperability SQL injection, threaten the confidentiality and integrity of academic ● Collaborative research platforms
between academic record systems and other applications. records.

Table 2. Problem Statement.

4. From among the technologies in Table 1, which do you intend to use? Justify why it is the best choice.

Due to its superior security, transparency, and automation capabilities, blockchain is the best option for redefining academic records. Its decentralized and immutable nature guarantees the accuracy of academic records, while its transparency inspires confidence among
stakeholders. Through the use of smart contracts, administrative duties are streamlined and errors are reduced. Moreover, blockchain technology facilitates global collaboration and recognition of academic credentials, thereby promoting educational mobility and
standardization. By leveraging these advantages, blockchain technology can provide an academic records management system that is secure, efficient, and reliable.
5. What motivates you in pursuing the study? (from “Technology” in Table 1)

The need for increased security, streamlined administrative processes, and international recognition prompted the study of blockchain-enabled smart contract issuance and verification as a means of revolutionizing academic records. By leveraging
blockchain technology, academic records can be safely stored, ensuring data integrity and reducing the risk of fraud. The automation of administrative duties through smart contracts reduces paperwork and improves efficiency. In addition, blockchain
technology facilitates global collaboration and recognition of academic credentials, thereby promoting educational mobility and standardization.
6. Give the Problem Statement of your research. Use the Justification (1 and 2 above) to formulate this.

The current management of academic records faces challenges in security, inefficient administrative processes, and a lack of standardized verification mechanisms, necessitating the development of blockchain-powered smart contract technology to revolutionize academic
record management (or academic credential verification).

Table 3. List of Chosen Technology. Add more rows, if necessary.

Technology Author/Year(Cite) Description Strength Weakness Application

Blockchain Cheng, J, et.al., Blockchain and smart contract for digital The platform provides enhanced security through Challenges include scalability restrictions, complexity Educational institution
(2018) certificate tamper-proof storage on the Ethereum blockchain in implementation and maintenance, and potential
and the InterPlanetary File System, transparent interoperability and integration issues with existing
verification enabled by smart contracts and systems.
blockchain records, and a cooperative and
decentralized infrastructure that incorporates the
BSCW document management system.

Smart Contracts Malsa, N, et. al., Framework and smart contract for The development and testing of a smart contract to A blockchain-enabled certificate verification system Educational institutions
(2021) blockchain-enabled certificate store and validate certificates, allowing employers to with smart contracts is proposed by the study. Corporate Training and Workforce
verification system using robotics readily verify job applicants' credentials. Students, However, possible weaknesses include a lack of Development.
employers, and educational institutions can use this discussion on smart contract code vulnerabilities,
smart contract to expedite the certificate verification scaling difficulties, and the complexity of integrating
process. robotics technology with the blockchain infrastructure.
Additional research is required to address these
limitations.

Distributed Ledger Technology (DLT) Bessa, Emanuel E; A blockchain-based educational record The study confirms that DLT, as exemplified by the Possible DLT weaknesses demonstrated in the study Educational Institution

Martins, Joberto SB repository BcER2 system, enables secure and authentic include the lack of a scalability evaluation, potential Corporate Training and Workforce

(2019) access to educational records. It showcases the difficulties in bringing stakeholders together, and the Development.
potential of DLT for improving certificate need for standardized asset representation. While the
management and distribution in the education study validates the security and authenticity
sector, with possible applications in corporate verification capabilities of DLT, additional research is
training and workforce development. necessary to address scalability concerns and assure
the secure and decentralized use of the proposed
system.

Cryptography Lemieux, V. (2021) Blockchain and recordkeeping: Editorial Cryptography has proven successful at providing A potential weakness of cryptography in this study is Record keeping
immutability and dependability to distributed ledger the "oracle problem," as cryptography-based
technology (DLT) record systems, addressing the decentralized trust systems are unable to
double-spending problem and ensuring secure and autonomously verify real-world events or guarantee
tamper-resistant transactions. the authenticity of manually entered records in DLT
systems. In addition, interoperability issues, the need
for human-centered design, and usable security
remain unresolved issues requiring additional
research and development.
Cybersecurity Abdullah, A, et. al., Educational Blockchain: A Secure The paper proposes the use of blockchain This study identifies manual authentication processes, Higher Education Commission
(2021) Degree Attestation and Verification technology, specifically the HEDU-Ledger, for potential vulnerabilities in smart contracts, and the
Traceability Architecture for Higher secure and auditable degree attestation in the need for a robust security framework to safeguard
Education Commission Higher Education Commission (HEC). The HEDU- against cyberattacks targeting the blockchain
Ledger is a decentralized system that guarantees infrastructure as cybersecurity vulnerabilities.
security, transparency, and cyberattack protection.
The system conforms to HEC's policies and offers
non-repudiation, provenance, and traceability.

Web Development Muhammad, S, and Website-based Certificate Management In this study, one of the strengths of web If appropriate security measures are not implemented, Training and Consultant Certification
Daning, n (2022) Information System Design in Training development is the ability to automate the manual the weaknesses in this study's web development
and Consultant Division certificate administration system using a web-based methodology could lead to security vulnerabilities.
information system. This enables efficient and Additionally, reliance on a web-based system may
effective administration of personal information, introduce internet connectivity requirements and
registration agencies, and certificates. The web- potential browser or device compatibility issues. To
based system offers employees convenience, address these weaknesses, careful consideration
enhances data organization, and enables efficient should be given to data preservation and system
data backup and systemization. stability.

Table 4. Proposed Model/System to Chosen Technology. The proposed modification must be unique and different from those in Table 5.
5. From the technology identified in Table 3 (whichever is applicable), what particular technology are you going to use?

Blockchain technology assures improved security, transparency, and automation, as well as the potential for worldwide collaboration and credential recognition. It provides a decentralized and tamper-resistant system for storing and verifying academic records, streamlines
administrative processes through smart contracts, and assures data integrity and privacy. The administration of academic records can be revolutionized by utilizing blockchain technology, thereby fostering trust, efficiency, and portability.
6. Will the propose study provide solution to the research problem as identified in Table 2? How?

Yes, the proposed study on blockchain-based smart contract technology for academic record administration will offer a solution to the aforementioned research problem.
The use of blockchain technology provides enhanced security through tamper-proof and transparent record storage, thereby reducing the likelihood of fraud and data manipulation. Integrating smart contracts automates administrative procedures, thereby reducing inefficiencies
and errors. In addition, the use of blockchain technology permits the creation of standardized verification mechanisms, ensuring the reliability and efficiency of academic credential verification.
By utilizing blockchain technology and smart contracts, the proposed study will streamline the administration of academic records, improve security, and establish a reliable and efficient system for academic credential verification. This will address the security issues, inefficient
administrative procedures, and absence of standardized verification mechanisms that plague the current administration of academic records.
7. Write down your research statement – that your proposed solution (Table 2 above) will solve the identified problem.

This study seeks to address the challenges associated with academic record administration by developing a blockchain-based smart contract technology solution that improves security, efficiency, and standardized verification mechanisms.
8. What Is the proposed application for your technology? What will be the final output (model, system, etc.)? Describe.

The proposed application is an innovative academic record management system powered by blockchain-based smart contracts. The final output will be a functional system that allows for secure storage, issuance, and verification of academic records on the blockchain. This
system aims to offer a tamper-proof and transparent platform for storing, sharing, and verifying academic records, ensuring data integrity, reducing fraud, and streamlining administrative processes between academic institutions, employers, and other stakeholders.

VioLert: Parking Violation Monitoring and Detection through Image Recognition

Table 1. Compilation of Technologies. In this section, list the technologies being used. Add more rows to the table, if necessary.
 Technology Strength Weakness Areas of Application

Machine Learning - Through image recognition, machine learning - The system's accuracy heavily depends on how ● Municipal Parking Management
algorithms can identify parking violations with high accuracy, good and varied the training data are. Lack of training data or biased ● Private Parking Facilities
minimizing false positives and enhancing the effectiveness of training data can result in subpar performance or potential biases in ● Traffic Control
enforcement. parking violation detection.
- Real-time image processing capabilities of machine - As image recognition technology involves taking and
learning models allow for the immediate detection and notification of processing pictures of cars and license plates, privacy issues are
parking violations as they happen. raised when it is used. Sensitive data handling and protection should
be handled with the appropriate care.

Artificial Intelligence - It is possible to accurately predict and identify potential - It can be difficult to comprehend the reasoning ● Intelligent surveillance and monitoring
parking violations by using artificial intelligence algorithms to analyze behind the decisions or predictions made by some AI models, such ● Predictive analytics for parking management
parking data and find patterns, trends, and anomalies. as deep neural networks, due to their lack of interpretability. ● Adaptive parking regulations
- AI models can continuously improve their performance - AI models can inherit biases present in the training
and adaptability in the detection of parking violations by learning from data, leading to biased or unfair outcomes in parking violation
new data, feedback, and real-time updates. detection. Ensuring fairness and mitigating biases require careful
data curation and model evaluation.

Image Processing - Images of a parking lot can be processed to extract - Variations in lighting, image resolution, or occlusions ● Automated parking enforcement
important features like vehicle position, license plate recognition, or can have an impact on image processing techniques and the ● Parking occupancy monitoring
parking space occupancy. accuracy of parking violation detection. ● Evidence gathering and documentation
- Real-time image analysis by image processing - Algorithms for image processing may only be able to
algorithms can quickly spot unauthorized vehicles, parking violations, process visual data and may not be capable of reasoning or
and other rule violations. contextual understanding.

Computer Vision - In parking lots, cars, people, and other objects can all - In extremely crowded or cluttered parking areas, ● Automated violation detection
be detected and tracked by computer vision algorithms, allowing for where occlusions or overlapping objects can make accurate ● Parking lot management
precise monitoring of parking infractions. detection and tracking difficult, computer vision algorithms may ● Traffic flow and safety
- Computer vision techniques can analyze parking lot encounter difficulties.
layouts and spatial relationships, aiding in the identification of - Complex computer vision techniques can be
unauthorized parking or blocked access areas. computationally demanding, requiring a lot of resources and
processing power.

Optical Character Recognition - OCR technology makes it easier to identify the - OCR performance may differ depending on elements ● Automated violation detection
vehicles involved in parking violations by accurately extracting and like plate design, font choices, plate conditions, or regional variations ● Ticket processing and management
recognizing license plate numbers from images or video feeds. in license plate formats. ● Analytics and reporting
- OCR can be integrated into already-existing networks - OCR accuracy may be impacted by elements like
of security cameras or mobile applications, streamlining the collection bad image quality, dim lighting, or occlusions that obscure the
and storage of information about violations. visibility of the license plate.

Data Analytics - Large amounts of parking violation data can be - Data analytics can help with proactive enforcement ● Violation hotspot identification
 processed and analyzed using data analytics techniques, which can measures and resource allocation by identifying patterns in historical ● Performance evaluation
reveal important insights into patterns, trends, and possible areas for data and forecasting future violations. ● Demand forecasting and planning
improvement. - Data analytics provides statistical insights but may
- Data analytics can help with proactive enforcement require domain expertise to interpret and apply the results effectively
measures and resource allocation by identifying patterns in historical in the context of parking violation monitoring and enforcement.
data and forecasting future violations.

Table 2. Problem Statement.

7. From among the technologies in Table 1, which do you intend to use? Justify why it is the best choice.

Given the importance of visual input to the proposal and being at the forefront of its objective, computer vision enables the system to extract relevant data from surveillance camera images, facilitating the precise identification of vehicles, license plates, and parking violations.
Furthermore, the potential of computer vision to improve enforcement efforts, optimize resource allocation, and streamline parking management is highlighted by real-time processing capabilities, adaptability to various parking environments, and seamless integration with
current surveillance systems.
.
8. What motivates you in pursuing the study? (from “Technology” in Table 1)

To increase effectiveness, accuracy, and fairness in parking management, a study on parking violation detection using these technologies is being pursued. Given the problems surrounding parking monitoring, utilizing these cutting-edge technologies, the study seeks to create
an automated system that can quickly and accurately identify violations, lightening the load on parking authorities, and maximizing resource use. The study also contributes to the vision of building sustainable, technologically advanced urban environments, aligning with the
objectives of smart city initiatives.
9. Give the Problem Statement of your research. Use the Justification (1 and 2 above) to formulate this.

A solution is required to increase the effectiveness, accuracy, and fairness of parking violation detection given the problems with resource allocation, ineffective monitoring, and insufficient enforcement that currently plague parking management.

Table 3. List of Chosen Technology. Add more rows, if necessary.

Technology Author/Year(Cite) Description Strength Weakness Application

Computer Vision Almeida et al., 2022 A Systematic Review on Computer Deep learning techniques typically produce slightly The analysis of the datasets also showed that some Parking Spaces
Vision-Based higher accuracy results. All but one of the surveyed features, such as the availability of video sequences
Parking Lot Management Applied on approaches used deep learning-based methods for and images taken in more varied conditions, such as
Public Datasets the car detection and counting task, clearly nighttime and snow, have not been incorporated
demonstrating that the authors favor using this when creating new benchmarks.
method

Artificial Intelligence Yin et al., 2019 A Deep Learning based Illegal Parking The suggested web-based analytical platform will The suggested platform's reliance on video footage Parking Spaces
Detection Platform increase the effectiveness of law enforcement in for identification and analysis could be one possible
spotting vehicles that have been parked illegally. flaw. Factors including video quality, illumination, and
The platform improves the efficiency of vehicle camera angles can add unpredictability and could
license plate detection and offers a method to impair the accuracy of vehicle plate detection and
predict where vehicles will park by utilizing deep parking location estimation, which could have an
learning algorithms and video analysis. This impact on the platform's performance.
provides a more automated and effective method of
managing car parking, especially in big, developing
cities.
Machine Learning Ng, 2019 Deep Learning-Based Parking On a single board computer, a Convolutional Neural The suggested system's reliance on the availability Parking Spaces
Detection System Using Structural Network (CNN) classifier is used to categorize and caliber of video data taken by IP cameras is one

Similarity parking space occupancy and identify automobiles of its weaknesses. Parking occupancy detection and
that are parked unlawfully. The use of unlawful parking identification accuracy and reliability
multithreading techniques and the addition of a can be affected by variables like video resolution,
revolutionary structural similarity (SSIM) decision camera angles, and weather conditions.
scheme both significantly optimize the system's
performance and quicken the detection procedures.

Image Processing Huang et al., 2022 Visual Detection and Image Processing The difficulty of complicated picture information and The proposed method's dependency on the Parking Space
of Parking Space Based on Deep uneven lighting in autonomous parking systems is availability and quality of the 360-degree panoramic Vehicle Security
Learning addressed by the deep learning-based visual photographs taken by the devised system is a
recognition and image processing method for possible limitation. The precision of parking space
parking spaces. The parking spot identification recognition may be impacted by any restrictions or
system receives a thorough input from the 360- distortions in the panoramic photography process. In
degree panorama system, which records the order to guarantee the method's robustness and
surrounding area of the car and creates a generalizability, further testing of the method's
panoramic aerial view. performance and efficacy in various real-world
settings and environments may be necessary.

Optical Character Recognition Bassam & Samann, Smart Parking System based on The upgraded Optical Character Recognition (OCR) The suggested system's reliance on the installed Parking Space
2020 Improved OCR Model model-based smart parking system that is being camera mounted above the parking spaces for
suggested responds to the growing demand for precise vacant parking slot recognition is a potential
effective parking space use in metropolitan areas. drawback. The system's capacity to accurately
The technology saves time and fuel by precisely identify and convey parking availability may be
identifying vacant parking spaces using an installed impacted by elements including camera angles,
camera and the OCR model, and it also informs lighting, and potential impediments.
vehicles in real-time about parking availability.

Data Analytics Gao & Ozbay, 2017 A Data-driven Approach to Predict Using a case study of New York City, the paper The study's dependence on particular data sources, Parking Spaces
Double Parking Events Using Machine provides a novel data-driven methodology for like parking tickets, 311 service calls, and data from Law Enforcement
Learning Techniques comprehending and forecasting the incidence of social media, could be one of its weaknesses. These
double parking in crowded urban locations. The data's accessibility and accuracy may vary depending
system offers a thorough method for assessing and on the area and may not fully account for instances of
forecasting double parking occurrences by merging double parking. The study's emphasis on New York
several data sources, such as parking infraction City as a case study might make it harder to
tickets, 311 service requests, and social media generalize its conclusions to other cities with diverse
information with street features. To discover double parking patterns and characteristics. Although
significant factors and reach a high prediction the framework obtains a high prediction accuracy of
 accuracy of 85%, three feature selection 85%, it is important to take into account any potential
approaches are applied, and Random Forests is drawbacks linked to the temporal and spatial
used as a machine learning methodology. variability of double parking incidents.

Table 4. Proposed Model/System to Chosen Technology. The proposed modification must be unique and different from those in Table 5.
9. From the technology identified in Table 3 (whichever is applicable), what particular technology are you going to use?

Due to machine learning's outstanding capacity to identify patterns and intricate relationships within data, it has been decided that this method is the best one for monitoring parking violations. Machine learning models' versatility allows them to continuously learn from fresh data,
assuring their effectiveness in identifying changing parking violation instances. Additionally, while maintaining high accuracy and reducing false positives, machine learning approaches can be scaled to handle large-scale parking spaces.

10. Will the propose study provide solution to the research problem as identified in Table 2? How?

Yes, the suggested study seeks to use image recognition technology for parking infraction detection to address the research problem. The goal of the study is to improve parking violation detection's efficiency, accuracy, and fairness by utilizing image recognition and machine
learning approaches. The work intends to increase monitoring capacities and enable more effective enforcement by building an image recognition algorithm on annotated data.
11. Write down your research statement – that your propose solution (Table 2 above) will solve the identified problem.

The project aims to address resource allocation concerns and improve the overall management of parking offenses by adopting real-time detection and alarm systems based on the image recognition model.
12. What Is the proposed application for your technology? What will be the final output (model, system, etc.)? Describe.

A parking violation detection system based on image recognition and machine learning is the technology's proposed application. A trained image recognition model that is capable of reliably identifying and categorizing parking offenses in real-time will be the system's ultimate
output. The system will evaluate visual data provided from security cameras or other sources and apply the trained model to it in order to find problems like illegal parking, parking in prohibited locations, or expired parking meters. The system's output will often include alerts or
messages to parking management systems or enforcement officers, giving them knowledge of the place, time, and kind of violations.