Project Background

Radio frequency identification devices (RFID) can be scanned invisibly, and these devices can contain private data. This has triggered concerns that personal information could fall into the wrong hands. Although RFID devices are convenient and widely used in many ways, security poses a concern. For instance, using an outdated RFID reader and old databases could allow attackers to hack the RFID readers and gain unauthorised entry.

By using SQL, this proof-of-concept seeks to demonstrate how RFID readers could be used to test the strength of existing RFID security solutions and from there, raise awareness of how security could be beefed up, such as using up-to-date RFID readers or updating the databases.

Key Benefits

• Raising Awareness - Raise awareness of the need to strengthen the RFID database or reader

Key Technology Used

• Python 3.0
• MariaDB
• libnfc
• Dongmin Kim’s ACS-ACR122U-Tool
  (Open Source Python tool for ACR122 NFC Reader/Writer)

Potential Opportunities

This proof-of-concept has the potential of becoming a means to test existing security solutions for vulnerabilities. Organisations can also use this information to increase the robustness of their security solutions.

Project By

• Tan Ying Jie, Dennis

Supervisor

Mr Lui Tien Heong Willie

Project Background

USB storage devices are commonly used due to their convenience and size, but they also pose information security risks if they have been tampered with. They can be used to spread malicious programmes and even connect to an organisation’s network to steal data. Even though antivirus scans can help detect and get rid of malicious programmes in USB devices, they are not foolproof.

The USB Sheep Dip is a miniature computer configured with multiple antivirus software specifically for scanning USB devices. A potentially malicious USB device can be plugged in and analysed for any malware or suspicious programme. It is also able to detect tampered USB devices. This allows it to be used as a portable front line defence in many different environments.

This project was submitted for the Tan Kah Kee Young Inventors’ Award 2017.

Key Benefits

• Powerful Detection - Identifies and detects malicious USB Human Interface Devices
• Frontline Defence - Acts as a first line of defense against USB devices
• Increased Security - Sanitises file transfers via USB devices.
• Increased Safety - Removes potential loss of data via USB devices

Key Technology Used

• C / C++
• Windows API
• OS Integrity Tools

Potential Opportunities

This solution can be enhanced with the following:

• Incorporated into any organisation’s network
• Converted into an automated solution

Project By

• Eyo Wei Chin

Supervisor

Mr Ng Cheng Tiong Jason

Project Background

Your computer has a virus! These days, there are multiple antivirus software in the market for consumers to purchase and download. Which antivirus software should you use? With so many choices, the question remains, which one provides the best protection against computer virus threats and malware?

“AntiVirus Avoider” is an encryption tool developed specifically to encrypt malware that bypass anti-virus detection. Through this tool, an antivirus software can be tested to see how well it detects encrypted malware installed in a computer.

Key Benefits

• Determine Antivirus Capability - Find out which antivirus software is better able to detect malware
• Simulate Cyber Attacks - Encrypt different malware for simulated cyber attacks
• Store Malware Safely - Avoid detection by antivirus software

Key Technology Used

• C# Programming Language
• Batch Script
• Cryptography
• MSBuild

Potential Enhancements

This solution can be enhanced with the following:

• Implementation of database
• Improvement to C++ programming language
• Change to internal decrypter
• Use of stronger encryption

Project Team Members

• Tham Chi Sheng, Gary (Project Leader)
• Nurul Huda Bte Musthapa

Supervisor

Mr Tan Teck Chai, Shaun

Project Background

Binary analysis, a process that is widely used to audit computers and detect potential malware, is a manual and tedious process. However, with machine learning advancing significantly in recent years, binary analysis tasks could be changed to make the task less time-consuming and more effective.

To do this, the Binary Function Classifier (BFC) was developed, to make the task of binary analysis less tedious. BFC generates the name of functions and their binary features without the need to analyse source codes, thus speeding up the binary analysis process significantly.

Key Benefits

• Accurate Predictions- Binary functions are classified into the various categories with an accuracy rate of 91.4%
• Improved Productivity- Binary functions are classified at about 50 functions per minute without the need for analysing source codes

Key Technology Used

• Ubuntu 14.04
• IDA Pro 6.8
• Scikit-Learn
• Rendezvous

Project By

• Yuen Si Hao

Industry Partner

DSO National Laboratories (DSO)

Supervisor

Dr Yap Chern Nam

Project Background

Dynamic malware analysis requires a security analyst to manually do a threat hunt on a virtualised environment. This takes up valuable time as there are many ways in which modern malware can keep itself hidden in a computer.

With Clairvoyance, the investigation process and time is greatly shortened as any changes and actions on a machine (files, registry, services, processes, network) would be recorded in a timeline. It would also actively monitor for threats, and upload any files that could be malicious to VirusTotal for a virus scan. Besides this, any signs of a programme trying to add itself to the system start-up would also be detected and flagged immediately, allowing investigators to perform their analysis.

Key Benefits

• Full System Coverage - Changes to any part of the system would be recorded
• Automatic Threat Hunt - Suspicious activities would be flagged
• Data Visualisation - Significant events would be plotted on a timeline and changes are represented on a pie-chart giving the analyst a quick overview of what is going on

Key Technologies Used

• Windows Management Instrumentation (WMI)
• Microsoft .NET Framework 4.62
• C#
• Python
• Batch Scripting

Potential Enhancements

This application can be enhanced with the following:

• Machine learning algorithm to more precisely identify the threat type
• Network analytics capability

Project By

• Chan Jian Hao

Industry Partner

ST Electronics (Info-Security)

Supervisor

Mr Tan Teck Chai, Shaun

Project Background

Computer peripherals like USB storage devices and Human Interface Devices (HIDs) like keyboards, mice, game controllers, are vulnerable to hackers. For instance, USB storage devices could be hacked to inject keystrokes that emulate keyboards when plugged into a computer.

This project demonstrates how USB HIDs could be hacked. Connecting the FireGoose* device to a computer, a malicious code could be uploaded and the FireGoose device would mimic keystrokes typed into a host computer. Using information from the key logging script, security solutions could be explored in order to develop robust solutions against such attacks.

*FireGoose is a HID device that simulates how malicious code could invade a computer, enabling users to identify their system’s weaknesses.

Key Benefits

• Versatile - Limited only by the boundaries of a keyboard
• Easy to Use - Plug and play solution

Key Technologies Used

• Arduino
• PowerShell

Potential Enhancements

This application can be enhanced with the following:

• Made compatible for different operating systems
• Integrated with a network sniffer tool

Project By

• Ryan Tan Kian An

Supervisor

Mr Ng Cheng Tiong, Jason

Project Background

Polymorphic malware such as virus, worm, Trojan or spyware are malicious intrusive software that are hard for anti-malware programmes to detect, leaving organisations open to cyber security breaches. To solve this problem, the technology of deep learning is used as it has a greater capacity to detect, identify and classify malware to a higher degree of accuracy.

Through the incorporation of deep learning techniques and classification algorithms, the Malware Classification System is able to classify malware datasets and their respective malware families. When incorporated into anti-malware programmes, the detection and identification of malware would be greatly enhanced.

Key Benefits

• Data Visualisation - Data is plotted on graphs to show a result of the classification
• Modular - Users can choose which classification method they desire
• Cross-platform - Executable across various platforms

Key Technologies Used

• C++
• Miniconda
• Python 2.7
  • Sci-kit learn 0.16.1
  • Numpy 1.9.2
  • Scipy 0.15.1
  • Matplotlib 2.0.0

Potential Opportunities

This application can be enhanced with the following:

• Integration with other classification methods and deep learning techniques
• Integration with anti-malware programmes or Security Operation Centres

Project Team Members

• Muhamad Afiq B Busari
• Wan Teng Chew Spencer

Supervisor

Mr Ng Cheng Tiong, Jason

Project Background

Security configuration audits are important as they protect a company’s system from hackers. Existing commercial tools require privileged administrative access, making them unsuitable for third-party configuration audits as companies would be exposing their sensitive data to outsiders. In addition, existing configuration audit tools also cannot effectively audit web-based management consoles, such as those which use Internet of Things (IoT) products.

PiloT is a configuration audit tool which allows for configuration audits to be done offline as well as from the saved pages from web-based management consoles, removing the need for outsiders to gain access to a company’s system. This method not only protects a company’s sensitive data, it provides an efficient way of auditing IoT products without using up too many resources.

Key Benefits

• Efficient - Increased productivity for auditors
• Unique - Fills gap in the market for offline configuration auditing tools
• Control - Access is controlled for system administrators

Key Technologies Used

• Java & JavaFX
• POI
• Cling Core
• HTMLUnit

Potential Enhancements

This application can be enhanced with the following:

• Automated auditing for more UNIX systems
• Testing Universal Plug and Play-enabled devices for security flaws

Project Team Members

• Tan De Yi

Industry Partner

Resolvo Systems Pte Ltd

Supervisor

Mr Tan Teck Chai, Shaun

Project Background

Most customers are not proactively monitoring the log data in their computer’s operating system as log syntax is difficult to understand. In addition, there are challenges in consolidating and visualising such data efficiently. Since there is no proactive monitoring of a computer’s log data, threats and abnormal events are not effectively flagged and the potential of the data is not maximised.

However, if organised in a meaningful way, log data can yield insights that are practical, usable and accessible, making the detection of threats more effective. Thus, the aim of Project Splash is to provide a platform for companies to achieve better log management and to reap the benefits of it.

Key Benefits

• Data Visualisation - Provides a platform to consolidate and visualise log data
• Informative - Creates useful insights for analysis
• Useful - Performs threat analysis for anomaly detection

Key Technologies Used

• Amazon Web Services (Lambda, EC2, S3, CloudFront, CloudWatch, VPC)
• ELK Stack (Elasticsearch, Logstash, Kibana)
• Infrastructure software (Lovebeat, Cowrie, Glastopf)
• Pentesting Tools (XSS-ME, SQL Inject me, DotDotPwn, Nikto, Nmap)

Potential Enhancements

This application can be enhanced with the following:

• Display of past year’s progress
• Checking for lags in any departments in the company
• Creation of useful insights for non-technical staff

Project Team Members

• Umar Mokhtar Ayob Putera (Project Leader)
• Voon Sze Hwee

Industry Partner

Banff Cyber Technologies

Supervisor

Mr Chan Yan Jau, Joe

Project Background

In the realm of digital forensics, additional tools are becoming increasingly necessary to ensure that the integrity of collected evidence remains intact.

Working like a Faraday* cage, the Radio Frequency Shielding Box was developed to block out all phone signals, so as to prevent the loss of evidence and to maintain the integrity of the evidence collected. This Box would aid computer forensics investigators in their seizure, transportation and examination of portable digital devices without tainting the evidence.

*A Faraday cage is an enclosure used to block electromagnetic fields.

Key Benefits

• Preserve Device Integrity - Ensures that a device’s information is carefully preserved throughout its acquisition process
• Reduce Investigation Time - Reduces investigation time by allowing for convenient extraction of information from confiscated electronic devices

Key Technologies Used

• Bash Scripting
• Python
• EAGLE Software

Potential Opportunities

This application can be enhanced with the following:

• Collaboration with mobile forensic firms to integrate other extraction software
• Use of Cellebrite’s UFED to extract both iOS and Android devices

Project By

• Wesley Sim Qian Dong

Industry Partner

Central Narcotics Bureau

Supervisors

Dr Yap Chern Nam

Mr Lim Chee Yong

Project Background

Most users of smart devices today do not know which programmes or applications are accessing their data without permission. During a programme installation process, administrator rights have to be given to a software for it to make necessary changes to the computer. In the process, users might have unwittingly exposed their personal data.

Personal Data Safeguard is a project that aims to return control of private data to the users. Users have access to a platform where they are able to view and investigate what has happened to their data with the use of a provenance logger (Progger) and Titan database.

Key Benefits

• Informative - Allows users to know what is happening to their data
• Reliable - Suitable for corporate entities to use in internal audits
• User-friendly - Neat and user-friendly data visualisation

Key Technologies Used

• Titan Database
• Progger (Provenance Logger)
• Gremlin Graph Traversal Language
• Gephi
• JAVA

Potential Opportunities

This application can be enhanced with the following:

• A fully integrated front-end for data visualisation
• An improved method of data visualisation
• An integration with ADAMS for the creation of dynamic modular queries

Project By

• Tan Boon Siang

Supervisors

Mr Tan Teck Chai, Shaun

Project Background

Online scams are on the rise in Singapore. According to The Straits Times, the number of scam victims more than doubled in a year in 2015. Most common amongst these scams are credit-for-sex scams, love scams, kidnap hoax scams and lottery scams. In 2016, about 150 people lost over $12 million in the DHL phone scam alone. There is a need to raise awareness so that less will fall victim.

ScamDefender aims to be a one-stop portal for all things scam-related to keep users better informed and safe from scams. It has scam prevention features such as ‘Newsfeed’ to keep users up-to-date on the latest scams, ‘Report a Scam’ so the community could be notified of new scams, and user data analysis (e.g. contact number, number of good and bad reviews) to produce an accurate user trust rating.

This project was submitted for Best Innovative ICM Student Project (Tertiary) at SiTF Awards 2016.

Key Benefits

• Early Alert - Early detection and warning against potential scams
• Trustworthy Sources - Generation of accurate trust ratings of users
• Easy Migration - Easy migration across e-commerce websites
• Victim Reduction - Reduction on the impact of widespread scams

Key Technologies Used

• Natural Language Processing (NLP)
• PHP and PostgreSQL
• Ionic Framework
• AngularJS

Potential Opportunities

This application can be enhanced with the following:

• Big data analysis
• Integration with e-commerce websites

Project Team Members

• Eyo Wei Chin (Project Leader)
• Ong De Zhi
• Sean Yap Rong En
• Seng Jun Yi

Supervisors

Mr Lim Kim Seng, Jet

Project Background

The work carried out by forensic first responders are important. Their prompt and correct handling of a computer network incident allows it to be contained and investigated. Their actions could adversely affect the forensic value of any data they collect. Hence, their effectiveness when responding to security alerts is crucial when it comes to protecting network infrastructures.

The aim of Owl PE is to aid forensic first responders to swiftly identify suspect files for further investigation. It allows them to carry out their work in identifying potential malware infections effectively, despite not having malware analysis resources as part of the incident response team’s organic capability.

Key Benefits

• Flexible - Modular approach to choose evaluation functions
• Effective - Detects malicious Portable Executable (PE) files effectively
• Efficient - Utilises an efficient algorithm to conduct string search
• Swift - Allows first responders of a Cyber Incident Response team to swiftly determine suspicious files for further investigation

Key Technologies Used

• C & C++
• wxPython

Potential Opportunities

This application can be enhanced with the following:

• Incorporation of entropy mapping analysis
• Expansion of heuristics

Project Team Member

• Patrick Kang Wei Sheng (Project Leader)
• Wong Wei Sheng

Supervisors

Ng Cheng Tiong Jason

Project Background

This Autonomous Litter Detection Vehicle is a free-roaming robot that can help to detect litter. With its intelligent programming, it is able to move around on its own in an enclosed area. The integration of machine learning and sensors allow the vehicle to detect and recognise litter such as pieces of tissue on the floor.

Key Benefits

• Reduces Effort - Lessens the need for cleaners to patrol places that are already clean and litter-free

Key Technology Used

• Ubuntu Linux
• Arduino Software (IDE)
• Robot Operating System (ROS)
• Nvidia DIGITS Machine Learning

Potential Enhancement

This application can be enhanced with the following:

• Send SMS notifications to cleaning staff
• Automatically pick up litter

Project Team Members

• Muhammad Azrul B Mohd Sanif (Project Leader)
• Muhammad Riduwan B Zainudin
• Raymond How

Industry Partner

Nvidia Corporation

Supervisor

Mr Tan Kok Hui
Mr Tan Sio Poh

Project Background

DIVA is a conversational bot which can provide information on courses and facilities in the School of Informatics & IT,  allowing visitors to learn about the School in an interactive and fun way.

DIVA is intelligent enough to understand different types of questions posed and provides answers to the questions posed. This smart bot is implemented with machine learning, a type of Artificial Intelligence (AI) that provides computers with the ability to learn without being explicitly programmed.

Key Benefits

• Interactive Solutions – Provides an interactive and fun way for visitors to know more about the School of Informatics & IT and its courses

Key Technology Used

• Microsoft bot framework
• Language Understanding Intelligent Service (LUIS) AI
• JavaScript
• Html and CSS
• Microsoft Azure Services
• Node.js

Potential Enhancement

This application can be enhanced with the following:

• Interactive Solutions – Provides an interactive and fun way for visitors to know more about the School of Informatics & IT and its courses

Project Team Members

• Huang Bo Hao
• Parmesher Singh
• Sin Jun Jie

Industry Partner

Nvidia Corporation

Supervisor

Mr David Wang
Mr Tan Sio Poh

Project Background

Have you always wanted your devices to automatically monitor the lighting or air conditioning in a room to conserve energy usage? With the Energy Conservation using IoT sensor, now you can.

This project involves the use of several portable sensors placed strategically to collect data such as temperature, humidity and noise levels in any room. The sensors then alert the user of the physical conditions in the various rooms and to take action to conserve energy. The user can then decide to adjust the thermostat of the air-conditioner if it is too cold or lower the blinds if the room is too bright.

The portable sensors also assist in energy conservation as the user is alerted if the air-conditioner is switched on even though the room is vacant.

Key Benefits

• Energy efficient – Reduces energy wastage
• Cost-saving – Lowers cost of energy consumption
• Automated – Data is automatically collected using sensors so that users can stay updated on room conditions

Key Technology Used

• Sens’it sensors
• Sumo Logic
• SIGFOX network

Potential Enhancement

This application can be enhanced with the following:

• Turn off the lights and air-conditioners automatically when not in use
• Set air-conditioners to the ideal temperature
• Control lights and air-conditioners remotely

Project Team Members

• Melisa Hariyanto
• Phee Jing Yuan

Industry Partner

UnaBiz Network Pte Ltd

Supervisor

Ms Lam Su Yi

Project Background

Would you like to explore your home in virtual reality? MagixHome, is a virtual reality interior design game. This game allows players to explore the interior and exterior room designs, including the surrounding environment, in order to make better decisions when designing their spaces.

In this game, players focus on exploring the in-game world without engaging in any active conflict like combat or stealth. This game allows players to focus on the game’s architecture, aesthetics and atmosphere.

Key Benefits

• Innovative – Introduces an exploration element to the interior design process
• Efficient – Allows designers to understand their environment before designing starts

Key Technology Used

• Unreal Engine 4
• HTC VR (Virtual Reality)
• Photorealistic 3D modeling

Potential Enhancement

This application can be enhanced with the following:

• Improve vehicular controls
• Enhance audio feedback for players
• Include performance support for lower-end users
• Upgrade day/night cycle and weather system

Project Team Members

• Wong Qin Liang (Project Leader)
• Benedict Low Ren Jie
• Muhammad Fadhil Bin Khamsin
• Teo Wei Shan
• Yap Zhan Jie

Industry Partner

SKY Optimum Technology

Supervisor

Mr Tan Hong Yap

Project Background

The bathroom is a high-danger zone for the elderly, where the slippery floors and small space increase the likelihood of falling. An early detection of the fall and quick rescue time can minimise potentially life-endangering injuries. With the Fall Detection System (Fall-D), you can quickly be notified if your loved one has fallen and can immediately seek help.

The Fall Detection System (Fall-D) is an Internet-of-Things solution comprising a Hexiwear wearable device, sensors and a mobile application.

The elderly user would use the Hexiwear device, while the Arduino-controlled motion and door sensors are affixed in the bathroom. An additional voice recognition sensor on a Raspberry Pi is also located in the bathroom to detect the elderly user’s voice. The system is activated when the sensor at the doorway detects someone entering the bathroom. If the elderly user falls, there are 4 ways it can be detected:

• Sudden acceleration recorded on the Hexiwear
• No motion detected in the bathroom
• No exiting motion detected after a period of time
• Voice command calling for help

An alert is immediately sent to the next-of-kin using the mobile application. Hence, quick action can be taken to save the elderly user’s life.

Key Benefits

• Cost Effective – The complete solution is affordable
• No Invasion of Privacy – The sensors are able to detect the fall with no need for intrusive cameras
• Specialised System – As the components are small, the solution can be customised to suit the given space

Key Technology Used

• Arduino
• Raspberry Pi
• Python
• Cordova

Potential Enhancement

This application can be enhanced with the following:

• Built-in network to transmit data even if the household does not have internet access
• Tailored to work in any area of the house and for any audience
• Utilised as an anti-theft solution to alert in case of burglaries

Project Team Members

• Benedict Sim Jun Hong (Project Leader)
• Cindy Leong XinYi
• H R Shafana Thasneem
• Halima Nabisathul Basariya
• Ramya Amirthalingam

Industry Partner

Microsoft Singapore

Supervisor

Mr Jackson Ng

Project Background

Can you see yourself as an immune system cell unit in an alien’s body?
In Cellulose, you can!

Taking the role of an immune system cell, you get to chase and eliminate viruses which are trying to infect the alien’s brain where you reside. Experience the thrill of a chase and pursuit game in the unique context of a microscopic environment in the circulatory system.

This Real Time Strategy game allows you to play the role of a commanding immune system cell that can produce numerous child cells, and then order them to chase down and defeat an invading virus.

Key Benefits

• Convenient - Play a Real Time Strategy game on a mobile phone
• Efficient - Low battery consumption
• Versatile - Runs on a range of devices

Key Technology Used

• Finite state machine (FSM)
• Hierarchical AI
• A* pathfinding
• Flocking behaviour
• Unity

Potential Enhancement

This application can be enhanced with the following:

• Incorporate realistic medical simulation of viruses
• Improve the AI’s real-time learning system
• Include online modes of playing the game (Co-Op & PvP)

Project Team Members

• Xu Jie (Project Leader)
• Chong Nian Kai
• Tan Xue Wen, Rayner
• Wong Qin Liang

Award

SiTF Best Innovative Award

Supervisor

Mr Tan Hong Yap

Project Background

Is your choice of lunch suitable for your health condition? Let FoodCase analyse a picture of the meal and make recommendations tailored to your health profile. Share pictures of your nutritious lunch with your friends to encourage one another to eat healthily.

FoodCase is a mobile application that can help you make more informed decisions on your food choices. It is an intuitive nutrition advisor that can identify food from images that you take using your phone. Using machine learning and artificial intelligence, FoodCase analyses the images and provides the nutritional information of the dish. It can also give intelligent recommendations to the user based on his/her health profile and location. FoodCase includes an integrated social networking element that allows you to keep updated with what your friends are eating.

Key Benefits

• Convenient – Food recognition from images is fast and automated so users do not have to manually input details about their meals
• Customised – Health recommendations are provided based on the user’s unique health profile and dietary requirements

Key Technology Used

• Computer Vision API
• Cordova

Potential Enhancement

This application can be enhanced with the following:

• Biometric API for user login
• Text analytics to identify keywords in user’s health profile
• Partnership with food delivery services to recommend healthy dishes that can be delivered to the user

Project Team Members

• Wong Jian Hui (Project Leader)
• Koh Thiam Chye, Raymond
• Lye Jun Wei, Ananda
• Nurul Hidayati Bte Mustazam

Industry Partner

Microsoft Singapore

Supervisor

Mr Mark Kor Chan Chiong

Project Background

Educating children about good posture just became fun! Posturiser is a 2D mobile game which uses a holistic approach to raise children’s awareness of good posture. While playing the game, fun facts and hints will appear to educate a player about maintaining a good posture.

The game comprises multiple mini-games. Play time is short to maximise player retention. Each mini-game has a subtle message regarding good or poor posture in daily situations.

Key Benefits

• Educational - Raises awareness for good posture
• Engaging - Key messages are presented in a fun manner
• Fun - Makes learning enjoyable and relaxing

Key Technology Used

• Unity
• Adobe Photoshop

Potential Enhancement

This application can be enhanced with the following:

• Modify the game to raise awareness about other health issues
• Include a leaderboard which is linked to a social platform

Project Team Members

• Sarvesh s/o Panirselvam (Project Leader)
• Calvin Lee Jian Wei
• Chan Jia Hur Galvin
• Chew Boon Yang Kenneth
• Leong Wei Ren
• Li Jing Xin
• Tam Chee Jian William

Industry Partner

KK Women’s and Children’s Hospital

Supervisor

Mr Douglas Finnigan

Project Background

Mirror Mirror on the Wall is a fun and entertaining face recognition programme. An easy to use application, Mirror Mirror on the Wall requires a user to stand in front of a webcam and take a picture. It would smartly guess a visitor’s estimated age, gender as well as mood. This is done using face recognition technology.

One way this application could be used is to capture patrons’ faces in a shopping centre where data about their age, gender and mood  group could be used for marketing purposes. Another possible use is for this application to be used in a hotel where the concierge could tell the customers' mood and can provide a customised meet-and-greet service. It could also be used in a store where customers’ age could be identified so that stores avoid selling alcohol or contraband to minors.

Key Benefits

• Engages customers with this interactive application
• Understands customers’ facial expression and age group
• Brightens up visitors’ day as the application will prompt them to smile

Key Technology Used

• Microsoft Face API
• C# programming

Potential Enhancement

This application can be enhanced with the following:

• Used as an interactive tool to gauge customers’ satisfaction
• Utilised by classroom or seminar speakers to understand their audience’s reception of the content

Project Team Members

• Huang Bo Hao
• Parmesher Singh
• Sin Jun Jie

Supervisor

Mr David Wang
Mr Tan Sio Poh

Project Background

Smart Classroom is an intelligent application that allows users to turn the lights on and off through an intelligent remote application remotely. This saves users the hassle of physically entering the rooms to turn the lights on and off, because now they can do so at the touch of their fingertips!

The Smart Classroom deploys automation products from Anjels Pte Ltd in order to have a centralised control of lights and electrical appliances. It is equipped with an access point that communicates with the intelligent remote application to enable users to control lights and electrical appliances in designated rooms.

Key Benefits

• Energy efficient – Reduces energy consumption
• Convenient – A one-stop control for appliances

Key Technology Used

• Anjels iNet Node
• Anjels i-Top Switches
• Linksys Cisco Router
• Wi-Fi Technology

Potential Enhancement

This application can be enhanced with the following:

• Implement energy saving light bulbs

Project Team Members

• Humairah Binte Abduh
• Nur Alwani Bte Akhbar

Industry Partner

Anjels Pte Ltd

Supervisor

Mr Ngeow Teck Liang, Simon

Project Background

The Smart Distribution Board (DB) leverages the Internet of Things and cloud-based technologies to enable smart home management. By placing sensors into the electrical distribution board that every household has, data from the sensors is transmitted to a mobile application.

In this way, households have a way of determining where, when and how much electricity is being used by each appliance in a household so as to save energy and monitor lifestyle habits. Available in both mobile and web formats, the Smart DB hopes to inculcate responsible electricity usage behavior.

Key Benefits

• Ease of Use – Installation is centralised, cost efficient and requires no maintenance
• Real Time Data – Real time data is captured and analysed to derive historical data and cost of electricity usage
• Safe Home Environment – Electrical safety is monitored

Key Technology Used

• Microsoft Azure Web Apps Service
• Microsoft Azure SQL
• Android Studio with Android Software Development Kit (SDK)
• SQLite
• Visual Studio with .NET Core Framework
• Socket Programming
• RStudio

Potential Enhancement

This application can be enhanced with the following:

• Automated identification of appliances without guided training

Project Team Members

• Ang Zhi Kai
• Isaac Ibrahim

Industry Partner

• Clean Energy Research Centre (CERC), School of Engineering, Temasek Polytechnic
• Housing and Development Board (HDB)
• Ministry of Education (MOE)
• Smart DB Pte Ltd

Supervisor

Dr Chia Chaw Poh

Project Background

The Smart Pond System is developed in collaboration with Changi Airport Group (CAG). This system automatically monitors water quality and senses water levels of Koi fish ponds within the airport. The system can also trigger notifications based on predetermined sensor thresholds, and includes an automated fish food dispensing system.

This solution comprises an Internet of Things (IoT) device as well as a mobile application which allows users to actively monitor pond conditions and pre-empt possible problems that may occur. This would help to speed up the response that is required to improve the conditions of the pond.

Key Benefits

• Automated: Fish food is automatically dispensed twice a day
• Convenient: Easily monitor various parameters of the water such as the pH level, temperature, oxygen and salinity
• Time-saving: Notifications sent to staff can help to trigger quick action to improve water conditions

Key Technology Used

• Arduino Mega, Arduino Yun, Arduino Uno and LinkIt ONE
• Grove system (Seeedstudio)
• Ubidots Cloud Platform
• Atlas Sensor Circuits
• Inter-Integrated Circuit Communication

Potential Enhancement

This application can be enhanced and applied to aquariums and fish farms.

Project Team Members

• Baratharam s/o Manogaran
• Kui Kai Ting
• Michael Tanri
• Stanley Ng Wee Tee
• Wei Minn

Industry Partner

Changi Airport Group (CAG)

Supervisor

Ms Andrea Cristina Ng

Project Background

Set in the mystical world, the mighty god, Zune, joins forces with a risen hero to save humanity from the evil Kroden. To save the world, they have to prevent incoming armies from breaching the shrine that seals Kroden’s power. Failure would spell the end for humanity.

Tower Power is a Virtual Reality (VR), first-person, co-op tower defence game. Play the role of Zune and place towers to strategically defend against your enemies. Alternatively, you can be the Hero on the ground fighting through hordes of enemies with your ever-powerful sword and shield. Work together to win! For the kingdom!

Key Benefits

• Entertaining – Fun and immersive VR experience
• Customisable – Two play styles to cater to first-person shooter, and strategy-loving players
• Social – Unique multiplayer VR experience

Key Technology Used

• Unity
• Autodesk Maya
• Adobe Photoshop
• HTC Vive

Potential Enhancement

This application can be enhanced with the following:

• Include other game modes such as Offline Single Player Mode or Party Mode
• Improve combat and strategy game systems

Project Team Members

• Tan Xue Wen, Rayner (Project Leader)
• Aaron Tan Jian Rong
• Chong Nian Kai
• Clement Kang Jie
• Jonathan Leong Jin Young
• Low Wai Leong

Supervisor

Mr Jonathan Pillai
Mr Tan Hong Yap

Project Background

Want to know more about TP’s latest news and events? Or explore the campus to find hidden spots to relax? The TP Interactive Mobile Application is for students and visitors looking to find out more about life in TP. It features interactive elements such as geo-tagged places of interest, location-based notifications as well as a chat-bot to clue you in on our diploma courses.

The application also includes a Smart Library where students can browse the catalogue of books available at TP’s Library. Not sure where the book you want is located in the Library? Fret not, as the app will navigate you to the correct section.

Key Benefits

• Interactive – Chat with the integrated chat-bot to find out more about TP
• Customised – Receive location-tailored notifications about nearby places of interest

Key Technology Used

• Microsoft Cognitive Services
• Microsoft Azure SQL and MySQL
• Microsoft Azure Virtual Machines on Ubuntu Linux
• Microsoft Bot Framework with Natural Language Processing (NLP)
• Estimote Bluetooth Low-Energy Beacons
• A*Star Pathfinding Algorithms
• Trilateration Algorithms
• Android
• Node.js
• PHP

Potential Enhancement

This application can be enhanced with the following:

• More training of the Natural Language Processing platform to improve the effectiveness of the chat-bot
• Additional beacons to increase indoor location accuracy

Project Team Members

• Tjai Yao Long, Shawn

Industry Partner

• Microsoft Singapore
• Temasek Polytechnic Library

Supervisor

Mr Yeak Shaw Wen

Project Background

NotifyNow is a web-based application where it is designed to assist the Legal and Compliance Department in the Government of Singapore Investment Corporation to automate the process of tracking whether staff have completed a quiz about legal and compliance requirements. This application sends email reminders to staff to complete the quiz and automates the daily reporting of staff who have not completed the quiz. Data about the completion of the quiz is presented in a dashboard via graphical charts for the administrators’ reference.

Key Benefits

• Efficiency – Improved process flow eliminates human error
• Data visualisation and management – Dashboard and graphical charts present data in highly visual formats to aid in tracking

Key Technology Used

• C#
• jQuery
• Bootstrap
• Highcharts
• JavaScript

Potential Enhancement

This application can be enhanced with the following:

• Enable interaction with other systems to ensure automated data entry
• Enable tracking of financial reports

Project By

• Cao Jiewei

Industry Partner

GIC Private Limited

Supervisor

Mr Lim Beng Kee, Ryan

Project Background

Initiated by Channel NewsAsia, “Future of Work” is a project which uses the Virtual Reality (VR) technology to envision how the advancement of technology can impact the way we work in the future.

Using estate planning as the context of this serious game, this potential software can help Estate Developers visualise and plan their estates. Features of the simulation include the construction of apartment blocks, selection of amenities, and placement of street lights in an estate. Instant feedback is also provided through the use of Big Data.

Key Benefits

• Realistic – Ability to envision the future living environment of Singapore
• Innovative - Use of VR Technology on a serious game

Key Technology Used

• Unity
• Autodesk Maya
• HTC Vive

Potential Enhancement

This application can be enhanced with the following:

• Include other genre-defining features to appeal to different audiences
• Incorporate big data analytics to provide more realistic simulations

Project Team Members

• Ambrose Wee Chong Yin (Project Leader)
• Eugene Tan Jianpeng
• Jared Koh Fu Jie
• Lim Jia Wei
• Tan Yong Kiat, Joshua

Industry Partner

Channel NewsAsia

Supervisor

Mr Jonathan Pillai
Mr Tan Hong Yap

Project Background

Voice-Clarity is an amazing audio signal processing mobile application for any android device which aims to reduce the background noise so as to improve the clarity of the speaker’s voice. This is especially critical in areas with high background noise, such as live conferences. The application suppresses the background noise in real time while the speaker is speaking, thus providing greater clarity for listeners.

This project establishes the foundation of signal processing for researchers and developers to build innovative Android applications.

Key Benefits

• Sound Clarity – Ensures the clarity of the speaker’s voice through noise reduction capabilities
• Removal of Unwanted Data – Enables the removal of unwanted raw data through the digital signal processing processing feature for Android

Key Technology Used

• Processing 3 IDE - Java Language
• Android programming

Potential Enhancement

This application can be enhanced with the following:

• Implementation for outgoing mobile phone calls
• Clear projection of speaker’s voice during video conferencing in a noisy environment

Project By:

• Jason Chew Jia Wang

Supervisor

Mr Alfred Teng Yee Wee