Tuesday, August 2, 2011

CAIRO Overview

Established in 1997, CAIRO is one of the Centers of Excellence in UTM. It has established itself as the leading R& D center in AI, Robotics and Automation in the country. Its main role is to acquire advanced technologies, to be involved in research, and to work with local industries to improve their product competitiveness. CAIRO’s mission is to be an internationally acknowledged research center. CAIRO actively collaborates with various universities in Japan, France, United Kingdom and others around the world in various research areas. The ability to turn ideas into commercialized products is also of primary importance to CAIRO. Correspondingly, a spin-off company, Augmented Innovations Sdn. Bhd (AISB), has been set up. The research outputs from CAIRO are improved further and commercialized through this company. In line with the re-structuring policy of Universiti Teknologi Malaysia, CAIRO has since expanded to include more research groups such as control, instrumentation and process tomography. CAIRO is now under a research alliance called ACE (Automation, Cybernetics, and Engineering Systems), a move by UTM in the restructuring process in order to encourage a more multidisciplinary research rather than a faculty-based.

This blog is to advertise our product and researches.

Wood Recognition System


Product Features
  • A PC-based wood recognition system
  • Use VSDP Library developed by CAIRO
  • Accuracy rate: 90-96%
  • Can recognize wood species in real time
  • Portable: use offline / online
  • Can be deployed on PDA
  • Internet based recognition using web service
  • First commercialize wood recognition software

Product Background
  • Currently, wood identification based on the anatomy of the wood is done by only a few certified personnel using their naked eyes
  • In order to be able to perform the job, the personnel need to undergone a very complicated training that require them to study up to 3000 species of wood in Malaysia jungle
  • Furthermore, the probability of human error or biasness in manual identification are high. Thus, there is an urgent need of a dynamic, automatic and systematic system to perform the wood recognition in high accuracy rate and in a short time

Market Size

Wood Industry: Safety issues, ensure the quality of wood-based product, timber grading (6th largest external trade of Malaysia)

Public agencies: To control illegal logging and related trade based

Research: Determining type of wood, understanding ecology and geological information of an area, study relationship between species and identifying a wood from existing object


How This Product Can Help Your Company
  • This system can be used anytime and anywhere as it is developed in web service
  • Save time, cost and human resources as it can recognize wood species in just 1 second
  • The identification of wood will be more organized and systematic
  • By having the system, authorities are able to prevent fraud in logging

Relevant Industries
Wood industry, research and authorities

E-HOUSE : Solution for Smart Home System



Product Features
  • The system gives us the complete control over the access, security, automation and energy savings
  • It consists of several input and output control system which are centralized to embedded system
  • User can access the system through monitor and control some application in the premise remotely such as lighting and air-conditioning
  • This system is also equipped with surveillance cameras to detect any suspicious movement when the security system is activated. The system will sound an alarm and send a short message to the user's mobile phone using GSM modem


E-House System Architecture


Relevant Industries
  • Medium size industries
  • Educational industries

Online Signature validation and Verification Using Neuro- Template Method

Our signature verification system
  • Features a fast, easy to use and good accuracy system
  • Using an artificial neural network with a new training technique for verification
  • Using a new threshold method where both the neural network output and selected global features are used to obtain an improved false acceptance and false rejection rate
  • Can be used for PC log-on and internet log-on

How the system works

During Enrollment
  • Signature data such as pen position and pen pressure with respect to time are captured using a digitizing tablet and act as the input to the system
  • Important features that can characterize the signature are extracted after data has been pre-processed to ensure uniformity
  • Signature data are validates using a correlation technique to decrease the variations among genuine signatures
  • Signature data are trained using a novel neuro-template method
  • Neuro-template of each person will be stored in database
During verification
  • User will sign in
  • Signature data of user is captured and pre-processed
  • Important features of signature data, and the neural network output are input to the threshold module
  • If these values are within the threshold levels set by the system, then the person will be authenticated and can gain access otherwise access will be denied
Advantages
  • Network environment needs accurate, fast and automatic authentication of users
  • PINs, password, smart card are methods of authentications but can be forgotten, misplaced, or stolen
  • Signature, a behavioral biometrics identification is the most accepted method of asserting a person's identity

Vision Plate II : Malaysian License Plate Recognition System

Overview

There has been a number of commercialized software that can be used for identification of license plates, which are mainly from the USA and Europe. However, they cannot be utilize in Malaysia since Malaysian license plates are differ in styles and formats. The license plate of vehicle s in Malaysia are divided into several different colours and formats. For example, the license plate for normal private vehicle has black background and white characters as shown in Figure1.Whereas he license plate for taxis has white background and black characters. Another type of license plates, which are occasionally introduced, are special plates such as Proton Satria, Sukom, Tiara and several others. These plates produced usually during the introduction of such vehicles in the market.

To overcome the problem of this complexity, CAIRO has developed a software known as VISIONPLATE which can recognize various license plates design especially in Malaysia. The second version of this system is currently being designed to overcome other problems as well such as various levels of illumination and blurriness, which are not incorporated in the first version.

Product Features

VisionPlate is a state-of-the-art, image recognition software package for the automatic visual detection, identification and recognition of vehicle license plates.

It is capable of detecting, locating and identifying the plate number of any vehicle.
The CCD camera takes images of incoming car when the loop sensors activated.

The images are transmitted towards the computer, typically via standard video signal(PAL/NTSC).

The video signal of the camera is received and digitized by a video capture device (a frame grabber card) which has been installed in the computer.

In the computer, the image recognition module of VisionPlate will process the digitized images. The steps of the recognition process are:
  1. The number plate is located on the images
  2. The located number plate is extracted from the image, and transformed to a uniform size position.
  3. Using segmentation, the characters of the plate are identified.
  4. The identified characters are recognized using Neural Network(NN) technology

The result of the recognition is the number plate in text(ASCII) format. It then can be processed further according to the specific application such as stored or looked-up in the database.

This output can also be linked to ticket printing machine, payroll system and etc. via serial port(RS232).

Application
  • Vehicle Access Control
  • Vehicle Entrance Admission
  • Road Traffic Surveillance/ Control
  • Vehicle Theft Prevention / Detection
  • Crime Prevention/ Detection
  • Vehicle Counting Automation
  • Highway Toll Collection
  • Public/ Private Car Park Management/ Automation /Control

Benefits
  • Reduced costs by automation
  • Efficiency of work-processes
  • Spare of time and resources
  • Improved service
  • Reduced human mistakes
  • Efficient protection of assets

Intelligent Diagnostics of Power Transformer (ADAPT V 3.0)

Overview
  • Fault in power transformers can cause extensive damage which does not only interrupt electricity supply but also result in large revenue losses
  • Transformers must be regularly monitored and maintained. Transformers of different made and capacities exhibit different characteristics and problems. Which make routine to maintain and diagnosis of such transformers rather difficult
  • The Malaysian power utility company Tenaga Nasional Berhad(TNB) needs an efficient transformer fault system
  • An out put of collaborative research between Tenaga National Berhad Research(TNBR) , a subsidiary of TNB and Centre for Artificial Intelligence and Robotics(CAIRO) of Universiti Teknologi Malaysia

Product Features

  • The transformer health is being monitored by entering the result of dissolved gas analysis in a database. Fuzzy Logic algorithm is used to detect if there is any irregularities in the result of dissolved gas analysis and interpret the fault. It is used as an aid for the engineers to monitor the transformer's health
  • This system is developed in web based environment

Highlight
  • Advanced database management information system for fast and easy retrieval and manipulation
  • Utilizing Total Intelligence Diagnostic Solution (TIDS) architecture as the core fault diagnostic engine
  • Incorporate Fuzzy Logic into the well known transformer diagnosis method, Key Gas, roger Ratio, TDCG and Logarithmic Nomograph
  • Web Accessible and Reporting for fast information delivery with no boundary
  • Sophisticated early auto email or SMS precaution alert system to avoid the risk of transformer explode
  • Empowered by Microsoft© SQL Server 2000 DBMS and Microsoft©Internet Information Services(IIS 4.0)

IC Inspection System


Product Features
  • Designed for quantity control purpose
  • Designed to inspect small scale size for medium size industries
  • Developed to overcome human error during inspection process
  • Manage to inspect 10,000 chip per hour which is 10 times compared to human capabilities
  • Developed 100% by CAIRO including libraries, software and hardware
  • Implementation can be done using embedded PC or commercial PC
  • Implementing image processing method in the whole process
  • System is compact, cost effective and quite portable compared to other kind of some solution
  • Using PLC and PIC to handle input and output
  • Applicable to be implemented for lab purpose in programming PLC with industrial specification
  • Sorting process will be done automatically according to chip character(crack,no crack, lead and labeling)
Relevant Industries
  • Medium size Industries
  • Educational Industries