Современные информационные технологии/1. Компьютерная инженерия

 

Sarana E.V., Zaitseva T.A., Zolot’ko K.E.

 

Dnipropetrovsk National University named after O. Honchar

 

 

Person identification in distance education

 

 

Dynamic adoption of distance learning technologies in education is taking place now. At the same time the issue about quality control of knowledge received in educational process is appearing. In the majority of cases using distance technologies happens when there is no visual contact of examiner and student. In such conditions student can use background and teaching materials, promptings etc. during testing, what considerably reduces objectivity estimation.

Using intellectual systems in distance learning process and knowledge control can be appreciably amend both the quality of learning process and verify obtaining knowledge. In practice we can achieve these objects using videoconference, i.e. visual contact between teacher and student. In this case it is easy to solve the problem of student identification. Direct audio-visual contact between teacher and student in distance learning can be realized using available current telecommunicational technologies, however such approach can’t solve person-identification problem entirely controlling his knowledge.

It is known that biometrics, which is used for person identification, can be classified according to recognition rate, stability to falsification and environment, and also for measurability. Concerning to this task we can mark out such characteristics as face, fingerprint, hand-grip, voice, iris, ear shape.

Due to development of sensor technologies, new dynamic methods of biometric person identification – analysis of brain electrical activity, cardiovascular system - appears recently. They can estimate current functional status of student simultaneously with identification, for example when he gets to the stress situation.

In the past ten years, it has been shown that combining biometric systems achieves better performance than techniques that use only one biometric modality. Taking into consideration specificity of application domain, we can ground choice of distance face recognition, fingerprint and EEG analysis.

Before studying starts we create a base of students’ biometric data. It is supposed to use 3D face photo, images received by scanning fingerprints, and EEG. These data can be used for student identification and control of self-sufficiency training

Studying EEG, capacity of definite frequency ranges when real movements or imaginary movements decrease on amplitude. Therefore during initial EEG registration, student makes several various imaginary hand and leg movements to ascertain unique EEG features, which has changed while this movements. The result of such calibration will show optimal channel for using. Under subsequent identification, registered EEG compares with previous record and computer makes conclusion about personality and his brain activity. Thus we can define the state of student’s brain and how objective his answers are.

Combining cordless sensors with contactless makes possible gathering data at distance. The program monitors electrical activity of the brain and sends information to a computer using wireless communications. Then it forms digital user’s portrait. All the information, received from the student (fingerprint, face image, EEG) is sent in a batch to examiner’s computer.

Combination of classical biometric methods and EEG analysis ensure new ways of integrating multifunctional systems of authentication (also called verification) and identification, which are rather reliable and quite acceptable because of its imbeddedness and optionality of oral communications. EEG-based biometry is an emerging research topic and may open new research directions and applications in future.

Thus a range of tasks demanded complex solving appears. It is detection of fingerprints, faces, EEG analysis and forming databases, biometric identification, transferring of required video information by lower speed communication channel, high security clearance, access control and automation, flexible control implementation, simultaneous process of working with data.

 

Bibliography:

 

1. S. Marcel, J.d.R. Millan Person authentication using brainwaves (EEG) and maximum a posteriori model adaptation //IDIAP Research report, 2007.

2. Ландэ Д.В., Фурашев В.Н. О цифровой идентификация личности //Открытые информационные и компьютерные интегрированные технологии. – Харьков: НАКУ, вып. 34, 2007, с. 127 – 135.