MEMS sensörlerine dayalı hareket analizi ve tanıma

This study covers the use of Micro-Electro-mechanical system (MEMS) sensors and communication with Bluetooth to transmit sensor data in order to understand the shape drawn into the air. It is important for people, especially children with speech disabilities, to be understood by the characters such as commands, shapes, letters or numbers that are given by their hand or arm movements. For this purpose, the MPU6050 which is type of inertial measurement unit, was used as a MEMS sensor to capture dynamic hand movement. This device is preferred because it is a small integrated structure that transfers data from the gyroscope and accelerometer sensors inside to the computer via Bluetooth. They can be placed easily on the hand and arm. In this study, some numbers, letters and geometrical shapes, which were generated by hand movements, have been recognized using hidden Markov model algorithm in MATLAB. The obtained data is transferred to a computer by means of wireless communication. The empirical studies have shown that the system can successfully recognize the generated gestures.