arevish / Brain-wave-detection-system

Project is focused on the detection and extraction of a brain wave signal with the help of analog as well as digital circuitry. Using active electrodes on human scalp, the brain signals were fed into a series of hardware and software stages. Simple conscious movements such as blinking caused a change in the detected waveform. Although the project was not successful in discriminating between different motions or utilizes the signal to control an electrical device, the team was able to successfully separate and display the alpha waves after filtering off all associated unwanted signals.

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Brain Wave Detection System

Project is focused on the detection and extraction of a brain wave signal with the help of analog as well as digital circuitry. Using active electrodes on human scalp, the brain signals were fed into a series of hardware and software stages. Simple conscious movements such as blinking caused a change in the detected waveform. Although the project was not successful in discriminating between different motions or utilizes the signal to control an electrical device, the team was able to successfully separate and display the alpha waves after filtering off all associated unwanted signals.

PROJECT IDEA: -

In this research paper we have designed a new instrument which reads the subject brain waves continuously and gives feedback in the form of led signal. In previous protocols, researchers concentrate on only drowsiness but they are not gives important while driving. Some of them have done this research using real car. But these protocols feedbacks only after the subject slept. So in order to rectify these problems we designed new protocol which identifies subject drowsiness before he is going to sleep. With the help of this, we are analyzing the brain wave signals. Human brain has millions of interconnected neurons. The interaction pattern between these neurons is represented as emotional states and thoughts. This pattern will be changing as the human thoughts change, which in turn produces different electrical waves. In the proposed system, an alarm with low power consumption is placed near the driver which will wake up the driver while he falls asleep during driving. The EEGsensor senses the brain signals of the driver and if he falls asleep it will send the signal to the embedded system for further processing. The signal is sent to the amplifier and the amplified signal is given to the comparator that compares the input signal (amplified brain wave signal) with the threshold voltage that is set according to different sensation states of the driver.

WORKING OF PROJECT: -

The proposed works are briefly described in this section. The system hardware consists of the signal acquisition module. The EEG signal is captured through the EEG electrode; the inbuilt amplifier will amplify the captured EEG signals. The EEG signals is then passed to the High Pass Filter which is capable of filtering 60Hz. Then it is given to the Low Pass Filter which is capable of filtering up to 5Hz. At the same time the amplified EEG signals is given to the LM339 comparator, the acquired voltage is compared with the reference voltage. From the obtained voltage ,the various states of the driver is indicated using LED whereas the red LED represents the active state of the driver, the yellow indicates the normal state of the driver and the green represents the drowsy state of the driver. The program for the whole module is dumped into the microcontroller. Whenever the drowsy state is detected, the microcontroller enables the motor to slow down and the drowsy state of the driver is indicated using LCD display and the alarm starts to ring.Fig.2.1.1, Shows the overall block diagram of the proposed system

project architecture

EXTRACTION OF BRAIN SIGNAL: -

The electroencephalogram (EEG) is a record of the electric signals generated as the result of brain activity. EEG provides important and unique information about the sleeping brain. While EEG signals have advantage in making accurate and quantitative assessment of alertness levels.EEG signal will be obtained from the Brain wave Sensor. EEG signal is quite small, ranges from 1Hz to 100Hz and amplitudes vary from 1µV to 100µV.

extraction of brain signal



eeg signal



result simulation


project.working.mp4

checkout the synopses , report and ppts for more details.

About

Project is focused on the detection and extraction of a brain wave signal with the help of analog as well as digital circuitry. Using active electrodes on human scalp, the brain signals were fed into a series of hardware and software stages. Simple conscious movements such as blinking caused a change in the detected waveform. Although the project was not successful in discriminating between different motions or utilizes the signal to control an electrical device, the team was able to successfully separate and display the alpha waves after filtering off all associated unwanted signals.