Unconscious driving state is one of the core reason of road accidents. Several people loss their life due to Hypnic jerks during driving. Various models are designed to alert the driver during fatigue state, still there are multiple loop holes to efficiently develop smart driver drowsiness mechanism
Real Time Driver Drowsiness Alert Mechanism
Unconscious driving state is one of the core reason of road accidents. Several people loss their life due to Hypnic jerks during driving. Various models are designed to alert the driver during fatigue state, still there are multiple loop holes to efficiently develop smart driver drowsiness mechanism. However, in this project we are targeting multiple input parameters such as human pulse rate, facial landmark detector (eye aspect ratio) for effectively recognize drowsiness state. As, human pulse rate and eye opening closing ratio is tremendously affected in fatigue condition. Moreover, this module will computing results in real time environment and making effective decisions. Furthermore, with the help of machine learning, we can also maintain a brief history of current driver in order to check the time span of driver fatigue state. So, it’s the core need of today’s era to install such smart modules in vehicles to alert the user beforehand, thus preventing him to fall asleep behind the wheel and cause an accident.
The following objectives of this project are as follows:
The hardware gadget comprised of pulse sensor, high resolution camera, raspberry pie 3 B+ with accessories, GSM module and Bluetooth module. However, camera is placed on the dash board in front of driving seat which will the capture the facial landmarks and maintain brief driver fatigue state history. Moreover, pulse sensor is attach on the wrist along with Bluetooth module and controller. Furthermore with the assistance of Bluetooth module (master), the pulse sensor value receive by Bluetooth module (slave) which is attach to the facial landmarks detecting module. Both of these input parameters play an effective role to compute and provide fruitful results in real time.
The ultimate benefit of this project is to reduce human death ratio in road traffic accidents. As, human life is the most precious creature on the earth which is non-comprisable. Moreover, we will also maintain driver drowsiness time slots history in order to schedule driver’s time table if necessary. Furthermore, this human drowsiness detecting mechanism also implemented in our industries where abundant of machines operated by manpower. So, the scope of this project is not limited to driver drowsiness, but it is also extended to other real time scenarios such as Airport, railways etc.

Above figure depicts the overall project flow and the components used in each stage. To be more precise, with the help of pulse sensor and smart cam, we can get the real time human pulse rate and facial landmarks. Moreover, by enabling master-slave Bluetooth communication b/w two modules we can recognize current driver drowsiness state and generate alarm in case of emergency. Furthermore, facial landmarks (eye feature classification) and machine Learning algorithm will process in the controller (slave) and the pulse rate reading will get from master controller (rasp berry pie). Further, driver’s Hypnic jerks time span history will maintain in online server for scheduling driver’s time table if necessary. Finally, all of the above project modules will code in Python due to its diversity and real time effectiveness as compare to other available programing languages.
| Item Name | Type | No. of Units | Per Unit Cost (in Rs) | Total (in Rs) |
|---|---|---|---|---|
| Raspberry Pie 3 B+ With accessories | Equipment | 2 | 12000 | 24000 |
| Raspberry Pie Zero | Equipment | 2 | 10000 | 20000 |
| 8Mp Raspberry Pie V2 Camera | Equipment | 2 | 5000 | 10000 |
| Cables/ adapter | Miscellaneous | 1 | 2000 | 2000 |
| Pulse Sensors | Equipment | 2 | 500 | 1000 |
| Bluetooth Module HC 05 | Equipment | 2 | 750 | 1500 |
| GPIO 40 pin | Miscellaneous | 2 | 500 | 1000 |
| Total in (Rs) | 59500 |
The analog transmission using AM (Amplitude Modulation) and FM (Frequency Modulation) tech...
AI BASED SEARCHING OF BEST CANIDATE: Our project will automatically choose the best candid...
Electroencephalography (EEG) is an electrode which can be placed over the skull to measure...
We proposed a face recognition system for KKKUK, based on computer vision and proposed for...
E-Commerce Websites are a great way of shopping online and having products delivered to ou...