| Idea Name: |
DriveCare |
| Slogan: |
Take care through DriveCare |
| Supervisor Name: |
Dr. Ahmad Rauf Subhani |
| Supervisor Designation: |
Assistant Professor |
| Supervisor School: |
CEME |
| Supervisor Department: |
Electrical Engineering |
| Contact number: |
+923008538215 |
| Email ID: |
[email protected] |
| Abstract: |
| Fatigued or emotional drivers can suffer from impairments that drastically reduce their safety on the road. Hence, we devise a safety system based on driver monitoring. It will utilize facial feature extraction and PPG signals to determine the driver’s state. ML models will be used to accurately determine the driver’s mood. If the driver is distracted, necessary actions will be taken. |
| What is the unmet need in society that your idea will fulfill ? |
| Studies show that the risk of a car accident can be up to 10 times greater when the driver is under the influence of strong emotions. Hence, there is a need to develop a system that not only detects erratic driving behavior but also takes preventive measures to avoid vehicular collision. |
| Who needs it ? How many would benefit ? |
| Anyone who owns a manual or semi- automated vehicle needs to adopt this system. system will continuously monitor driver's emotions that will help predict their behavior as well as take appropriate measures to avoid accidents. This will keep them safe until they reach their destination. |
| How will the solution works |
| Camera captures the facial images of a driver and sends it to a Raspberry Pi module that will extract useful features from them. By tracking driver’s face, gaze and emotions, the system can detect drowsiness, road rage and other potential safety issues using ML model and computer vision. If the driver is not alert, an alarm will be played to warn the driver to remain focused. Also, a wristband containing a PPG sensor will also determine driver’s state through pulses to maximize accuracy. |
| Who are your competitors ? How is your solution different |
| Majority of the work being done on this issue is either purely based on identifying facial features or determining physiological signals. However, our model uses a hybrid approach using PPG signals in addition to images. So, we have improved the accuracy of the model and minimized error in results. |
| Status: |
new |
| Entry Date & Time: |
2022-12-15 (1415) |
