Driver Monitoring System
To encourage efficient and safe driving, drivers are screened and evaluated on driving habits. Advanced Driver Monitoring System or Driver Fatigue Monitoring System to detect and monitor behavior and fatigue levels of drivers are emerging which makes the cars more intelligent for avoiding accidents on roads. Systems are being developed for real-time monitoring of vehicles which controls the speed of the vehicle and fatigue level of the driver to prevent accidents. The primary components of such a system will be microcontrollers along with some sensors like an eye blink, gas, impact sensors, alcohol detecting sensor, and fuel sensors. GPS and Google Maps APIs are used to track the location of the vehicle which can be sent to a predefined number in the system.
Faststream Technologies’s Driver Monitoring System, using a single low-power in-vehicle camera and advanced vision technologies, provides reliable detection of driver drowsiness and distraction, alerting the driver to reduce the chances of serious accidents and thus providing a safer and comfortable drive. Our solution is offered to OEMs and Tier1s for pre-installment into their cars and trucks.
Faststream’s Driver Sense monitors in real-time the driver’s state, detecting signs of drowsiness as well as distracted driving. Recognition of enrolled drivers and detection of actions (such as wearing a seatbelt and holding a cellphone) is also available. We use a single low-power camera inside the vehicle and complex vision and AI algorithms that enable reliable driver monitoring on the edge.
Real-time Driver’s Drowsiness Detection with Neural Network
Faststream’s driver fatigue detection system used Convolution Neural Networks (CNN) for detection of early drowsiness and fatigue of the driver which helps to prevents accidents.
Working Principle
- The optical sensors monitor the driver’s eye and pupil movement and head position.
- The machine vision algorithm analyzes the video feed in real-time.
- We can anticipate driver drowsiness by applying a Recurrent Neural Network over a sequence frame driver’s face. We used a dataset to shape and approve our model and implemented repetitive neural network architecture multilayer model-based 3D Convolutional Networks to detect driver drowsiness.
- When an anomaly is registered, an alarm sound is played. If the driver does not restore his attention on the road, a second louder alert is activated. _ The system sends the info about the anomaly to the dispatching center (if deployed)
Fatigue Detection System Hardware Used
- Optical Sensors
- Central Unit
- Loudspeaker
- Fasteners
- Calibration Button
- Antenna
- Dispatching Platform
Faststream helped one of the clients to ADAS — Advanced Driver Assistance Systems next age of vehicle technology by implementing the Internet of Things (IoT) with sensors, feel free to contact us for more detail info@faststreamtech.com .