Inspiration
A large portion of Road Accidents (78.4%) are caused due to the driver fatigue and inattention. This includes over speeding, driving under the influence of alcohol or drugs, and hit and run cases. Other causes of road accidents include such as the fault of cyclists, pedestrians, or drivers of other vehicles. Fewer accidents are caused due to neglect of civic bodies (2.8%), defects in motor vehicles (2.3%).
What it does
1.Facial Recognition 2.Emotion Recognition 3.Drowsiness Detection 4.Head Movement 5.Object Detection 6.Blind Spot Detection
How we built it
So, to avoid this we have come up with the idea of ADAS. Advanced Driver-Assistance Systems (ADAS) are electronic systems that assist drivers in driving and parking functions. Through a safe human-machine interface, ADAS increase car and road safety. Our Advanced Driver Assistance Systems usage is not limited to cars it can be used in any vehicle be it cars, self-driving cars, trucks, buses, etc. Our ADAS has facial recognition, emotion recognition, drowsiness detection, head position detection, object detection, blind spot detection and even driving duration as metrics to avoid accidents.
Challenges we ran into
GPU issues and sending data to the cloud
Accomplishments that we're proud of
The entire system is working fine and integrated perfectly
What we learned
Deep Learning And Computer Vision
What's next for ADAMS-Advanced Driving Assistance And Monitoring System
6.1.In many conditions the system might detect the driver as being inattentive while the vehicle is at halt, to avoid this the use of additional sensors like the Inertial Measurement Unit for detection of the state of the car, i.e. Driving or Stopped can be implemented.
6.2. Use a pre-built alcohol sensor to detect whether the driver is drunk or not and send the details to the authorities through the cloud (IoT) to avoid any kind of accidents.
6.3. To alert the driver about future events e.g. if a person is crossing or there is a speed breaker and the driver is not alert and is not taking the necessary action then he can be alerted to avoid any kind of accidents.
6.4. Use timers in trucks, in India many truck drivers drive for a longer time than the stipulated time which can lead to accidents and lose money as they carry a huge amount of load, this system will alert the authorities about the driver and for how much time he has been driving.
6.5. Object detection outside the car like important Sign Boards (physically challenged signboards) using OCR or Custom Machine
Learning Algorithms can be used to alert the driver to drive slowly.
The google drive link has the entire documentation of the project The video is for a longer duration due to the content sorry for the inconvenience.
Log in or sign up for Devpost to join the conversation.