Drowsiness-Detection-and-Alert-System

A countless number of people drive on the highway day and night. Taxi drivers, bus drivers, truck drivers, and people traveling long-distance suffer from lack of sleep.

Due to this, it becomes very dangerous to drive when feeling sleepy. To counter the issue, we can develop a machine learning model, to alert the driver in case of extreme drowsiness by real time detection of specific characteristics that are showcased by the driver such as Eyes Closed, Yawning, Head bend, Head tilt, etc.

The first part of the work focuses on monitoring a driver vigilance by real time tracking of the driver’s facial features i.e, eye status, yawning. The second part of the research mainly contributes to the methods of classification of these facial features in drowsiness class. In the third part of research, we will analyse the facial expressions of the driver using a well-developed machine learning model. In the last part of the research, we will alert the driver in the case of exceeding a fixed drowsiness score given by the system based on the facial expressions of the driver. For each stage of the research we present and discuss our experimental results.

We can create this Machine learning model by using Python libraries such as TensorFlow, Keras, OpenCV, Numpy, PyGame, and Pandas and implementing Convoluted Neural networks on the Previous Set Database.

The major outcomes of the research will benefit the upcoming generation of intelligent vehicles and increasing road safety for both driving and pedestrian public. This will also contribute in increasing alertness of the drivers and it can be developed into even more sophisticated system which can directly contact the operators in case of extreme drowsiness of the driver reducing the chances of accidents and improving revenue of the company as well as saving many precious lives.

Keywords:

Convoluted Neural Networks, Driver Alertness, TensorFlow, Image Dataset, Drowsiness