Abstract:

Abstract Traffic accidents pose an increasing threat to society, and researchers are dedicated to preventing accidents and reducing fatalities, as highlighted by the World Health Organ-ization. One significant cause of accidents is drowsy driving, which often leads to severe injuries and loss of life. The objective of this research is to create a fatigue detection sys-tem that can effectively minimize accidents associated with exhaustion. The system uti-lizes facial recognition technology to identify drowsy drivers by analyzing eye patterns through video processing. When the level of fatigue surpasses a predetermined thresh-old, the system alerts the driver and adjusts the vehicle's acceleration accordingly. The implementation of OpenCv libraries, such as Haar-cascade, along with Raspberry Pi fa-cilitates seamless integration of the system. This dissertation evaluates advancements in computational engineering for the development of a fatigue detection system to miti-gate accidents caused by drowsiness. It offers valuable insights and recommendations to enhance comprehension and optimize the system's effectiveness, ultimately leading to safer road travel.

Keywords: