A MACHINE LEARNING APPROACH TO EYE BLINK DETECTION IN LOW-LIGHT VIDEOS

Inadequate lighting conditions can harm the accuracy of blink detection systems, which play a crucial role in fatigue technology, transportation and security applications. While some video capture devices are now equipped with flashlight technology to enhance lighting, users occasionally need remember activate this feature, resulting slightly darker videos. Consequently, there is pressing improve performance systems detect eye accurately blinks low light This research proposes developing machine learning-based system see flashes low-light The Confusion matrix was conducted evaluate effectiveness proposed system. These tests involved 31 videos ranging from 5 10 seconds duration. Involving male female test subjects aged between 20 22. measured using confusion method. results indicate that by leveraging learning approach, achieved remarkable 100% detecting within However, necessitates further development account for more complex diverse real-life situations. Future studies could focus on sophisticated algorithms expanding conditions. Such advancements would contribute practical application broader range scenarios, ultimately enhancing its technology.