Deep convolutional network based real time fatigue detection and drowsiness alertness system

<span>Fatigue and drowsiness detection techniques based on the external features are under progress, methods of facial feature extraction require further development. This paper discusses innovative processes, efficient methods, recent advancements in field fatigue detection. In this proposed model, a wide application is planned artificial intelligence by defining fundamentals human-computer interaction, expression recognition driver fatigue-sleepiness determination. research outlines an effective three-phase strategy for detecting drowsiness. Viola Jones used to detect traits these three phases. Detection yawning tracking once face has been identified, segmenting skin, system becomes lighting invariant portion itself, focusing chromatic components reject most non-face image backdrops. The color eye carried out template matching with correlation coefficient. vectors each above phases concatenated, binary result obtained. analysis sound successive frames into non-fatigue states classified. If time state exceeds threshold, will alarm. </span>