A study on the context aware system for intelligent automobile black box

The automobile black boxes may help identify the cause of an accident once it occurs. Recently, automobile black boxes develop into intelligent ones with more features than simple recording. The intelligent black boxes can make decisions and take responsive actions based on video analysis. This paper proposes the intelligent black box system that can prevent theft and damage to a car based on situation awareness as well as protect privacy from unauthorized personnel with long distance face recognition. The proposed algorithm detects and recognizes face as well as protects privacy depending on the distance between the object recorded by the black box and the car. It also generates warning automatically against dangerous situations through situation awareness. Tests on the algorithm prove that it can recognize an approaching object, identify it, and set the alarm if needed, depending on each risk level. Also, it does not provide face information for unauthorized personnel with long distance face recognition to protect privacy.