This video consists of 3 segments: eyelid and eye gaze ...



This video consists of 3 segments: eyelid and eye gaze tracking, head pose and facial expression tracking, and fatigue detection by combined feature tracking. Only a single camera is utilized to record the facial data of the human being.

• Eyelid and eye gaze tracking

• Head pose and facial expression tracking

•

• Fatigue detection based on combined feature tracking

-----------------------

[pic]

Three parameters, which measure the motion of a human head: PAN, TILT and SWING are computed. The size of the opening mouth is measured as a representation of the facial expression. The colored graphics marked on the left image indicate the feature positions of the facial features being tracked and the normal direction of the human face.

[pic]

The system computes 3 parameters over time:

ACES: measures average eye closure time speed.

PERCLOS: measures the percentage of eye closure.

PERSAC: measures percentage of saccade eye movement.

The bouncing red dot on the image at the upper right corner of the screen indicates the driver’s gaze direction. The image at the upper left corner shows the human eye being tracked. The found position of the eyeball is outlined in white.

The upper left image monitors the eyeball movement and the right upper image represents the head pose and mouth tracking result. The lower image outputs the composite fatigue index, which is computed based on combining individual fatigue parameters. The red circles drawn on the images indicate issuing the “drowsy driver” alert generated by the system.

[pic]

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download