P. 1


|Views: 2,169|Likes:
Published by Kiran Kumar

More info:

Published by: Kiran Kumar on Feb 09, 2011
Copyright:Attribution Non-commercial


Read on Scribd mobile: iPhone, iPad and Android.
download as PPT, PDF, TXT or read online from Scribd
See more
See less






What is an artificial passenger? 

Natural language e-companion. eSleep preventive device in cars to overcome drowsiness. Life safety system.  

What does it do? 

Detects alarm conditions through sensors. Broadcasts pre-stored voice messages preover the speakers. Captures images of the driver.  

Voice recognizer or speech recognizer.  .Devices that are used in AP The main devices that are used in this artificial passenger are:are: Eye tracker.

About AP  The AP is an artificial intelligence² intelligence² based companion that will be resident in software and chips embedded in the automobile dashboard. The system has a conversation planner that holds a profile of you.  . including details of your interests and profession.

 . A microphone picks up your answer and breaks it down into separate words with speechspeech-recognition software. A camera built into the dashboard also tracks your lip movements to improve the accuracy of the speech recognition.

the system judges you to be alert and tells the conversation planner to continue the line of questioning.  . If you reply quickly and clearly. A voice analyzer then looks for signs of tiredness by checking to see if the answer matches your profile. Slow responses and a lack of attention are signs of fatigue.

the voice analyzer assumes you are dropping off and acts to get your attention. If your response is slow or doesn·t make sense. .

  . Detecting frequency of up and down nodding and left to right rotations of the head within a selected time period.Detecting driver vigilance  Aiming a single camera at a head of a driver. Determining frequency of eye blinkings and eye closings.

eye-  Data are stored as a series of x/y coordinates related to specific grid points on the computer screen. .HOW DOES TRACKING DEVICE WORK?  Data collection and analysis is handled by eye-tracking software.

 Our head tracker consists of tracking the lip corners. eye centers. ´Occlusionµ of the eyes and mouth often occurs when the head rotates or the eyes close. mis-  . so our system tracks through such occlusion and can automatically reinitialize when it mis-tracks. and side of the face.

Representative Image .

Eye Tracker .

Monitoring System .

Track lip corners using dark line between lips and color predicate even through large mouth movement like yawning.Tracking includes the following steps  Automatically initialize lips and eyes using color predicates and connected components.  .

  . Determine eye blinking and eye closing using the number and intensity of pixels in the eye region. Determine rotation using distances between eye and lip feature points and sides of the face. Construct a bounding box of the head.

Mark the lip pixels in the image is important. BLUE. GREEN)RGB) are marked in the image offline. The lip and eye colors ((RED.  .

 Each pixel has an Red(R). G. B components.  . go to this location in the RGB array indexing on the R.0*( j*j+k*k+i*i )/(2*sigma*sigma)). Green)G). This array location can be incremented by equation (1): exp(í1. For a pixel that is marked as important. (1) where: sigma is approximately 2. and Blue(B) component.

. is marked as important multiple times. If a color. Pixel values that are marked as unimportant can decrease the value of the RGB indexed location via equation (2) as follows: exp(í1. its new value can be added to the current value. (2) where: sigma is approximately 2.0*( j*j+k*k+i*i )/(2*(sigmaí1)*(sigmaí1))). or pixel value.

and the largest non-skin components above the nonlips are marked as the eyes First a dark pixel is located  . Another RGB array is generated of the skin colors. The values in the array which are above a threshold are marked as being one of the specified colors.

 The system goes to the eye center of the previous frame and finds the center of mass of the eye region pixels Look for the darkest pixel. which corresponds to the pupil This estimate is tested to see if it is close enough to the previous eye location   .

so the eye motion between individual frames should be relatively small. Feasibility occurs when the newly computed eye centers are close in pixel distance units to the previous frame's computed eye centers. This kind of idea makes sense because the video data is 30 frames per second. .

the system searches a window around the eyes and finds all non-skin connected components in nonapproximately a 7×20 pixel window. and finds 7× the slant of the line between the lip corners using equation (5).y 1 is the coordinate of a feature. . If new points are too far away. (( y 2 íy 1 )/( x 2 íx 1 )) (5) where: x 1 . This equation finds the slope between two points in general. and x 2 .y 2 is the coordinate of the other corresponding feature.

. These two stages are called the eye black hole tracker. The system selects the eye centroids that have the closest slant to that of the slant between the lip corners using equation (5).

 The detection of eye occlusion is done by analysing the bright regions. in the first frame of each sequence we find the brightest pixel in the eye region and use this as the eye white color. To determine what eye-white eyecolor is.  . If not blinking is happening. As long as there are eye-white pixels in the eyeeye region the eyes are open.

the system declares that driver has his eyes closed for too long.If the eyes have been closed for more than approximately 40 of the last approximately 60 frames. .

Output  Onsite alarms within the vehicle Remote alarms  .

2) Water craft such as boats. 3) Trains and subways. .Other applications 1) Cabins in airplanes.

SUMMARY  Method for monitoring driver alertness Sufficient time to avert an accident.   . Monitoring of full facial occlusion of the driver.


You're Reading a Free Preview

/*********** DO NOT ALTER ANYTHING BELOW THIS LINE ! ************/ var s_code=s.t();if(s_code)document.write(s_code)//-->