Image analysis – Applications – Personnel identification
Reexamination Certificate
2000-06-29
2004-10-26
Ahmed, Samir (Department: 2623)
Image analysis
Applications
Personnel identification
C280S735000
Reexamination Certificate
active
06810135
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
This invention relates generally to a frame differencing technique for eliminating background interference in a human presence detection system and, more particularly, to a frame differencing technique for eliminating background interference in a human presence detection system employed in a vehicle for proper airbag deployment.
2. Discussion of the Related Art
The National Highway Traffic Safety Administration requires passenger side airbags on all passenger cars manufactured after 1998. Airbags have saved thousands of lives, but safer airbag deployment systems can save even more lives. Proposals have been submitted by the National Highway Traffic Safety Administration to require advanced airbags that are safer and more efficient. Thus, there is a need in the art for an advanced airbag deployment system that determines airbag deployment depending on the detection, identify and tracking of a person in the passenger seat of the vehicle.
Future airbag deployment systems will be able to identify and track a person in the passenger seat of the vehicle. Based on this information, the system will provide signals whether to fire the airbag, fire the airbag at low velocity or fire the airbag at normal high velocity during a crash event, depending on whether a person is seated in the passenger seat, the size of the person and the position of the person relative to the airbag deployment door. In one example, if a person is detected in the passenger seat, and is some small distance (for example, within three inches) from the airbag door, then the airbag does not fire during a crash event. If a person is detected in the passenger seat, and is close (for example, between three and eight inches) to the airbag door, the airbag is fired at a lower velocity during a crash even. If the person is detected in the passenger seat, and far enough away (for example, more than eight inches) from the airbag door, then the airbag is fired at normal high velocity during a crash event.
Present passenger sensing systems typically include ultrasonic, weight, infrared and/or electromagnetic sensors to detect a passenger for airbag deployment. Ultrasonic sensing systems sense the motion of objects within the passenger seat area and determine whether the object is moving closer or farther away from the sensor location. However, ultrasonic sensors don't identify the nature of the object, and thus can't tell if it is a person or some other object, such as a bag of groceries. Similar to ultrasonic sensors, microwave sensors employed in active doppler radar systems can track objects, but cannot identify human presence. Weight sensors identify objects in the passenger seat based on applied pressure, but don't consider the passenger's location relative to the airbag. Passive IR sensors acquire thermal images of the passenger seat, but these systems are very costly. Present active IR sensors- sense the relative location of the passenger relative to the airbag, but cannot identify human presence. Electromagnetic systems include LC resonant circuits where body capacitance is used to detect presence and identify objects, but these systems can not track the objects.
U.S. Pat. No. 5,835,613 issued to Breed et al., Nov. 10, 1998, discloses a vehicle interior monitoring system that claims to identify, locate and monitor persons in the passenger compartment of the vehicle. The monitoring system employs infrared emitters that illuminate the interior of the vehicle, and charge couple device (CCD) arrays that detect the radiation. Outputs from the CCD arrays are analyzed by computational devices that employ pattern recognition algorithms to classify, identify or locate the content or objects in the passenger seat. The pattern recognition system for determining vehicle occupants disclosed in the '613 patent employs complicated software that must learn the shape of an individual in all kinds of lighting situations under various conditions. Additionally, employing pattern recognition in this manner is limited in its ability to track the individual as he or she moves around in the passenger seat. Further, the ability to identify and track humans by general pattern recognition is questionably unreliable. Pattern recognition cannot identify who the person is, only detect an object's shape.
What is needed is an improved airbag deployment sensor and system that is able to effectively identify and track a person in the passenger seat of a vehicle. It is therefore an object of the present invention to provide such a sensing system for vehicle passenger airbag deployment.
SUMMARY OF THE INVENTION
In accordance with the teachings of the present invention, a human presence detection system is disclosed that employs a frame differencing technique for subtracting out background interference from images generated by the system. The system includes an infrared source that generates a beam of infrared radiation, and an infrared detector that receives infrared radiation reflected from objects in the path of the beam. Face recognition software is employed to determine the presence of a person from the reflected radiation. The infrared source is pulsed and the detector is synchronously shuttered to the pulses so that image frames are generated at different times, where one frame includes reflected radiation and background radiation and another frame includes only background radiation. The frames are subtracted to separate out the background radiation.
In one embodiment, the detector includes a pixel array of photodiodes and first and second capacitor storage sites for storing the image frames on a single CMOS chip. The storage sites are subtracted in a summation device that is also on the CMOS chip. In an alternate embodiment, the frames are stored and subtracted at an off-chip site.
REFERENCES:
patent: 4625329 (1986-11-01), Ishikawa et al.
patent: 4648052 (1987-03-01), Friedman et al.
patent: 4708882 (1987-11-01), Asano et al.
patent: 4720189 (1988-01-01), Heynen et al.
patent: 5001557 (1991-03-01), Begle
patent: 5008946 (1991-04-01), Ando
patent: 5829782 (1998-11-01), Breed et al.
patent: 5835613 (1998-11-01), Breed et al.
patent: 5845000 (1998-12-01), Breed et al.
patent: RE36041 (1999-01-01), Turk et al.
patent: 6060989 (2000-05-01), Gehlot
patent: 6082764 (2000-07-01), Seki et al.
patent: 6144366 (2000-11-01), Numazaki et al.
patent: 6298311 (2001-10-01), Griffin et al.
patent: 6373557 (2002-04-01), Mengel et al.
patent: 10-147206 (1998-06-01), None
patent: 10315906 (1998-12-01), None
patent: 11020606 (1999-01-01), None
patent: 11112968 (1999-04-01), None
patent: 11201822 (1999-07-01), None
Pending U.S. Berenz et al. patent appln. Ser. No. 09/607,090, filed Jun. 29, 2000 entitled Human Presence Detection, Identification and Tracking Using a Facial Feature Image Sensing System for Airbag Deployment.
Pending U.S. Berenz et al. patent appln. Ser. No. 09/621,772, filed Jul. 21, 2000 entitled Application of Human Facial Features Recognition to Automobile Security and Convenience.
J. Krumm, G. Kirk; “Video Occupant Detection for Air Bag Deployment”, Sandia Commerically Valuable Information, SAND97-0408, Feb. 1997, pps. 1-20.
Berenz John J.
Dunbridge Barry
McIver George W.
Niesen Joseph W.
Shreve Gregory A.
Ahmed Samir
Tarolli, Sundheim Covell & Tummino L.L.P.
TRW Inc.
LandOfFree
Optimized human presence detection through elimination of... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Optimized human presence detection through elimination of..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Optimized human presence detection through elimination of... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3289112