Television – Special applications – Observation of or from a specific location
Reexamination Certificate
2000-01-11
2004-06-08
An, Shawn S. (Department: 2613)
Television
Special applications
Observation of or from a specific location
C348S143000, C348S161000, C340S932200, C340S933000, C340S988000, C340S989000, C340S990000, C382S104000, C382S105000
Reexamination Certificate
active
06747687
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to computer recognition systems, and more particularly to systems that record the passing of a vehicle at one camera location and then recognize the same vehicle passing a second camera at another location.
DESCRIPTION OF THE PRIOR ART
There are many reasons that a robot that could recognize individual vehicles could be useful. Traffic ticketing systems that photograph speeders and red-light-runners and mail the drivers notices of violation are well known. Such systems generally require a high quality photograph that clearly shows the car and its license plate are needed for convictions. Pulnix America, Inc., (Sunnyvale, Calif.) markets several cameras for such systems that can image high quality photographs even under adverse conditions. The Netherlands Government recently tried a freeway speed-enforcement system that used a “matcher” technology. A three kilometer stretch of the A2-freeway between Rotterdam and Amsterdam was outfitted with two sets of three cameras, one camera for each lane. A “visual signature” of the vehicle and time-of-passing information was sent to a database from the first set of cameras as each car entered the measured three kilometer trap. As each car passed the second set of cameras at the end of the trap, its visual signature and time-of-passing information was matched with that in the database to compute speed. If a violation was detected, the license-plate identity was processed from the visual signature and tickets were issued to the speeders. The “privacy” of non-violators was respected by their license plate information never being extracted into human-readable form if the tests indicated they had not been speeding.
Similar matcher technology was used in the first attendentless video tolling system on the Canadian 407-Highway. Ontario license plates of non-transponder equipped vehicles are automatically read by computer from camera images collected at all 128 entry and exit ramps. The non-Ontario vehicles, or ones with non-readable plates, caused the matcher technology to pair entry and exit vehicles images to assist later inspection by a human.
As can be expected, such recognition systems are not able to extract complete license plate information in every case. But there are many useful applications where only the vehicle needs to be recognized, and the license plate information is of no consequence. This is similar to recognizing a person's face, but not being able to remember their name.
SUMMARY OF THE INVENTION
It is therefore an object of the present invention to provide a recognition system within the video cameras that recognizes the coming and going of particular vehicles.
It is a further object of the present invention to provide a recognition system within the video cameras that automatically computes the residence times of particular vehicles in certain areas.
Briefly, in a vehicle time-of-entry and time-of-exit system embodiment of the present invention, the passing of a vehicle through an entrance generates a “trigger-t
0
” signal. An entrance video camera provides realtime images of entering vehicles. The trigger-t
0
signal causes one video frame of the image of vehicle to be stored or “frame-grabbed” in an entrance subsystem. A time-of-entry clock-time signal is used to time-stamp the image which is then forwarded on a connection to a matcher. The passing of a vehicle through an exit generates a “trigger-t
1
” signal. An exit video camera provides realtime images of exiting vehicles. The trigger-t
1
signal causes one video frame of the image of vehicle to be stored or “frame-grabbed” in an exit subsystem. A clock-time signal is used to time-stamp the image which is then forwarded on a connection to the matcher. An output automatically provides a length-of-stay for each exiting vehicle according to its time-of-entry. The matcher uses the unique characteristics in the images of the vehicles to declare a match. While a license plate number is unique, reading any or all of a license plate is not necessary for the extraction of visual signatures related to particular cars. For example, in the limited variety of cars that would patronize a particular parking lot, the color, trim, style, tires, wheelcovers, collision damage, etc., of each vehicle can be used in combination to distinguish individual ones.
An advantage of the present invention is that a recognition system within the video cameras is provided that can be used in a ticketless parking system.
Another advantage of the present invention is that a recognition system within the video cameras is provided that does not require special equipment to be carried by any vehicles being monitored to quickly find pattern matches.
A further advantage of the present invention is that a recognition system within the video cameras is provided that does not require the licenses plates to be readable of any vehicles-being-monitored to compute length-of-stay in a parking lot.
REFERENCES:
patent: 4603390 (1986-07-01), Mehdipour et al.
patent: 4671650 (1987-06-01), Hirzel et al.
patent: 4774571 (1988-09-01), Mehdipour et al.
patent: 5687249 (1997-11-01), Kato
patent: 5845227 (1998-12-01), Peterson
An Shawn S.
Law Offices of Thomas E. Schatzel
Pulnix America, Inc.
Schatzel Thomas E.
LandOfFree
System for recognizing the same vehicle at different times... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with System for recognizing the same vehicle at different times..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and System for recognizing the same vehicle at different times... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3308956