GB2393837A

GB2393837A - Traffic violation processing method

Info

Publication number: GB2393837A
Application number: GB0317855A
Authority: GB
Inventors: David Freinkel
Original assignee: Individual
Current assignee: Individual
Priority date: 2002-07-31
Filing date: 2003-07-30
Publication date: 2004-04-07
Anticipated expiration: 2023-07-30
Also published as: IL151025A0; GB2393837B; GB0317855D0

Abstract

A method of processing images of moving vehicles to provide documentary evidence of traffic violations comprises the steps of receiving a plurality of video images having an associated traffic violation data set, analysing the images using a processor and producing a condensed set of images. In one embodiment the analysing step includes using OCR techniques to produce a digital representation of the vehicle license plate and a quality evaluation of the image. The method includes the further steps of manually observing the condensed set of images to verify the occurrence of a violation. In a further embodiment the processor may control the speed and direction of the video playback according to the time data recorded in the violation data set.

Description

Trafflc Violation Processing Method FIELD AND BACKGROUND OF THE INVENTION

The invention relates broadly to a traffic violation processing system, and, 5 in particular, to a method of recording and storing vehicle images and textual information (e.g., alphanumeric data from the license plate of a monitored vehicle) and processing the stored data to provide an integrated evidentiary record for traffic violation enforcement purposes.

In law enforcement, a reliable witness that is incapable of perjury is 10 needed to substantiate the actions taken by the law enforcement officer, and to protect the law enforcement officer against false allegations by the persons involved in the incident. An excellent witness of this type is video recording of the incident, which can reviewed after the incident by the officer himself so he can prepare an accurate written report, investigators, prosecutors and/or judges to 1 S witness firsthand the incident as it actually happened. These video recordings eliminate conflicting individual interpretations of the incident, since it was recorded while it was happening. As a result, lengthy trials based on the individual interpretations will largely become a thing of the past.

Existing traffic monitoring systems often include cameras configured to 20 take a photographic image of the violators. Sometimes, the traffic monitoring systems are located in a monitoring vehicle, e.g., a police car. In other cases, the l

systems are not located in a vehicle, but rather are stationary, positioned close to the roadway, e.g., on the ground or elevated on a pole.

United States Patent No. 5,381,155 to Gerber discloses a photographic system including a video camera using a high density pixel CCD (Charge 5 Coupled Device) image converter, a flash unit, a frame "grabber" which is activated when the speeding vehicle reaches the exact position for the car and driver to be photographed, and a video recorder or frame memory which may be solid-state RAM (Random Access Memory) capable of recording an entire frame. The date, time of day, license plate number and associated information 10 are added to the recorded frame. The entire frame is then transmitted, over a telephone line, to police headquarters where it is recorded and printed out, for later use.

United States Patent No. 5,677,979 to Squicciarini, et al., discloses a video incident capture system that integrates the outputs of a video camera, a radar 15 unit, a wireless microphone, a remote control and a wireless microphone to produce a comprehensive video recording of an incident from its beginning to the end.

Sometimes, the systems are controlled by an operator who is present during monitoring. The systems may also be operated automatically, without an 20 operator being present. In either case, the images captured by the camera on film are typically stored in a film magazine. The photographic images recorded on film by the camera are often used to form evidentiary records for purposes of proving the existence of a violation.

One of the shortcomings associated with storing images on film is that the number of images that can be stored tends to be limited by the space in the film magazine. The system's ability to record violations is thus limited by the capacity of the film magazine. When the film magazine reaches its maximum capacity, 5 the system can no longer record images of violators. Because it is often difficult if not impossible to accurately estimate the number of violators at a given location, it is also difficult to determine when the capacity of the film magazine has been reached.

Another shortcoming of these earlier systems is that the operator must 10 often make frequent trips to a central processing location to deliver the film for developing and processing. The need to make such frequent trips can occupy a great deal of time. The expenditure of time is magnified when numerous traffic monitoring systems are located in different geographic locations at inconvenient distances from central processing location.

15 Yet another shortcoming of the earlier systems is the labor-intensive process of matching the vehicle in each photographic image with registered owner information, in order to prepare traffic citations or an evidentiary record.

For example, after the image is developed, the photograph is manually examined by an operator to identify the license plate number. Next, the license plate 20 number is correlated with a listing of registered vehicle owners to determine the name of the owner, after which the traffic citation is prepared. This is done typically by manually inputting information relating to the traffic violation, then

mailing the traffic citation to the registered owner. This cumbersome process is inefficient and results in high costs and expenditures of time.

A continuing need therefore exists for a traffic violation processing system that overcomes one or more of the above-mentioned shortcomings.

SUMMARY OF THE INVENTION

The present invention is a method of processing images of moving vehicles in order to provide a compact and integrated evidentiary record for traffic violation enforcement purposes. The method disclosed herein 10 significantly reduces the video footage and associated data requiring manual verification, such that the labor-intensive processing procedure is streamlined.

According to the teachings of the present invention there is provided, a method of processing images of moving vehicles to provide a compact and integrated evidentiary record for traffic violation enforcement purposes, the 15 method including the steps of: (a) providing a plurality of images, each of the images including a vehicle image; (b) automatically monitoring a license plate field within each image of the images, so as to produce: (i) a digital

representation of the license plate field, and (ii) a quality evaluation of the digital

representation; (c) automatically producing, based on the monitoring, a 20 condensed set of images; (d) manually observing each image in the condensed set of images to visually verify a license plate number for each particular subject vehicle represented in the plurality of images, and (e) manually observing

vehicle speed information associated with each subject vehicle to identify a traffic violation.

According to further features in the described preferred embodiments, the 5 plurality of images is a video footage.

According to further features in the described preferred embodiments, the quality evaluation is compared with a minimum quality threshold.

According to further features in the described preferred embodiments, the video footage is evaluated according to the presence of a vehicle e.g. bus lanes.

10 According to further features in the described preferred embodiments, the method further includes the step of: (f) automatically monitoring a speed window within each of the images using optical character recognition (OCR) software, and (g) automatically determining, based on the monitoring and at least one pre determined criterion if a speed violation is severe enough to warrant a legal 1 5 action.

According to further features in the described preferred embodiments, the speed window includes a mark for denoting whether a speed violation is associated with a particular image of the images.

According to further features in the described preferred embodiments, the 20 speed window includes an image of a speed value associated with the particular subject vehicle.

s

According to further features in the described preferred embodiments, the plurality of images includes a plurality of subject vehicle images for at least one particular subject vehicle, and wherein the plurality of subject vehicle images is condensed on a weighted, multiple parameter basis.

5 According to further features in the described preferred embodiments, the multiple parameter basis includes OCR quality evaluation and speed value.

According to further features in the described preferred embodiments, the plurality of images includes a plurality of subject vehicle images for at least one particular subject vehicle, and wherein at least one image of the subject vehicle I O images is selected on a weighted, multiple parameter basis.

According to further features in the described preferred embodiments, at least one image of the subject vehicle images is a sole image.

According to further features in the described preferred embodiments, the speed value is range-oriented.

15 According to further features in the described preferred embodiments, the speed value is classified in one of at least three speed ranges.

According to further features in the described preferred embodiments, each image of the plurality of subject vehicle images is converted to a digital image. 20 According to further features in the described preferred embodiments, each selected image from the subject vehicle images has a display block output associated therewith.

According to further features in the described preferred embodiments, each of the plurality of images is a digital image.

According to further features in the described preferred embodiments, the quality evaluation is compared with a minimum quality threshold in step (c).

5 According to yet another aspect of the present invention there is provided a method of accumulating information from a passing vehicle so as to enable substantially immediate communication with the passing vehicle, the method including the steps of: (a) identifying a vehicle in which a traffic violation has been committed; (b) identifying a mobile phone number of a mobile phone 10 disposed in the vehicle, and (c) automatically composing and sending a message to the mobile phone, the message relating to the traffic violation.

According to further features in the described preferred embodiments, the traffic violation is a speed violation.

According to further features in the described preferred embodiments, the 15 method further includes the step of: (d) providing a system including at least one mobile phone detector.

According to further features in the described preferred embodiments, the system includes at least two mobile phone detectors, and wherein a fixed distance is interposed between the detectors.

20 According to further features in the described preferred embodiments, the system further includes an SMS transmitter for sending a message to the mobile phone.

According to further features in the described preferred embodiments, the system further includes a database containing mobile phone numbers associated with particular license plate numbers.

According to further features in the described preferred embodiments, the 5 method further includes the step of: (e) reading a license plate of the vehicle in which a traffic violation has been committed.

According to further features in the described preferred embodiments, the system further includes an automatic license plate reading (LPR) device.

I O BRIEF DESCRIPTION OF THE DRAWINGS

The invention is herein described, by way of example only, with reference to the accompanying drawings. With specific reference now to the drawings in detail, it is stressed that the particulars shown are by way of example and for purposes of illustrative discussion of the preferred embodiments of the present 15 invention only, and are presented in the cause of providing what is believed to be the most useful and readily understood description of the principles and

conceptual aspects of the invention. In this regard, no attempt is made to show structural details of the invention in more detail than is necessary for a fundamental understanding of the invention, the description taken with the

20 drawings making apparent to those skilled in the art how the several forms of the invention may be embodied in practice.

In the drawings:

FIG. 1 is on overhead view of a traffic monitoring system positioned in a monitoring vehicle next to a roadway, FIG. 2 is a drawing showing a traffic image, including a vehicle sub image and a license plate area (LPA) sub-image, 5 FIGS. 3a,3b,3c are flow diagrams showing the processing of images and traffic data in which the input information includes a plurality of video images, in accordance with a specific embodiment of the invention, FIG. 4 is a flow diagram showing the processing of images and traffic data for a digital image input, in accordance with another specific embodiment of the 10 invention, and FIG. 5 is a schematic diagram of a set-up for accumulating information from a passing vehicle so as to enable immediate communication with the driver, passenger, and/or owner of the vehicle.

DESCRIPTION OF THE PREFERRED EMBODIMENTS

15 The present invention is a method of processing images of moving vehicles in order to provide a compact and integrated evidentiary record for traffic violation enforcement purposes. The method disclosed herein significantly reduces the video footage and associated data requiring manual verification, such that the labor-intensive processing procedure is streamlined.

20 The principles and operation of the method of the present invention may be better understood with reference to the drawings and the accompanying description.

Before explaining at least one embodiment of the invention in detail, it is to be understood that the invention is not limited in its application to the details of construction and the arrangement of the components set forth in the following description or illustrated in the drawing. The invention is capable of other

5 embodiments or of being practiced or carried out in various ways. Also, it is to be understood that the phraseology and terminology employed herein is for the purpose of description and should not be regarded as limiting.

One aspect of the invention involves processing traffic images. Referring now to the drawings, Figure I illustrates a video camera 10 monitoring a moving 10 vehicle 14 to determine whether a traffic regulation (in this case, the speed limit) is being violated.

The particular device used for image capture is beyond the scope of the invention. Traffic monitoring systems with video cameras are generally well-

known. Traffic monitoring systems that include a video camera are disclosed 15 United States Patent No. S,677,979 to Squicciarini, et al., which is hereby incorporated by reference for all purposes, to the extent not inconsistent with this invention. As shown in Figure 2, a traffic image that includes the image of vehicle 14 is captured using video camera 10 (not shown). The term "traffic scene image" 20 16 refers to the image of the entire scene captured by the camera. The "vehicle image" 18 is actually a sub-image, i.e., a portion of the traffic scene image that includes the image of the vehicle. The "LPA image" 22 is the sub-image portion of the vehicle image containing the license plate (or license tag) . Each passing

car, or alternatively, passing cars selected manually by a policeman or operator, or automatically according to a pre-determined criterion, is filmed, producing a video footage. The video footage is typically 25 frames per second and runs 2-7 seconds per filmed car, such that approximately 50-l 75 frames are produced for 5 each filmed car.

As is well known in the art, and as exemplified in the above-referenced United States Patent No. 5,677,979, interposed between video camera 10 and the Video Recorder is an "ON SCREEN DISPLAY" (OSD) circuit, which superimposes specified information on the video signals generated by video 10 camera 10. The specific information is in the form of alpha- numerical characters arranged in the form of a display block (also known as frame overlay, overlay imprint, etc.) 24 normally located near the bottom of the recorded image, as shown in FIG. 2. The specific information superimposed on video signals generated by video camera 10 is permanently recorded on the videotape along 15 with the incident recorded by video camera 10 and becomes a permanent part of the recorded information.

As shown in FIG. 2, display block 24 preferably contains the month, date and year in the conventional numerical format as well as the time in hours, minutes and seconds. Display block 24 will also contain an identification of the 20 filming source (e.g., stationary unit I.D. and location; law enforcement vehicle I.D., in the case of a mounted video system, etc.), the speed of an identified vehicle, the speed limit, the speed of the law enforcement vehicle, when applicable, and the date and time of the image capturing. Typically and ll

preferably, a separate display block, termed "speed window" 26, is used to display the speed of vehicle 14. Alternatively, speed window 26 contains a mark, such as an asterisk, in the event that vehicle 14 is speeding, or speeding relative to some pre-defined criterion (e.g., 10% above the speed limit).

5 The inherent problem with current video-based violation capturing systems is that it is difficult and time consuming to extract the information necessary to prove a traffic violation. A single recorded video instance of a moving vehicle, whether it is actually violating the law or not, may have between 2 and 7 seconds of video footage associated with it. The total time of a video cassette handed in 10 to the back office for processing might be up to 3 hours, with perhaps 50% of the recorded vehicles meeting the violation criteria. Typically, an operator has to review the entire video in real time, note when a violation occurs, stop the playback at the correct position, and manually extract and record the information.

The time taken for this manual violation extraction process can be anywhere 15 between 10 seconds and one minute, depending on the diligence and dedication of the reviewer. Thus, the back office processing of a 3 hour video cassette containing 7 second instances of moving vehicles, of which 50% require further violation processing, will consume between 5 and 15 hours.

In order to ensure a speed enforcement presence on the road, law 20 enforcement departments invest much capital, expertise and expense in multiple front-end enforcement systems, but without due diligence for the difficulties in back office processing of the captured media. To make matters worse, strict government control over front-end violation capturing equipment limits the

options available for systems that lend themselves to efficient back office processing. Thus, the law enforcement departments are hard-pressed to find methods for making currently-used systems much more efficient and less labor intensive. 5 The traffic violation video recognition (TVVR) system and method of the present invention automatically reads and extracts information from video footage. The TVVR system reads the license plate of the vehicle of the violator, and the overlay imprint of the violation information. The information is stored by TVVR in digital format.

10 As shown in Figure 3a, a video footage input 30 is automatically monitored in step 31 by an automated optical monitoring (AOM) unit, without human supervision and/or intervention. The TVVR system determines when a violation begins. This is achieved by continually reading the recorded vehicle speed image (e.g., as it appears in speed window 26, as shown in Figure 2) and 15 producing a digital vehicle-speed value, and subsequently reading and processing this digital value in step 32, to determine whether the speed exceeds the legal limit or some other predefined criterion.

In the event that speed window 26 contains a mark, such as an asterisk, for indicating that the vehicle is speeding, speed window 26 for each video frame is 20 monitored in step 31 by the AOM unit to determine (step 32) whether a particular frame contains evidence of a traffic violation.

In the event that the optical monitoring in step 31 proves negative, i.e., that no traffic (speed) violation is detected (or that the violation is not of interest

l based on the above-mentioned pre-defined criterion), the video footage undergoes editing (step 34), and the particular frame is removed. Preferably, a condensed version of video footage input 30 is produced. In this case, if the optical monitoring in step 31 proves negative, the particular frame is simply not 5 saved or copied to the condensed output (step 35).

If the optical monitoring in step 31 detects a violation, the information from display block 24, which is usually in digital format, is output (step 36) as a log file, or to a computer via dynamic data exchange (DDE) or the like. The information from display block 24 includes frame number information (or the 10 like) for associating with a particular frame or image in the condensed output (step 3 S) produced by the editing process (step 34). The length of the condensed output, in terms of frames or images, is often less than 15% of the length of the video footage input, and may be less than l %, as will be illustrated herein below.

Consequently, the verification process (step 37), which was heretofore labor 15 intensive, is significantly shortened and simplified.

As is evident from the description provided above, the condensed output

may be in the form of video footage. Alternatively, the condensed output may be in digital form.

In a preferred embodiment, the optical monitoring in step 31 includes OCR 20 monitoring. Preferably, step 31 includes reading the image containing the registration plate of the subject vehicle of a particular frame. In processing step 32, the OCR recognition quality associated with the registration plate number is determined, and compared with a predetermined quality value. In the event that

the OCR recognition quality is below the pre-determined quality value, the video footage undergoes editing (step 34) such that the particular frame is either removed, not copied, or not saved in digital form.

As mentioned hereinabove, the input video footage is typically 25 frames 5 per second and runs 2-7 seconds per filmed car, such that approximately 50-175 frames are produced for each filmed car. It would be advantageous to have a greatly reduced number of frames, even for those vehicles for which a traffic violation has been recorded. One possibility would be to evaluate, in step 31, which frame belonging to a particular filmed car has the best OCR recognition 10 quality for the registration plate number. This processing step is within the realm of existing OCR technology, and is known to those skilled in the art. All other frames associated with this particular filmed car are not saved in (or copied to) the condensed output 35. Subsequently, in verification step 37, the operator views the individual selected frames, preferably one for each vehicle, and l 5 determines, based on the speed value appearing in the frame overlay, whether a violation has occurred. The operator also views the registration plate number, to make a positive identification. The success rate for this operation is greatly enhanced by the selection of the frame having the best OCR recognition quality for the registration plate number.

20 It has been observed by the inventors that in many or even most instances, the frame containing the best resolution for the registration plate number image is not identical with the frame containing the maximum speed value. In a preferred embodiment, processing step 32 selects a single frame of a particular IS

filmed car, on a weighted, multiple parameter basis. Typically, the parameters include OCR recognition quality and speed value. An example is provided below, based on illustrative parameter values presented in Table 1.

The optimal frame, based solely on OCR recognition quality of the vehicle 5 plate number, is frame 0 (86%). Unfortunately, this particular frame has the lowest speed value, 89 kph, among all 10 frames. A processing step that selects the optimal frame based solely on speed value would select frame 10 (103 kph).

Here, however, the OCR recognition quality of the vehicle plate number is the lowest of all frames. It might be advantageous to select the frame having the 10 highest speed value for a pre-determined threshold of the OCR recognition quality of the plate number. Thus, for a cutoff value of 80% (min.), the selected frame would be frame no. 175, having a speed value of 99 kph.

Alternatively, a weighted formula including the parameters of plate resolution quality and speed value might yield frame no. 125 as the optimal 15 frame: the speed is relatively high (96 kph) and the plate image resolution quality is extremely high (84%) as compared with the other frames.

It must be emphasized that the OCR resolution quality of the speed value is high for all frames, which is typical for many applications, however, this parameter could also be included in the selection criteria provided to the 20 processor.

1b

TABLE 1

Frame No. OCR Resolution Oualit,v Speed Value OCR Resolution Ouality -

Vehicle Plate No. (KPH) Speed Value 0 86% 89 92%

25 85% 89 92%

50 83% 91 90%

75 82% 92 93%

100 83% 94 90%

125 84% 96 92%

150 80% 97 78%

175 81% 99 92%

200 80% 101 91%

225 78% 103 91%

Finally, other pre-determined criteria could be included, e.g., customized 5 criteria of a particular law-enforcement agency. For example, if the speed limit in the previous example is 70 KPH, the customized criteria might include the following ranges: RANGE 1: up to 82 E(PH no fine RANGE II: up to 90 KPH fine 10 RANGE III: up to 100 KPH severe fine RANGE IV: above 100 KPH suspended license

In this case, the criteria might be range-oriented, with all frames having a speed value in excess of 100 KPH (i.e., frame nos. 200 and 225) being examined for best OCR resolution quality to obtain the optimal frame, and/or with all with all frames having a speed value between 90 KPH and 100 KPH (i.e., frame nos. 50, 5 75, 100, 125, 150, and 175) being compared for best OCR resolution quality to obtain the optimal frame for Range III. Alternatively, the optimal frame number in each range could be selected, and the processor could decide between them based on pre- determined criteria. Finally, it is possible to select more than one frame to represent a particular filmed car.

l 0 In a further embodiment, processing step 32 uses the presence of a license plate to determine if a violation has occurred. E.g., bus lane violations.

Figure 3b is a variation of Figure 3a, wherein the AOM monitoring (step 31) is replaced by signal detection (step 31a). The detected signal is produced by the interpretation of the digitized overlay imprint of the violation data on the 15 video footage.

Figure 3c is a variation of Figure 3b wherein the video footage is played (output) by a video player both to a processor and to an AOM. The video footage contains summary information, typically video information, within the

footage. In AOM monitoring step 30a, violation information from the summary

20 information in the video footage is selectively identified (e.g., by identifying marked frames) and output as a digital violation summary report to the processor.

In processing step Job, the time delay or interval between each of the violations is determined from the digital violation summary report. These time

l8

intervals are fed to the video player such that the video player will accurately fast-forvard to the next violation (step 30c), thus saving considerable processing time by disregarding irrelevant footage.

Preferably, the processing step 30b is self-synchronizing, in that an event 5 that is time-identified by the AOM unit is synchronized with the identical event that is to be found in the playback of the full video footage that is input to the processor. This synchronization can be performed one or more times over the time events that are stored within the digital violation summary report.

Preferably, the synchronization is performed frequently, and mostpreferably, the 10 synchronization is performed each time a time event is retrieved/detected from the digital violation summary report. This compensates for various inaccuracies

in the time recordings, particularly those associated with the full video footage.

Preferably, processing step 30b distinguishes between actual violation footage and violation summary footage, such that the processor automatically

15 and deterministically identifies the frame location within the entire footage played by the video device. This means that no human operator need supervise or intervene in the processing, such that the process is fully automatic through licence plate recognition (LPR) processing step 42.

In one preferred embodiment, the processor identifies signal frequencies 20 and/or time sequences to distinguish between actual violation footage and violation summary footage. For example, a negative time interval between

violations would indicate that the footage has passed from actual violation footage to violation summary footage. Similarly, an increase in violation

A

frequency, above a pre-determined level, for a sustained period, indicates that the footage has passed from actual violation footage to violation summary footage.

Preferably, fast-forvarding step 30c is performed such that the video player changes from fast-forvard mode back to play mode at a predetermined 5 time-interval prior to the calculated time of the next violation signal. The pre-

determined time interval enables the image of the video player to stabilize in time for video footage editing (step 34). In step 34, the condensed set of video images is further condensed. A select number of images for each of the violations is captured and saved, and output to a licence-plate recognition 10 processing step (step 42). Similarly, any requisite violation data (e.g., date and time of the violation) from step 34 is output to licence-plate recognition processing step (step 42).

Preferably, at least two images per violation are captured in step 34. In this case, licence-plate recognition processing step (step 42) further condenses 15 the image data, such that a single best image, as determined by the LPR results, is stored for subsequent manual verification (step 37) . The requisite violation data for each violation is stored in a log file in step 42, and is presented for manual verification (step 37) along with the best image.

Optionally, licence-plate recognition processing step (step 42) is 20 performed off-line. In this case, after video footage editing (step 34) has been performed on a set of images corresponding to a particular violation, fast-

forvarding step 30c is performed to arrive atlnear the next violation, using the above-described time-interval data. This process continues until all violations as

have been processed through step 34, and the condensed output and log of violation data are output for processing in licence-plate recognition processing step (step 42).

Figure 4 is a schematic diagram based on Figure 3a, as applied to an input 5 of a sequence of digital images (step 40), produced, by way of example, by a digital video camera. The sequence of digital images is automatically monitored in step 41 by optical character recognition software (OCR), without human supervision and/or intervention. Typically, the speed associated with each digital image is recorded in digital form, hence, the OCR monitoring is usually needed 10 solely for reading the image containing the registration plate of the subject vehicle. Based on the digital images, associated speed values, and OCR monitoring and evaluation data from step 41, a processor then produces (step 42) and outputs (step 45) a condensed digital output containing digital images for manual 15 verification in step 47. The processing in step 42 may be performed in several ways: (1) removal of images in which a speed violation has not occurred, or in which the speed violation is below a pre-determined value; (2) removal of images in which an OCR evaluation parameter is below 20 a predetermined value, and/or (3) select one (or more than one) digital image for each vehicle (or violating vehicle, as in (1)), based on a combination of speed value and OCR evaluation parameter criteria.

Subsequently, in verification step 47, the operator views the condensed digital output images, typically one image per vehicle, and determines or confimns, based on the speed value appearing in the frame overlay (and/or in the display block output in step 46), the occurrence of a violation. The operator 5 views the registration plate number on each digital output image, to make a positive identification.

The selection of one (or more than one) digital image for each vehicle based on a combination of speed value and OCR evaluation parameter criteria has been described in detail hereinabove, in conjunction with Figure 3a. The 10 output of information from a display block (step 46) as a log file, or to a computer via dynamic data exchange (DDE) or the like, in which each set of display block information is associated with a particular digital image, is substantially identical to step 36 in Figure 3a.

The length of the condensed output, in terms of frames or images, is often 15 less than 1% of the length of the digital image input. Consequently, the verification process (step 47) is extremely fast.

Another aspect of the present invention is a method of accumulating information from a passing vehicle so as to enable immediate communication with the driver of the vehicle, and to inform him of the recently witnessed 20 speeding (or other traffic) violation, and/or to inform him on outstanding fines.

The speed of the vehicle is determined, and if it is above the local limit, an SMS with the violation infonnation is automatically and immediately sent to the detected or otherwise identified mobile phone. The method is operative for

vehicles traveling in both directions. The information transmitted to the mobile phone can include any of the following: 1. The vehicle is traveling above the local limit 5 2. Local speed limit 3. License plate number 4. Owner details 5. Vehicle speed 6. Outstanding fines (count) 10 7. Outstanding fine (total amount) 8. How many (other) times this particular mobile phone has been detected in a speeding vehicle Mobile Phone Number Identification The mobile phone number of the violator can be identified by various methods: l. By database lookup of the mobile phone number against the detected license plate number 20 2. By a mobile phone number detecting device 3. By accessing the GPS positioning system to which the passing mobile phone (owner) subscribes 4. By accessing the locating service provided by the mobile phone network providers Methods 2- 4 identify the presence of the mobile phone in the vehicle, but do not positively determine that the phone belongs to the driver or car owner.

Method 1 positively determines the mobile phone number associated with the detected license plate number. In many or most cases, it is the owner of the 30 vehicle who is driving when the traffic violation is committed. In most other

instances, the driver is driving with the permission of the registered owner, such that the message is of value. Finally, on the infrequent occasion when the driver is driving without the permission of the registered owner, the message may be of extreme importance to both lawenforcement agencies and to the owner, e.g., in 5 the event that the vehicle has been stolen Speed Measurement The speed of the vehicle can be determined by: 10 1. At least one conventional detector, e.g., a single conventional detector placed at the side of the road; 2. Two mobile phone number detectors placed at a known distance apart, along the length of the road.

15 Modes of Operation (Fieure 5) Option 1 Site A includes a mobile phone number detector and a storage device. Site 20 B includes an additional mobile phone number detector, database, storage device and SMS transmitter. Site A detects a mobile phone number from a passing vehicle and records it. Site B detects a mobile phone number from a passing vehicle, records it and determines whether the same mobile phone number exists at site A. If so, the speed of the mobile phone traveling from site A to site B is 25 calculated. If the speed is above the local limit or above some other pre-

determined criterion, site B transmits an SMS to the speeding mobile phone containing all or some of the information listed above.

Option 2 Site A consists of mobile phone number detector, conventional speed measuring device, database and SMS transmitter. The speed of a vehicle passing 5 site A is determined by a conventional speed measuring device. Substantially simultaneously or shortly thereafter, the mobile phone number is detected. If the speed of the vehicle is above the speed limit, an SMS is immediately, automatically transmitted to the mobile phone containing all or some of the information listed above.

Option 3 Site A consists of a conventional speed measuring device, LPR (license plate recognition) device, mobile phone number detector, database and SMS 15 transmitter. The speed of a vehicle passing site A is determined by a conventional speed measuring device. The license plate of the passing vehicle is read by an automatic LPR device, and substantially simultaneously, the mobile phone number is detected. The detected license plate number is compared to a database to extract fine history information. This information, and if necessary, 20 current speed violation information, is immediately transmitted via SMS to the detected mobile phone number.

Although the invention has been described in conjunction with specific embodiments thereof, it is evident that many alternatives, modifications and 25 variations will be apparent to those skilled in the art. Accordingly, it is intended to embrace all such alternatives, modifications and variations that fall within the as

spirit and broad scope of the appended claims. All publications, patents and patent applications mentioned in this specification are herein incorporated in

their entirety by reference into the specification, to the same extent as if each

individual publication, patent or patent application was specifically and 5 individually indicated to be incorporated herein by reference. In addition, citation or identification of any reference in this application shall not be construed as an admission that such reference is available as prior art to the

present invention.

Claims

WHAT IS CLArMED IS: I. A method of processing images of moving vehicles to

provide a compact and integrated evidentiary record for traffic violation enforcement purposes, the method comprising the steps of: (a) activating a video-playing device to provide a plurality of images, each of said images including a vehicle image, each of said images having an associated traffic violation data set; (b) providing said plurality of images and said associated traffic violation data to a processor, and (c) processing said images using said processor so as to produce a condensed set of images having traffic violations and a corresponding condensed set of digital violation data, wherein said processor utilizes time data within said associated traffic violation data set to control a playback speed of said video-playing device.
2. The method of claim 1, wherein said playback speed is a forward playback speed.
3. The method of claim 1, wherein said playback speed is a reverse playback speed.
4. The method of claim 1, further comprising the step of: (d) processing said condensed set of images and said condensed set of digital violation data in a licence-plate recognition (LPR) processor to produce a second condensed set of images and a second condensed set of digital violation data, said second set of data containing digital license plate data.
5. The method of claim 1, further comprising the step of: (e) manually observing each image in said condensed set of images to visually verify a license plate number for each particular subject vehicle represented in said plurality of images.
6. The method of claim 1, wherein said associated traffic violation data set is a digital data set.
7. The method of claim 6, wherein said digital data set is obtained from an optical data set prior to step (b).
8. A method of processing images of moving vehicles to provide a compact and integrated evidentiary record for traffic violation enforcement purposes, the method comprising the steps of: (a) providing a plurality of images, each of said images including a vehicle image; (b) automatically monitoring a license plate field within each image

of said images using optical character recognition (OCR) software, so as to produce: (i) a digital representation of said license plate field, and

(ii) a quality evaluation of said digital representation; (c) automatically producing, based on said monitoring, a condensed set of images; (d) manually observing each image in said condensed set of images to visually verify a license plate number for each particular subject vehicle represented in said plurality of images, and (e) manually observing vehicle speed information associated with each said subject vehicle to identify a traffic violation.
9. The method of claim 8, wherein said plurality of images is a video footage.
10. The method of claim 9, wherein in step (c), said quality evaluation is compared with a minimum quality threshold.
11. The method of claim 9, further comprising the step of: (f) automatically monitoring a speed window within each of said images using optical character recognition (OCR) software, and (g) automatically determining, based on said monitoring, if a speed violation has occurred.
12. The method of claim 9, further comprising the step of: (h) automatically monitoring a speed window within each of said images using optical character recognition (OCR) software, and (i) automatically determining, based on said monitoring and at least one pre-determined criterion if a speed violation is severe enough to warrant a legal action.
13. The method of claim 11, wherein said speed window includes a mark for denoting whether a speed violation is associated with a particular image of said images.
14. The method of claim I 1, wherein said speed window includes an image of a speed value associated with said particular subject vehicle.
15. The method of claim 11, wherein said plurality of images includes a plurality of subject vehicle images for at least one said particular subject vehicle, and wherein said plurality of subject vehicle images is condensed on a weighted, multiple parameter basis.
16. The method of claim 15, wherein said multiple parameter basis includes said quality evaluation and speed value.
17. The method of claim 11, wherein said plurality of images includes a plurality of subject vehicle images for at least one said particular subject vehicle, and wherein at least one image of said subject vehicle images is selected on a weighted, multiple parameter basis.
18. The method of claim 17, wherein said multiple parameter basis includes said quality evaluation and speed value.
19. The method of claim 17, wherein said at least one image of said subject vehicle images is a sole image.
20. The method of claim 18, wherein said speed value is range oriented.
21. The method of claim 18, wherein said speed value is classified in one of at least three speed ranges.
22. The method of claim 17, wherein each image of said plurality of subject vehicle images is converted to a digital image.
23. The method of claim 17, wherein each selected image from said subject vehicle images has a display block output associated therewith.
24. The method of claim 8, wherein each of said plurality of images is a digital image.
25. The method of claim 24, wherein in step (c), said quality evaluation is compared with a minimum quality threshold.

3l
26. The method of claim 24, wherein said plurality of images includes a plurality of subject vehicle images for at least one said particular subject vehicle, and wherein said plurality of subject vehicle images is condensed on a weighted, multiple parameter basis.
27. The method of claim 26, wherein said multiple parameter basis includes said quality evaluation and speed value.
28. The method of claim 26, wherein said at least one image of said subject vehicle images is a sole image.
29. The method of claim 27, wherein said speed value is range-

oriented.
30. A method of accumulating information from a passing vehicle so as to enable immediate communication with the driver of the vehicle.
31. The method of claim 30, where the accumulated information is the mobile phone number of the mobile phone in the vehicle.