I-Cube or Integrated Intelligent Imaging (I³) focuses of proactive crime prevention using real time links to multiple internal and external databases in order to generate a warning prior to crime occurring, allowing anticipation of the offence. All the projects I-Cube is involved with have a central theme of using images, as this allows both the real time and covert nature of the proactive crime prevention systems to operate to the best effect.

HOW THE SYSTEM WORKS: As the person or car approaches the entrance a camera captures a number of images from which the license plate or face are extracted. These images are saved, allowing this information to be used in subsequent investigations. The face or license plate is then compared to local databases, allowing greeting of repeat visitors and external databases, such as insurance companies, stolen and suspicious vehicles, SAPS, Metro, etc. If the vehicle is listed as stolen, appropriate action should then be taken. If the vehicle is listed as never have been registered, this is a good vehicle to investigate further. Within a short period of the system being in operation, most vehicles and people have been logged and registered, allowing typically limited resources to be focused on the few remaining exceptions. The ability to leverage the equipment to operate in a marketing role, while still actively focusing on proactive crime prevention, has major advantages in a cash limited environment. If you have more than 50 people a day passing a point where we can place a camera, I-Cube has the experience and ability to turn this into a revenue stream. Please contact I-Cube to explore this further.

WHAT IS REQUIRED: The key to proactive crime prevention using real time links is that the system does not require human intervention, it is fully automatic. The camera automatically captures the image, the facial or license plate is automatically compared to both local and remote databases and if required an alarm is generated. The I-Cube solution can operate locally but using a wide range of communication means (including ADSL, 3G, GPRS, iBURST, Wireless or satellite) a real time link to multiple databases can be established, allowing immediate updates of stolen vehicles as they are added to SAPS or the insurance databases.

The system requires a camera linked to a computer, running the I-Cube Intelligent Software. The software is trained to recognise the item to be logged or counted, be that people, cars or different types of products. Once the item is logged, it can then be compared to pre-set rules or databases, allowing an alarm to be generated if the incorrect number of products exit or a vehicle enters after hours or is listed as not allowed. The ability to link vehicle colour, weight, and driver, biometric and other info as required to the license plate gives the I-Cube system tremendous advantages.

I-Cube can integrate with your existing system or provide dedicated cameras, computers and networks. Please contact I-Cube with your requirements and existing equipment.

WHAT DOES IT COST: I-Cube can provide database links only, software only, some hardware and software or a complete solution, fully installed and maintained, if required? The cost can be a capital amount, a rental per month, a cost per transaction or shared revenue. Costs range from 0.13c per chicken (based on 100 000 birds a day) to 3c per car (based on single entry / exit lanes) to R1 950.00 per month to R49 500.00 for the software and hardware. Please contact I-Cube for your specific requirements.

The following illustrates some examples of the users of this technology. Please contact I-Cube or your local IT / CCTV / DVR / biometric supplier for a customised proposal, taking into consideration your requirements, existing solutions and possible revenue sources.

SCHOOL: The entrance of the school is covered with a camera, which is linked wirelessly to a PC running the I-Cube software. The students, staff and parents are enrolled into a local database, which allows specific messages to be placed on an electronic display, while visitors are directed to the visitor parking area and then to reception. The schools limited manpower resources can then focus on the visitors, rather than the known parents or staff. Please contact I-Cube for more details.

DISTRIBUTION CENTRE: The ability to link the license plate, driver facial image, vehicle type and colour, container number and weight of the truck and weight of goods listed on the invoice prevents unauthorised removal of any items. Productivity per vehicle is up while the number of hours wasted at weigh bridges has been substantially reduced, as the distribution centre no longer allows vehicles out which would be fined for overloading. Please contact I-Cube for more details.

GOLF ESTATE: The existing cameras at the entrance and at key points through the estate were linked to the I-Cube LPR software allowing all visitors and residents to be logged at a number of points through the estate. The system was linked to a national database of suspect and stolen cars, allowing proactive crime prevention. When an alarm is detected, the armed reaction team are called into action, supporting the estate front end guards. A side effect of the I-Cube LPR solution is that the average speed of the vehicles is automatically determined, ensuring the estate remains children friendly. Please contact I-Cube for more details.

OFFICE BLOCK: Parking, or rather the lack of parking within the office block was the reason to call I-Cube. Each tenant has a certain number of bays allocated for staff and visitors. When these bays were occupied and more vehicles were allowed into the office block, major congestion resulted, causing severe frustration for both visitors and office workers. I-Cube utilised the existing cameras to count the number of vehicles into and out of the office block. When the number of vehicles approached critical level, a mobile LPR system was deployed and staff parking in visitor areas were requested to remove their vehicles. Failure to comply led to banning of those vehicles for a certain number of days, again utilising the I-Cube enabled LPR solution. Please contact I-Cube for more details.

NEIGHBOURHOOD WATCH: The problem faced by the neighbourhood watch was not that of limited resources or money, but of how to prevent known criminals from entering the area patrolled by the neighbourhood watch. The exiting dome cameras were linked to the I-Cube LPR solution, plus a mobile LPR solution was provided. All vehicles and people entering the neighbourhood watch area were recorded and match against the known criminals. If matched, these people were approached and asked to explain where they were going and what they were doing. Car theft was reduced substantially while contact crimes no longer occur. Please contact I-Cube for more details.

TRAFFIC DEPARTMENT: The department had a recovery rate of fewer than 30% of all fines issued. I-Cube introduced a road block LPR solution which allowed the traffic department to generate an alarm when a vehicle with an outstanding fine or warrant was detected. These vehicles were then automatically pulled over to a parking area where the drivers were informed of any fines or warrants were issued. The drivers had the option to pay any amounts outstanding via credit card, check or cash via the I-Cube supplied mobile payment systems. The number of outstanding fines has improved substantially to more than 70%. Please contact I-Cube for more details.

TAXI ROUTE MONITORING & ENFORCEMENT: If taxi routes are not regulated commuters along non profitable routes are not serviced and violence between competing taxi associations often results. If the taxi routes are monitored with the I-Cube LPR solution illegal operators are immediately identified and prevented from operating. Where two of these monitoring points are linked, average speed determination can be determined and the number of trips per taxi per day can be calculated and communicated to the taxi owners. Please contact I-Cube for more details.

A NUMBER OF OTHER EXAMPLES EXIST: Please contact I-Cube for more details.

Cameras set to snare city bus lane motorists

MOTORISTS who use bus lanes are set to face fines under a system that could net hundreds of thousands of pounds.

Council officials are drafting a report that recommends putting cameras at three city centre locations as a pilot scheme, before more are introduced across Liverpool.

The level of fines has not yet been decided, but they can legally be set at the same rate as those for parking offences – up to £60, reduced to £30 if paid within 14 days. Last night, the council said cameras were needed because cars were frequently using bus lanes, causing delays to public transport and cancelling out any benefits of designated no-car routes.

The worst stretch, at a location which has not been disclosed, is abused by more than 100 drivers an hour at certain times of the day.

Motoring organisations last night claimed the scheme would be an infringement of drivers’ civil liberties.

But the city’s Lib-Dems leader, Cllr Warren Bradley, said the cameras could help traffic flow in Liverpool. Labour group leader, Cllr Joe Anderson, said fines would be seen as another tax and called for a review of bus lanes as he said they often “cause more problems than they solve”.

But the city’s largest bus operator, Arriva, said the plan was a positive step towards improving the city’s infrastructure and promoting environmentally friendly travel.

A decision on the scheme will be announced at the end of next month.

If approved by the city’s executive board, it could be introduced before the end of the year.

Cllr Bradley said cameras may be a necessary evil to ease traffic through the city.

“Bus lanes are there for a reason, to ease public transport access and egress to and from the city and encourage greener bus travel.

“If cameras can help the council to enforce the lanes, then they may be a good thing.”

Last night, council officials said no decision had been made on sites.

But three hot-spots up for consideration are likely to be Lime Street, St John’s Lane, and Chapel Street, where bus lanes are already in operation.

Councils were given the power to enforce bus lane offences under the Traffic Management Act 2004, the same act that handed them responsibility for parking enforcement.

Police can still give fixed penalty notices for motorists caught in bus lanes, which normally operate during morning and evening rush hours on week days, but unlike cameras officers cannot be constantly on guard.

Last night, Cllr Anderson said: “I don’t think this is a good idea. There’s a huge number of bus lanes in the city that don’t work.

“In many places they cause more problems than they solve. People will see this as another tax, like parking. What we have to remember is that bus lanes should be there to improve public transport. The aim is not to punish drivers, and this seems a ridiculous way to achieve that aim.”

He said the council should sit down with Merseyside police and Merseytravel to review the city’s bus lanes and scrap the ones that do not work well.

Paul Biggs, of the Association of British Drivers, who is against bus lanes and cameras because he believes they infringe drivers’ rights said: “Bus lanes cause problems themselves.

“All the cars have to pile into another lane, it causes traffic, and when you turn left you have to zip into a bus lane.”

A spokesman for the city council said: “The whole purpose of bus lanes is to give priority to public transport where appropriate.

“However, they are frequently misused and we are looking to carry out proper enforcement using new technology.”

“These are only proposals. If they are agreed, there will be a lot of publicity and consultation.”

Phil Stone, managing director for Arriva North West and Wales, said: “Bus priority lanes form a vital part of any city centre transport plan and we welcome any proposal that seeks to enforce them.

“By doing this, it would ease congestion on our routes at busy times of the day, thus making bus travel an even more attractive alternative to the car.”

Question

Subject: OCR for License Plates
Category: Computers
Asked by: dustymack-ga
List Price: $50.00 Posted: 05 Jan 2005 09:46 PST
Expires: 04 Feb 2005 09:46 PST
Question ID: 452412

I'm interested in seeing the various techniques people have used to perform OCR (Optical Character Recognition) on License Plates and their respective accuracy rates (both potential accuracy and actual empirical accuracy). Such techniques would include various template matching methods and neural network architectures. Of particular interest are links to research papers explaining particular methods and software proof-of-concepts.

<img border="0" src="http://pagead2.googlesyndication.com/pagead/imp.gif?event=noiframe&client=ca-answers_728x90&dt=1117580060468&lmt=1117580060&alternate_ad_url=http%3A%2F%2Fanswers.google.com%2Fanswers%2Fgoogle_adsense_script.html&prev_fmts=4_answers_sl&format=4_answers_sl&output=html&channel=a_placement_a_on_a_q_on_q&url=http%3A%2F%2Fanswers.google.com%2Fanswers%2Fthreadview%3Fid%3D452412&region=answer_text&cc=93&u_h=768&u_w=1024&u_ah=768&u_aw=1024&u_cd=32&u_tz=120&u_java=true" />

Answer

Subject: Re: OCR for License Plates
Answered By: maniac-ga on 09 Jan 2005 18:58 PST
Rated:

Hello Dustymack, The general approach apparently taken by several vendors can be summarized as follows: - capture image(s) and select the best (or several good) frame(s) - select the "license plate" portion of image / rectify - isolate the characters for recognition - identify the characters followed by some post processing steps where the identified characters are used (usually a database look up). For an example of this, see http://visl.technion.ac.il/projects/2003w24/ and scroll down a little for a flow chart and several samples of the steps used in this student project. http://www.licenseplaterecognition.com/ A tutorial on license plate recognition by one of the vendors. Includes how license plate recognition can be used for a variety of applications. Specific methods used include: - template matching; match each letter to a specific example. Noted in several papers that this can be quite difficult to apply in general. - structural matching; you look at the shape and connections of a letter. For example, comparing between B, D, 9, and 6 you could recognize the loops and relative position (top, bottom, center) and size but have difficulty distinguishing betwen B and 8. - neural networks; trained with several samples can be resistant to noise but may require a large investment in time / training with different character shapes (font). A very good summary of the issues with automatic recognition of license plates at http://www.photocop.com/recognition.htm Scroll to the bottom of this reference as well for an extensive list of suppliers. If you are interested further in the use of photography in trafic enforcement, I suggest starting at the home page http://www.photocop.com/ and checking several other links as well. Some describe methods used to obscure or work around the use of photo systems in police enforcement. There is also another list of companies (with links) that provide more turn key solutions. A Dutch site, using video capture plus speed measurement to help enforce speed limits and fining the car owner. http://itctraffic.com/videoenforcement.htm#License%20Plate%20recognition Scroll up / down for several good sections of information. In particular note the claimed system accuracy of over 99% on "identification" (apparently is a vehicle present) but only 78% on license plates. Also note this takes images at three locations (start, 800 meter, 3000 meters) for the speed measurement so there is a need in this system to recognize the "same vehicle" in each image. Apparently the Dutch also found it necessary to change the font of their license plates to improve recognition. See http://www.sunpig.com/martin/archives/2003/09/20/dutch_car_license_plates_and_traffic_control/ for an example. Somewhat short on details but does talk about resolution, types of cameras, and related issues at http://www.machinevisiononline.org/public/articles/archivedetails.cfm?id=1206 A look at the more general problem of identifying text in a video and extracting that text: http://www.informedia.cs.cmu.edu/documents/vocr_ieee98.pdf Also has references to license plate capture from video (1997 IEEE conference paper). A recent conference (October 2004) with several OCR / vehicle license plate recognition papers http://www.ieeesmc2004.tudelft.nl/?menu=program.&slotid=149 A couple hundred references to techincal papers found with the search phrase license plate recognition citeseer For example, a couple clicks from the second reference brought me to http://citeseer.ist.psu.edu/572231.html A paper titled "New Methods for Automatic Reading of VLP's (Vehicle License Plates)" which includes several references to / from this paper as well. I strongly suggest using citeseer for technical references if you have specific concepts or phrases to research. Google also has a recent service at http://scholar.google.com/ where I entered the phrase vehicle license plate recognition and had another 800+ references. Note that in this case, the references are free but the full document may require a fee for access. A FEW vendor sites: http://www.anpr.net/041129/index.htm http://www.platerecognition.info/1102.htm Adaptive Recognition Hungary, producer of both hardware and software for license plate recognition. http://www.saic.com/products/transportation/iis/ SAIC provides turnkey and custom systems for recognition as well as integration with with related systems (e.g., recognize loads on trucks, driver photo identification). SAIC claims over 95% accuracy on identification numbers with higher rates if backup data is available. Also indicates a rate of vehicles that can flow through their recognition site (roughly 100 to 300). http://www.cormactech.com/neunet/download.html Not specific to license plates, but this vendor of Neural Network software includes an OCR data sample to train / recognize if you want to get a feel for the issues involved. http://www.htsol.com/Products/SeeCar.html High Tech Solutions, provides several related vehicle license plate systems. Claims use at 2004 Olympics in Greece. http://www.jrwald.com/blue/license/newtech/LPD.html A more restricted system, recognizing license plate numbers during manufacture of the plates (but very high speed - 1000 per hour). Free and for fee software / algorithms are available from links at http://www-2.cs.cmu.edu/~cil/v-source.html for a variety of computer vision applications (including license plates). Search phrases used include: ocr automobile license plate automatic recognition automobile license plate license plate recognition --Maniac

Request for Answer Clarification by dustymack-ga on 14 Jan 2005 09:15 PST
Some of those are very good links. Thank you. I'm specifically interested in the accuracy ratings of specific methods, the links are just a bonus.

Clarification of Answer by maniac-ga on 17 Jan 2005 17:47 PST
Hello Dustymack, I am sorry the answer was not completely satisfactory. Let me clarify the accuracy values I had previously found and provide some additional references. The Dutch system claims only 78% accuracy on license plate recognition (LPR). The SAIC system claims about 95% accuracy for LPR and when combined with a data base lookup, higher rates [presumably by comparing "close" matches with a threshold]. This gives you a range of accuracies - a review of the system implementations should give you an idea of how these accuracies are achieved. Some additional accuracy references are available through searches such as LPR license accuracy VLP license accuracy including sites such as: http://www.cbp.gov/xp/CustomsToday/2001/December/custoday_lpr.xml over 90% accuracy using an automobile sensor, flash unit, rapid camera speed, and interface to government plate databases. Addresses all US and Mexican states, and all Canadian provinces. http://www.singaporegateway.com/optasia/perform.htm A system in place in Singapore with 99.7% accuracy of "good plates". Limited to plates issued in Singapore. http://www.alphatech.com/secondary/techpro/ISD/news-phoenix-lpr.html Description of a system in place at Sky Harbor Airport in Phoenix. Reads 92% of all "readable" plates and within 2 digits 99% of the time. Note this method of doing matches with less than perfect data. Please note there is little information relating the specific methods to achieve the accuracy indicated. I may be able to produce a table like this Method A + B + C => Accuracy X Method A + D + C => Accuracy Y Method E + D + C => Accuracy Z or is the information provided sufficient? --Maniac

Request for Answer Clarification by dustymack-ga on 19 Jan 2005 16:31 PST
Such a table would be absolutely perfect. In particular, when referencing a method I'm interested in the specific method used and not so much the implementation.

Clarification of Answer by maniac-ga on 23 Jan 2005 18:13 PST
Hello Dustymack, The sites that describe specific methods and accuracy appear to be even more rare than I expected. The following summarizes what I could find for specific systems. [1] IPMS Vehicle License Plate Recognition System http://www.singaporegateway.com/optasia/techinfo.htm Methods used for image capture: - standard CCTV, using a single frame (even or odd) sized so characters are at least 18 pixels high - lighting, options for continuous or strobed, visible or near IR - single or multiple images captured Methods for OCR: - breadth first AI for locating plate location in image - no real specifics on OCR method except it allows for "fuzzy" matches, most likely a neural network with confidence levels Sample plates read / accuracy: http://www.singaporegateway.com/optasia/plates.htm This vendor claims over 99.7% accuracy for Singapore plates. [2] US Customs Example http://www.cbp.gov/xp/CustomsToday/2001/December/custoday_lpr.xml Methods used for image capture: - high speed "video camera"; says sensor is sensitive enough for 1/10000 second image but does not state it captures at that rate - lighting appears to be near visibile IR strobe - image capture of both front and rear of vehicle Methods for OCR: - text implies it categorizes plates by issuing agency (e.g., US state, Canadian province, presumably to aid in recognition - no detail on method for individual character recognition - has back end interface to national databases for verification; may be used to "correct" recognition of plates Sample plates read / accuracy: http://www.singaporegateway.com/optasia/plates.htm Article claims 90% accuracy for Canadian, Mexican, and US plates. The other online references I found would either describe methods or accuracy but not both in the same context. Another reference I happened to find while searching that you may find interesting. http://users.erols.com/lnelson/lpir.html Shows a series of images captured in natural light and comparing to use of near IR lighting and cameras (scroll down to section starting with "Optimizing License Plate Design..." --Maniac

dustymack-ga rated this answer:
Excellent research; but does not include originally requested data.

Comments Log in to add a comment

Subject: Re: OCR for License Plates
From: jerome0001-ga on 05 Jan 2005 14:40 PST

don't know that much about it, but here in ontario, canada hwy 407 uses OCR for billing. http://www.407etr.com Maybe you can contact some one there for more info.

Subject: Re: OCR for License Plates
From: johnfrommelbourne-ga on 05 Jan 2005 22:52 PST

Citylink in Melbourne Australia have just started a service whereby they now advertise that their latest technology will mean that object we stick on inner windscreen of car that is read by scanning technology when we go through one or more sections of tollroad is now no longer required( although still preferred) as latest technology can simply read licence plate( here, registration plate, with great accuracy and record the $1.50 (or whatever fee was) against owner of that registration plate. Is that of any interestor relevance?? If so just advise John From Melbourne

Subject: Re: OCR for License Plates
From: javisv-ga on 08 Jan 2005 02:44 PST

F

Important Disclaimer: Answers and comments provided on Google Answers are general information, and are not intended to substitute for informed professional medical, psychiatric, psychological, tax, legal, investment, accounting, or other professional advice. Google does not endorse, and expressly disclaims liability for any product, manufacturer, distributor, service or service provider mentioned or any opinion expressed in answers or comments. Please read carefully the Google Answers Terms of Service.

If you feel that you have found inappropriate content, please let us know by emailing us at answers-editors@google.com with the question ID listed above. Thank you.

The I-CUBE Web site is packed with information on our product lines. We invite you to explore the site and download the technical documentation, news items, photos, description of sample installations, system simulations and recognition demos.

License Plate Recognition

License Plate Recognition for a wide range of applications including Parking, Access Control, Logging all vehicles & alarm when Wanted Vehicles detected.