GB2638438A

GB2638438A - An automatic number plate recognition system and a method of operating an automatic number plate recognition system

Info

Publication number: GB2638438A
Application number: GB2402516.5A
Authority: GB
Inventors: Mendel Sudak Menachem
Original assignee: Individual
Current assignee: Individual
Priority date: 2024-02-22
Filing date: 2024-02-22
Publication date: 2025-08-27
Also published as: GB202402516D0

Abstract

An Automatic Number Plate Recognition (ANPR) processing system receives, from an image capture device (e.g. camera 22,Fig.4), image or video data of a monitored region (e.g. parking space 39,Fig.4); processes it to identify a region of interest therein and produces a data file comprising image data 68 (e.g. an excerpt) regarding a vehicle in the region of interest. The data file may be transmitted to a remote server (44,Fig.4) for processing. The camera may be integrated into a bollard (33,Fig.4), fence post or brick and powered by a rechargeable battery or microgenerator only in response to an activation switch. The switch may be arranged in the ground (45,Fig.4) and trigger the camera if a force exceeding 300kg is applied to it, e.g. by a vehicle entering or leaving the region of interest.

Description

An Automatic Number Plate Recognition System and A Method of Operating an Automatic Number Plate Recognition System The present invention relates to an automatic number plate recognition (ANPR) system and to method of operating an ANPR system.

ANPR systems are well known, having been available and in use for many (>20) years. They are used in many fields including commerce, security, law enforcement and more. Their use in the management of car parks is a recent development and has been known to the present applicant since 2021.

An ANPR system typically, though not essentially, consists of two parts.

The first is an image capture device such as a camera that is arranged to capture images of and record number plates of vehicles passing the camera's viewpoint regardless of lighting conditions. One way in which such systems operate is by using infrared lighting with a filter provided over the lens that blocks the IR light. In this way, it is only the reflection of the number plate that is shown rather than the entire image (of the vehicle and the number plate).

The characters of the number plate are not reflective of the IR light. Accordingly, the characters of the number plate appear black against a white image representing the rest of the number plate. This provides a clear and easily identifiable image of the characters of the number plate such that a vehicle can be reliably identified. Little other information is provided by such a system.

ANPR systems are known including cameras with two lenses, one of which has a filter (as described above) whilst the other does not and therefore provides a useful overview image.

The second part of an ANPR system is a processor or computing function that includes optical character recognition (OCR) software that functions to look for a rectangle or square with images of characters in it, and then translates those images into editable or processable characters.

Some ANPR cameras are 'smart' and have the analytics software built in while others are arranged to provide the captured information to a connected processing system that could be remote. In other words, the camera captures an image in one of the ways described above and then in some way passes the captured image on to a remote system or server that is arranged to perform the analysis. The analysis is typically performed on another system such as a computer on site or, in some cases, streamed to a server remote from the image capture device which contains stored software required to execute the processing. The remote server could be cloud based.

Conversely, some cameras do not use filters or IR imaging, but simply capture images of sufficient quality for provision to the analysis software to carry out its image processing task of identifying the registration number or details of the vehicle in question.

The use of ANPR cameras in car park management can encourage parking abuse rather than deter it because it effectively provides a duration of free parking in a car park unlike having to pay for a short while on street.

In 2015 the present inventor invented the system described in GB2536470. This system includes an ANPR camera that is used on the entrance/exit of a vehicle to a car park to log the registration plate on arrival and/or departure. Further cameras are provided that track where vehicles park and where the occupants of the vehicles go once they have parked. This can be useful in determining whether the vehicle driver is entitled, say, to free parking in the space in question.

The data processing volumes are significant and so it can be difficult to capture sufficient data and to process it correctly. Indeed, although the system described in GB2536470 works well, the processing is presently performed on a central server that has a high end graphics processor. A number of cameras are provided and the images they capture analysed to enable as much information as possible to be determined about a user's movements after the vehicle has been parked. The amount of data that can be processed centrally is dependent on the bandwidth and capacity of the server and volume of movement/detail being analysed. The server is typically a larger piece of equipment which makes some local installations i.e. at generally the same location as the car park in question, cumbersome if not technically unfeasible. The limits on capacity render large scale deployment impractical.

As mentioned above, GB2536470 discloses a system in which ANPR is used in combination with tracking of a vehicle's user.

GB2608799 discloses a vehicle charging station comprises a processor which is programmed to communicate with a vehicle to obtain a unique vehicle identifier code.

The charging station may comprise an ANPR camera.

GB2598257 discloses a bollard for holding a camera within its interior at one of multiple positions and/or a camera mount for so mounting a camera. The camera mount may comprise a U-shaped holder for holding the camera and flanges forming a 'top-hat shape. The camera may be (or may be part of) an ANPR system.

US7530761 discloses a system and method for electrical power generation utilizing vehicle traffic on roadways. A disclosed power generation system comprises one or more standard hydraulic cylinders, one or more electro-mechanical generators, and one or more hydraulic accumulators coupled to a hydraulic motor.

According to a first aspect of the present invention, there is provided a processing system for use with an automatic number plate recognition (ANPR) system, the processing system comprising an image processing device arranged to: receive video data from an image capture device the video data including video of a monitoring region; process the video data to identify regions of interest within the monitoring region; and produce a data file in dependence on the processed video data, the data file comprising image data regarding a vehicle in the region of interest.

In an embodiment, the production of the data file comprises determining from captured video regions or frames of interest and providing this as an excerpt from the video data. The system also operates receiving still image as an input, i.e. instead of video data. In a corresponding manner the system functions to process the image data to identify regions of interest within the monitoring region; and produce a data file in dependence on the processed image data, the data file comprising data regarding a vehicle in the region of interest.

In an embodiment, the system is arranged to capture an image of a vehicle in a region of interest and movement of a user associated with the vehicle.

In an embodiment, the processing system is arranged to store the produced data In an embodiment, the processing system comprises a transmitter to transmit the produced data file to a remote server for processing.

Thus, a method and system is provided in which the analytics load is reduced which therefor enables the processing to be performed locally. Dedicated hardware is provided to perform the functionality.

Furthermore, in embodiments, detailed analysis is only preformed when relevant movement is detected. Rather than analyse the high resolution images captured by the camera, the field of view is divided into sections or regions. Only sections or regions that have a change in pixel groups are sent for analysis which may be on a unit attached to the camera or at a central server location. Typically, a region may correspond to a whole car park space or it may be a region of a car park space where the registration plate of a (parked) vehicle is usually located.

Speciality hardware has been developed and is provided including the components required for the specific processing at hand. Embedded computers already exist but a device that has a capture card, graphics processor, external memory modules, comms and additional processing, in a compact unit, is not currently available.

According to a second aspect of the present invention, there is provided a automatic number plate recognition system comprising: a video camera arranged to capture video of a vehicle in a monitoring region; a processing system according to the first aspect of the present invention. file.

According to a third aspect of the present invention, there is provided a method of operating an automatic number plate recognition system, the method comprising: receiving video data from an image capture device the video data including video of a monitoring region; processing the video data to identify regions of interest within the monitoring region; and producing a data file in dependence on the processed video data, the data file comprising image data regarding a vehicle in the region of interest.

According to a further aspect of the present invention, there is provided processing system for use with an automatic number plate recognition (ANPR) system, the processing system comprising an image processing device arranged to: receive image data from an image capture device the image data including an image of a monitoring region; process the image data to identify one or more regions of interest within the monitoring region; and produce a data file in dependence on the processed image data, the data file comprising image data regarding a vehicle in the region of interest.

The concept of video analytics is well known and is an ever developing technology, particularly within the Al world. However, present systems and technologies require extensive hardware resources resulting in the systems being largely unfeasible or impractical in most applications. A method is therefore required to make efficient use of hardware resources so that systems can be deployed more efficiently.

Embodiments of the present invention will now be described in detail with reference to the accompanying drawings, in which: Figure 1 is a schematic representation of a known automatic number plate recognition system; Figure 2 is a schematic representation of an example of an automatic number plate recognition system; Figure 3A is a more detailed schematic representation of the ANPR system of figure 2; Figure 3B is a schematic view of a cark park; Figure 4 is a schematic representation of another example of an automatic number plate recognition system; and Figure 5 is a schematic flow diagram showing an exemplary method of data or image processing.

Figure 1 is a schematic representation of a known automatic number plate recognition system. It is similar the system shown and described in the present inventor's previous application GB2536470.

Referring now to Figure 1, there is shown a car park 2 associated with a destination (not shown), such as a supermarket. The car park 2 includes a parking area with and a plurality of parking spaces 4. The car park 2 also incorporates a monitoring system in the form of a monitoring controller 6 and image capture means 8 in the form of a camera 8; arranged and fixed to capture images of the car park and vehicles that might enter or leave it.

In this embodiment, the camera 8 includes some means 10 of connection to the monitoring controller 6. The camera 8 may be operable to capture image data both in the visible spectrum and in the infrared spectrum.

In this example, the supermarket (not shown) owns and operates the car park 2; which is intended for the exclusive use of customers of the supermarket. The supermarket will typically include an entrance through which users enter and so the movement of users from their vehicles to the entrance can be determined and captured by one or more cameras 8 such as that shown in the fioure.

In this example, only customers of the supermarket are authorised to use the car park 2 and such customers are limited to, for example; hvo hours of parking time.

Traditionally, monitoring of car parks 2 to ensure compliance with rules has been carried out by one or more members of staff or other operatives, hereinafter referred to as parking attendants. The purpose of the monitoring system shown in Figure 1 is to minimise the requirement for such parking attendants. The monitoring data may also be used in customer loyalty schemes, for example where vehicle identification data is associated with a profile of such a loyalty scheme.

The monitoring controller 6, typically includes a processing means in the form of a processor, memory means in the form of a memory for storing data, ANPR means in the form of an ANPR module for extracting vehicle registration information from image data, a tracking module for tracking the movement of vehicles and occupants of vehicles and a transceiver, such as a wireless transceiver with an antenna.

It will be appreciated by those skilled in the art that any number of devices may be used to provide the functionality of the monitoring controller 6, such as a personal computer, laptop, tablet or smart phone. In the example shown, communication between the camera B and the monitoring controller 6 is via a network such as the Internet or a LAN 12.

As explained above, the system shown in and described with reference to Figure 1, works well. However, there is a desire to simplify the data processing and further to enable data in respect of a larger number of parking spaces or regions to be processed and hence in respect of a larger number of vehicles and users of the system to be tracked.

Referring to Figure 2, an ANPR system 14 is shown, in the form of a bollard 16.

The bollard 16 has a housing 18 and a base 20 for coupling the bollard to a floor surface, such as that of the car park itself.

An image capture device 22 is provided arranged within the housing. Coupled to the image capture device or camera 22 is an image processing device 24 for receiving video data from the image capture device. The image processing device 24 is provided as a dedicated hardware device for performing the required functionality, as will be described in greater detail below.

A processing devoice or system is provided for use with an automatic number plate recognition (ANPR) system. Typically, and as will be explained in greater detail below, the processing system includes an image processing device arranged to receive video data from an image capture device the video data including video of a monitoring region. The system is also arranged and configured to process the video data to identify regions of interest within the monitoring region and to produce a data file in dependence on the processed video data, the data file comprising image data regarding a vehicle in the region of interest.

In other words, first of all the captured video is processed to identify within it regions of interest. Depending on the size of the monitoring region, these regions of interest could correspond to an entire car parking space or to just the parts of a car parking space that are expected to include a car registration number or plate. These are then selected and further processed to produce a data file which could also be an image file, which would include image data regarding a vehicle in the region of interest. In other words the processing done locally identifies regions of an image or a frame of video that includes features of interest and cuts these to produce a data file for onward transmission.

The image processing device 24 is small, compact and efficient and arranged to deliver video analysis without the cumbersome extras associated with standard computing devices. Typically, it includes a microprocessor and a data filter. The processing of the data will be described in further detail below. However, an important feature of the image processing device 24, is that it is able, to perform a number of functions locally to the image capture device 22.

Specifically, the image processing device 24 is arranged locally to the image capture device 22, to process the captured video data, identify regions of interest within frames of the video data, and to produce a data file in dependence on the processed data. The produced data file can then be forwarded to a central server or to a different processor for the performance of ANPR functionality. The data file can be an image of a part of a vehicle including the registration details or it could be a text file indicating the registration details determined based on the captured video or an image file produced.

The image processing device 24 may be provided as a dedicated independent device, i.e., with its own separate housing or it may be integrated into or formed as part of the camera device 22 or bollard structure 18 itself. Thus, the local processing of the data and the production of a data file which can easily be transmitted to a central server means that the ANPR system is easily scalable since a dedicated device 24 can be provided with each image capture device or camera 22 provided within or associated with a particular car parking space.

In conventional systems, it has been known to stream video data from a video capture device to a central server which then performs known functions of ANPR. By contrast, in the present system, analysed data in the form of a data file is transmitted. This typically will mean that the amount of data that needs to be transmitted is significantly smaller than the video files that would have been transmitted in previous known systems.

The current applicant has recognised that a bottleneck in existing systems is the bandwidth available for transmission of the video data from the camera of an ANPR system (e.g. the camera 8 in the system of Fig. 1) to the monitoring controller 6 (again of the system of Fig. 1). This has now been addressed since the bandwidth required for transmission of a processed data file is significantly less than that required for the transmission of raw video data.

Figure 3A shows a schematic representation of the camera 26 arranged coupled, via a wired link 28 to an image processing device 24. The image processing device 24 is arranged to provide as an output 26 a data file which can be transmitted either wirelessly or via wired connectivity to a central server (not shown in Figure 3A). The image processing device 24 comprises an input unit 30arranged to receive raw video data from the camera 10.

A graphics processor 34 is provided arranged to receive from the input unit 30, the received video data. A memory and communication module 32 is provided to receive the processed data from a graphics processor 34 and to communicate or forward the data onwards for full ANPR processing.

The system 26 is arranged such that the graphics processor 34 is configured to determine when movement is detected to have occurred. Furthermore, it is arranged to break down the high resolution images received from the camera 26 into sections. Only sections that are determined to have changed are processed or sent for analysis.

Referring to Figure 3B, an example is shown of a schematic view of a car park, divided into regions 36. The regions 36 in this example correspond to car parking spaces within the car park.

As can be seen, a number of the regions 36 have no vehicles parked in them and accordingly during processing by the graphics processor 34, data associated with the unoccupied car parking spaces is not processed. It is not included in any generated data file and not forwarded to an ANPR system.

Over time only sections that have a change in pixel groups are sent for analysis as full ANPR. Detection can be made at the input unit 30 (which typically includes processing and memory functions) of which of the parking spaces 36 have experienced change. The input unit 30 can, in a preferred example, store at any point in time a last-captured image of the car park (this can be taken as a frame from the received video) and can then compare this to a subsequently captured image or frame from a video to determine if a change has occurred in any of the parking spaces 36. A simple and robust way is thus provided for enabling selection of the parts of an image for forwarding as a digital file for full ANPR processing.

An easily scalable system is provided for ANPR together with the tracking of vehicle users in a way that has not previously been possible.

Referring again to Figure 3A, a schematic representation of the image capture device 26 is shown. The image capture device or camera 26 typically comprises a lens 38. It further includes a processor capable of capturing an image via the lens 38. Optionally, memory 40 is provided arranged for storing captured images before being communicated via the connection 28 to the input unit 30 on the image processing device 24.

Figure 4 shows an example of an ANPR assembly including plural systems of the type described above with reference to Figures 2, 3A and 3B.

Each system in this example includes an image capture device, which as well as functioning in the manner described above with reference to figures 2, 3A and 3B, also has a trigger which may be mechanical or heat triggered and therefore be capable of being activated by a vehicle, such as a car passing over a cable or a heat sensor. Accordingly, the image capture device, such as a camera, utilizes less power and therefore renders power by a battery pack as a feasible technical solution.

Referring to Figure 4, processing devices 24 are associated with each of a number of bollards or posts 33, 35 and 37. An image capture device in the form of a digital camera is fixedly or removably mounted within each of the bollards and is arranged to capture images of the associated parking spaces 39, 41 and 43, respectively.

Each of the bollards is connected either via a wired connection or wirelessly to a region 45, 47 and 49 of the ground surface. This region could be a speed bump or simply a strip of the ground surface at the entrance and exit of the parking space in question.

The regions 45, 47 and 49 have arranged on them a pressure activated switch (not shown), arranged to trigger the associated digital camera. The system is arranged such that when the speed bump or region is driven over by a vehicle, such as a car, the pressure activated switch causes a signal to be provided to the associated camera. The signal could be an on/off signal and/or it could preferably include power which, is arranged to activate the camera and cause it to capture an image of the region in front of it, i.e., or where its image capture region happens to be.

Thus, it will be appreciated that the pressure activated switch could be a switch that causes a power source within the camera to be connected to the camera's circuitry, thus enabling it to take a picture. Alternatively, it is also possible that no permanent power source is provided within the camera and instead the pressure activated switch includes power generation functionality. In this case, in response to pressure being applied (by the vehicle driving over the region and activating the switch) sufficient power is generated to activate the camera and cause it to take a picture, and subsequently to store it or to transmit it to wherever it might be needed for analysis.

In other words, in one nonlimiting example the pressure activated switch is a microgenerator positioned such that when a vehicle enters or leaves the designated region, the vehicle engages with the microgenerator to cause it to generate power for the image capture device. The power is communicated to the image capture device typically by a wired connection which causes the image capture device to "wake up" and capture the image of the vehicle entering the space. The power generated is preferably sufficient to also cause the image capture device to transmit the captured image to a connected server for processing in known ways.

The pressure activated switch provided on the speed bump is activated when a vehicle, such as a car, drives over it. In an alternative example, the switch is provided in the form of a heat sensor which is triggered to provide an on/off signal and/or to the camera when a vehicle is in close proximity to it or engages it by driving over it.

In addition, the system is preferably arranged such that when the vehicle leaves, again, the pressure activated regions are again triggered which in turn will provide an activation signal (as described above) to the camera in question. This causes an image to be captured. Thus, there are two opportunities to capture an image of the vehicle; when the vehicle enters the car park space and also when it leaves. This enables a determination of the length of stay of a car in the space to be made which can be useful information for a car park operator, not least to enable a determination of charges to be made.

However, it will be appreciated that at all other times, and irrespective of how long the vehicle is parked, when the vehicle is merely in position in space 39, 41 and 43 and not interacting with the respective activation region 45, 47 and 49, no power will be provided or drawn by the camera in question.

In other words, the ANPR system can be entirely unaffected by the presence of the vehicle in the spaces 39, 41 and 43. The camera in question for a respective space is only activated when required to capture an image of the vehicle either leaving or arriving within the space 39, 41 and 43. This means that the power drawn from the power source in the actual camera is minimal. For most time that passes, no power at all is needed. This means that a small replaceable power source such as a battery, e.g. AA, AAA etc or fuel cell, can be used and can be expected to have a long active life whilst still enabling efficient and correct functioning of the ANPR system.

It will thus be appreciated that the system of Figure 4 is analogous to that of Figures 3A and 3B, except that in Figures 3A and 3B a single camera is used to capture images of multiple car parking spaces, whereas in Figure 4 each of the parking spaces 39, 41 and 43 is provided with its own dedicated camera and processing device 24.

Figure 5 is simplified schematic flow diagram showing steps of an exemplary method.

Initially at step 60 a video is captured of a monitoring region. The monitoring region could be a single cark parking space or a region of a cark park that includes multiple car parking spaces.

At step 62, the video is processed to identify regions of interest, this can be done using known image processing tools and be based for example on character recognition, text extraction or other such known tools. As explained above the method is equally effective and applicable if still images are captured by the image capture device. The benefit will be that instead of transmitting large field of view images which will be of larger file size, a smaller text or extracted image can be transmitted.

From this, at step 64, a file is generated. This can be an image file 68, say, including images of a vehicle registration (front and/or rear) or it could be a simpler and smaller text file 70 including a character recognised registration number or identification. The data file, at step 66 is then forwarded onwards for processing by an ANPR system to identify the vehicle in question. It will be appreciated that instead of having to forward the larger captured video file, by the performance locally of some image or video processing a smaller data file is created for onwards transmission.

It will be appreciated that in the example described above with reference to Figures 3A and 3B the regions of interest 36 identified for including in a data file included areas within a car park that included a whole car or vehicle. In contrast, in the example described above with reference to Figure 5 smaller regions of interest were identified, i.e. only those parts of an image taken from a video that included the actual registration number or plate. The principle at hand is the same which is that region of interest are identified within a larger image and only these are used to form a data file text or image for use in subsequent ANPR processing.

Embodiments of the present invention have been described with particular reference to the examples illustrated. However, it will be appreciated that variations and modifications may be made to the examples described within the scope of the present invention.

Claims

Claims 1. A processing system for use with an automatic number plate recognition (ANPR) system, the processing system comprising an image processing device arranged to: receive video data from an image capture device the video data including video of a monitoring region; process the video data to identify regions of interest within the monitoring region; and produce a data file in dependence on the processed video data, the data file comprising image data regarding a vehicle in the region of interest.
2. A processing system according to claim 1, in which the production of the data file comprises determining from captured video regions or frames of interest and providing this as an excerpt from the video data.
3. A processing system according to claim 1 or 2, in which the system is arranged to capture an image of a vehicle in a region of interest and movement of a user associated with the vehicle.
4. A processing system according to claim 1, arranged to store the produced data file.
5. A processing system according to claim 1, comprising a transmitter to transmit the produced data file to a remote server for processing.
6. A processing system according to any of claims 1 to 5, comprising an activation switch, responsive to an input to trigger the image capture device automatically to connect the image capture device to power when a vehicle enters or leaves the region of interest.
7. A system according to claim 6, comprising a power source, for powering the image capture device, the power source being disconnected from the image capture device in normal use unless activated by a vehicle.
8. A system according to claim 7, in which the power source is a rechargeable power source such as a rechargeable battery provided within the image capture device.
9. A system according to claim 7, in which the power source is a microgenerator responsive to the activation switch to generate power for provision to the image capture device.
10. A system according to claim 9, in which the microgenerator is provided as a pressure activated generator positioned such that when a vehicle enters or leaves the designated region it engages with the microgenerator to cause it to generate power for the image capture device.
11. A system according to any of claims 6 to 10, in which in a disconnected state there is no circuit for power loss from the power source to the image capture device.
12 A system according to any of claims 6 to 11, in which the activation switch is a pressure activated switch.
13. A system according to any of claims 6 to 12, in which the pressure activated switch is arranged to be activated by movement of a vehicle, the image of which is to be captured.
14. A system according to claim 13, in which the switch is arranged on a region of ground surface in the vicinity of the image capture device, the switch being arranged to activate the image capture device in response to movement of the vehicle over the region of ground surface.
15. A system according to claim 13 or 14, in which the switch is activated only if a threshold force is applied to the mechanical switch, the threshold force being greater 30 than 300KG.
16. A system according to any of claims 'I to 15, in which the image capture device is integrated into a bollard, a fence post, a brick or other feature.
17. A processing system for use with an automatic number plate recognition (ANPR) system, the processing system comprising an image processing device arranged to: receive image data from an image capture device the image data including an image of a monitoring region; process the image data to identify one or more regions of interest within the monitoring region; and produce a data file in dependence on the processed image data, the data file comprising image data regarding a vehicle in the region of interest.
18. An automatic number plate recognition system comprising: a video camera arranged to capture video of a vehicle in a monitoring region; a processing system according to any of claims 1 to 17.
19. A method of operating an automatic number plate recognition system, the method comprising: receiving video or image data from an image capture device the video or image data including video or an image of a monitoring region; processing the video or image data to identify regions of interest within the monitoring region; and producing a data file in dependence on the processed image or video data, the data file comprising image data regarding a vehicle in the region of interest.