Systems and methods for the digitisation of motor racing circuits and the consumer-controlled viewing of motorsports events

Abstract

A method 300 of creating a digital terrain mode! (DTM) of a racetrack, comprises: obtaining 304 digital boundary mode! (DEM) of the racetrack, wherein the DBM comprises a plurality of reference point

Description

FIELD OF THE INVENTION

[0001] The present disclosure is concerned with systems and methods for the creation of digital representations, such as a Digital Terrain Model (DTM), of motor racing circuits. More particularly, though not exclusively, the present disclosure is directed toward improvements in or relating to systems and methods for creating such digital representations which enable viewers (or other interested parties) of a particular motorsports event to manipulate the digital representation of the event in accordance with their own requirements or preferences. The present disclosure also extends to the creation of a three-dimensional digital surface model (3D-DSM) which encompasses the motor racing circuit (racetrack) and its surroundings which are a collection of non-uniform surfaces. It is to be appreciated that whilst the present disclosure is primarily concerned with digital representations on a motor racing circuit, it is also envisaged that the disclosures may be equally applicable to the creation of digital representations of driving surfaces or any physical features for a multitude of purposes, BACKGROUND OF THE INVENTION

[0002] The field of motorsports attracts millions of viewers on a regular basis. Typically, motorsports events are broadcast via either traditional broadcasting methods (e.g. live or previously recorded television) or are streamed through alternative channels such as on internet viewing platforms. However, in both cases, the images which are broadcast to viewers are subject to available views which are recorded by cameras positioned around the track of the motorsports event and in pit lanes of the event in question. Such cameras may be either static or dynamic. Whilst many cameras can be positioned around the track in order to provide viewers with a plurality of alternative views (which in some cases may be specifically selected by a viewer), the necessarily finite nature of the number of available cameras means that some aspects of the event in progress will not be captured. These missed aspects may be of interest to some viewers and as such, the resulting broadcast experience for that viewer may be suboptimal.

[0003] For example, one particular viewer may have a keen interest in one participant in the motorsports event and may simply wish to view that participant for the entirety of the event. However, since the broadcast experience is not tailored for the interests of one particular viewer, it will typically not be possible for the finite number of cameras to only capture the progression of that participant during the race (as they will also need to capture the progress of other participants). As such, the viewer may not be able to obtain the viewing experience that they wish to have when viewing the event. Furthermore, even were the progression of the particular participant to be persistently captured, the available view(s) of that participant will be limited to the images captured by the available cameras positioned at the predetermined camera locations and may not be able to be offered the view desired by the viewer,

[0004] The limited available views offered by existing systems may also be detrimental in other aspects of motorsports. For example, a race controller, marshal or adjudicator at a motor racing event will often need to have accurate and instant replays of events within the race, such as a vehicle possibly transgressing a driving boundary limit of the race track. The available views offered by existing systems may not be sufficient to determine conclusively whether such a boundary limit has been crossed. In some instances, such an incident may be entirely unavailable to view which prevents a required decision from being made in the required manner. Further, it can be advantageous for a racing team of drivers and engineers to have access to a plurality of views of a racing vehicle as it travels around the racetrack in order to train the driver or to improve the vehicle’s performance. Such views may not be available in existing systems.

[0005] In some systems, virtual models may be available which enable views of the racetrack to be manipulated in a virtual space. However, such models are unable to provide both the feature recognition and the high accuracy required to create virtual models containing ail the information necessary for high-fidelity viewing of motorsports events. In particular, these models are unable to provide the level of detail required for such a model to effectively and satisfactorily replace the live images captured by cameras placed around a racetrack. They are certainly unable to effectively replace these views when the view includes a vehicle travelling at a high speed.

[0006] It is an object of the present invention to overcome one or more of the problems described above. SUMMARY OF THE INVENTION

[0007] According to a first aspect of the present embodiments, there is provided a method of creating a digital terrain model (DTM) of a racetrack, the method comprising: obtaining digital boundary model (DBM) of the racetrack, wherein the DBM comprises a plurality of reference points used in the creation of the DBM, the plurality of reference points including locations of a master control station (MCS) and a set of secondary control stations (SCS), and each reference point being determined with respect to the locations of other reference points and the boundary of the racetrack; receiving: a plurality of aerially visible markers, each marker being positioned proximate a location of a reference point of the DBM and respectively at a predetermined position from each reference point; and captured ariel images of the racetrack including the visual markers; using the predetermined positions of the visual markers from the respective reference points to specify the locations of the reference points as a plurality of ground control points (GCPs) of the DTM; and orthorectifying the captured images with the DBM of the racetrack using the plurality of reference points of the DBM as GCPs for the DTM.

[0008] The receiving step may comprise receiving captured ariel images of the racetrack including a boundary of the racetrack; and the using step may further comprise adding the boundary of the racetrack as determined by the DBM, to the plurality of GCPs used in the orthorectifying step.

[0009] In some embodiments, the obtaining step further comprises obtaining the DBM of the racetrack which includes one or more marked features of the racetrack and the receiving step further comprises receiving captured ariel images of the racetrack including the one or more marked features; and the using step comprises adding the one or more marked features as determined by the DBM, to the plurality of GCPs used in the orthorectifying step.

[0010] The one or more marked features may include a raised kerb of the race track and / or a starting grid line.

[0011] In one embodiment, the visual markers comprise visible QR codes or barcodes.

[0012] The method may further comprise receiving dynamic data tracking an object moving on the racetrack and combining the dynamic data with static data of the DTM to create a fully digitised racing circuit (DRC) model that comprises both static and dynamic data.

[0013] The dynamic data may comprise data captured from a fixed location relative to a reference point.

[0014] In some embodiments, the obtaining step comprises obtaining a DBM having an accuracy of <10mm at the boundary of the race track. [001S] The method may further comprise capturing the ariel images of the racetrack including the visual markers.

[0016] The capturing step may comprise capturing photogrammic aerial images from an ariel drone flying at an altitude equal to the width of the racetrack above one of two boundaries of the racetrack. In one embodiment, the capturing step comprises capturing photogrammic aerial images from a camera angled at 45 degrees from the racetrack surface towards the other one of the two boundaries of the race track.

[0017] In some embodiments the DBM of the racetrack comprises a plurality of discrete data points representing different adjacent points along a boundary of the racetrack and the method further comprises fitting a curve to a plurality of discrete data points to determine a continuous digital representation of the boundary of the racetrack.

[0018] Preferably, the method further comprises creating a visual representation of the 3D-DSM. This can then be directly output to a screen for imaging. Alternatively, the method may further comprise transmitting the DTM or a visual representation of the DTM to a third party server using a wide area network.

[0019] The present disclosure also extends to a method of creating a three dimensional (3D) Digital Surface Model DSM of racetrack, the method comprising creating a DTM as described above; supplying elevation images of features in areas adjacent the race track; and combining information derived from the elevation images into the DTM to create the 3D-DSM.

[0020] The method may further comprise acquiring feature information regarding peripheral objects in geographical areas adjacent the racetrack, and combining the feature information with the DTM.

[0021] The acquiring step may comprise acquiring predetermined static digital models of the peripheral features of objects. Alternatively or in addition, the acquiring step may comprise acquiring predetermined dynamic digital models of the peripheral features of objects which are modelled in movement, namely are moving.

[0022] In some embodiments the method further comprises getting feature information regarding peripheral landscape in geographical areas adjacent the racetrack, and combining the feature information with the DTM.

[0023] The obtaining step preferably comprises obtaining a DBM having an accuracy of <10mm at the boundary of the race track and the acquiring or getting step preferably comprises acquiring or getting features information having an accuracy of <50mm in the geographical areas adjacent the racetrack.

[0024] The present disclosure extends to a non-transitory, computer-readable media storing instructions that, when executed by the one or more processors, cause the one or more processors to carry out the method described above. [002S] According to another aspect of the present disclosure there is provided a system for creating a digital terrain model (DTM) of a racetrack, the system comprising: a 3D-DSM creator processor configured to: obtain digital boundary model (DBM) of the racetrack, wherein the DBM comprises a plurality of reference points used in the creation of the DBM, the plurality of reference points including locations of a master control station (MCS) and a set of secondary control stations (SCS), and each reference point being determined with respect to the locations of other reference points and the boundary of the racetrack; receive a plurality of aerially visible markers, each marker being positioned proximate a location of a reference point of the DBM and respectively at a predetermined position from each reference point; and captured ariei images of the racetrack including the visual markers; use the predetermined positions of the visuai markers from the respective reference points to specify the iocations of the reference points as a plurality of ground control points (GCPs) of the DTM; and orthorectify the captured aerial images with the DBM of the racetrack using the plurality of reference points of the DBM as GCPs for the DTM. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order that the disclosure may be more readily understood, reference will now be made, by way of example, to the accompanying drawings in which: Figure 1 is a schematic diagram showing an embodiment of a system for the creation of a static 3D-DSM (Digital Surface Model) of a motor racing track, which includes a processing system, drones and a wireless communication network; Figure 2 is a schematic block diagram showing the processing system of Figure 1 in greater detail; Figure 3 is a schematic block diagram showing the inputs and outputs of the 3D-DSM creator processor shown in Figure 2; Figure 4 is flow diagram showing a method of operation of the processing system of Figure 1; Figure 5 is a photograph of an exemplary racing track with an overlay of a digital model of the racetrack and the pole-mounted sensors of a tracking system for use with an embodiment of the present invention; Figure 6 is a graphical diagram showing overlapping fields of view of the pole-mounted sensors around the racetrack of Figure S; Figure 7 is a schematic diagram showing a pole-mounted sensor used as part of a tracking system of Figure 5; and Figure 8 is a schematic diagram showing a cloud event hub used to distribute the 3D-DSM created by the system of Figure 1. DETAILED DESCRIPTION

[0027] In existing arrangements, remote viewing of motorsports events relies exclusively on TV or video cameras placed at fixed, or occasionally moving, locations around a racing circuit, many of which require a manual operator and all of which together create huge volumes of video data, which requires a high data transmission bandwidth, to be distributed live and viewed, or recorded, categorised with metadata, and subsequently retrieved and viewed. Even for viewers who are present at the event, there will typically be a similar reliance on such technology in order to broadcast event progress on an in situ video screen. It is an objective of the current disclosure to remove the requirement for any TV or video technology or operators in the creation of live and recorded high-fidelity motorsports visual viewing streams and to reduce dramatically the amount of data involved (by at least 104, 4 orders of magnitude).

[0028] The present disclosure describes systems and methods to create fully digitised motor racing circuits (and areas proximate to the circuits) that allow viewers of a motorsports event to use digital streaming media devices such as televisions, phones, tablets, and computers connected to the internet, together with the live or recorded streaming services described herein, to display high-fidelity video streams of the event from viewpoints which they can define and control in real time. The high- fidelity video streams are created without the use of any TV or video cameras. The viewing may be of a live event or of one previously recorded by the system. The viewers may be spectators at the motorsport event or may be remote from the location of the event. The viewpoints that can be defined and controlled are unlimited in terms of the position, pose, field of view, magnification, etc. of a virtual camera which may be static or dynamic, the dynamic viewpoint controlled by the viewer or selected from predefined options such as ‘follow a specific car’ or ‘from a virtual drone above the leading car’, etc. There may be multiple viewpoints from multiple virtual cameras displayed on a split screen. The video streams may be supplemented by audio feeds and may include augmentation features such as advertising, technical information, social media feeds, etc. This disclosure is part of a broader series of patent and patent applications (see WO2Q21 / 214496A, WO2022 / 003343A which are incorporated herein by reference and co-pending applications set out in Annexes 1 and 2) that describe the creation of fully digital racetracks and transport systems capable of producing real-time data streams for use by multiple different users and across a plurality of different geographic locations.

[0029] It is to be appreciated that the fully digitised motor racing circuits and surrounding areas created by the systems and methods described herein are of sufficiently high fidelity that they may be used to in effect replace content captured by trackside camera infrastructure. In some embodiments, real time tracking data may also be utilised by the systems and methods to generate real time representations of vehicles on the racing circuit, in effect enabling the generated digitisation of a motor race in progress in real time.

[0030] The systems and methods disclosed herein may also utilise aspects of other co-pending applications of the applicant in order to enhance the created digitised motor racing circuits for the purposes described herein.

[0031] The disclosure provided in Annex 1 describes the creation of highly accurate digital boundary models (DBMs) of boundary limits on driving surfaces such as roadways, car parks and racetracks. The present disclosure, in accordance with some aspects, describes how to integrate those for racetracks with less accurate but more easily established digital pictorial images and point cloud maps of the driving surfaces and their proximate and distant surroundings so that optimally harmonised, static, three-dimensional digital surface models (3D-DSMs) are created covering the significant geographical areas involved in motor racing and transport systems (which can include both the driving surface of the circuit as well as proximate surrounding areas). The integrated and harmonised 3D-DSMs have many useful applications, examples of which will be given here for motorsports. Transport system and smart city applications are described in other disclosures. In particular, the 3D-DSMs can be integrated with the high-accuracy dynamic vehicle tracking systems described in WO2021 / 214496A, and with established systems, methods and technologies for computer generated imagery as used in film, virtual reality, personal computing, gaming and simulators, to provide the high-quality, consumer-controlled viewing experiences of motor racing events as described herein.

[0032] It is to be appreciated that commonly accepted terminology defines a digital terrain model (DTM) as representing bare earth topography, void of surface features. More specifically a DTM is a 3D topographic model of the bare Earth that can be manipulated by computer programs. It approximates a part or the whole of the continuous terrain surface by a set of discrete points with unique height values over 2D points. The data files contain the elevation data of the terrain in a digital format which relates to a rectangular grid. The available data is in the form of an irregular triangular mesh. A digital surface model (DSM) represents the top surfaces of features elevated above the earth, such as trees, vegetation and buildings. A 3D-DSM is a DSM (two-dimensional surface model) extended by the addition of an extra dimension (orthogonal to the two dimensions of the DSM) and extending over non-continuous elements. For example, if the DSM provides longitude and latitude coordinates, the 3D-DSM can add an extra dimension and provide altitude information of non-continuous surfaces. A 3D-DSM can provide facade representations of vertical or steep features that are not represented accurately in a DSM.

[0033] In order to enable the functionality described herein, and in particular to generate a digitisation of a motor racing circuit and, in some embodiments, representations of vehicles racing on such a circuit in real time, which are, in effect, able to satisfactorily replace images captured by a camera and presented to viewers, the generated digitisations need to be of sufficiently high fidelity and accuracy. If the accuracy and fidelity is not sufficient, then the digital representation of the motor circuit and / or race in progress may not be able to be reproduced in such a manner so as to effectively replace the provision of camera obtained imagery.

[0034] There are established methods for creating DTMs of roadways and vehicle racetracks and 3D-DSMs of surrounding features covering large geographical areas, using for example drone mounted Light Detection &Ranging (LIDAR) scanning equipment, or drone mounted photogrammetry equipment, or satellite imaging. However, none of these techniques provide both the feature recognition and the high accuracy required to create DTMs and 3D-DSMs containing all the information necessary for high-fidelity viewing of motorsports events.

[0035] Reference [1] examines the accuracy achievable by airborne LIDAR scanning (fixed wing, long duration flying at 500-600m altitude over large areas) and demonstrates that the planimetric (i.e. x-y or latitude-longitude) limit of achievement is around 50cm RMS (root mean squared). Reference [2] demonstrates that elevation accuracies of better than 10cm, which should be indicative of achievable planimetric accuracies, can be attained by drone mounted LIDAR scanning and by drone mounted photogrammetry, i.e. rotary wing, short duration, flying at 50 to60 metres altitude over relatively small areas. Reference [3] also concludes that the elevation and planimetric accuracies achievable for UAV photogrammetric surveying employing 'structure from motion' techniques is around 5 toWcm using low-level flying regimes, which become increasingly uneconomic over large geographical areas. The relative and absolute inaccuracies of all these airborne survey methods are those that remain after enhancement by orthorectifying the imagery to match a network of accurately surveyed ground control points (GCPs) spread over the survey sites, with an optimal density of 30 / km2 for vertical accuracy and 6 / km2 for horizontal accuracy (Ref [4]). The absolute accuracies have in general been assessed using either highly accurate GPS systems (e.g. differential GPS, GPS+RTK...) with 1 to3 cm accuracies or (in some cases and) Total Station surveying equipment which can give relative positional accuracies between validation checkpoints of <10mm RMS (see Ref [1]) and with great care and ideal conditions, as good as 1 to2mm.

[0036] The method of some embodiments the present disclosure described herein in a first aspect, builds directly on the above knowledge for the creation of highly accurate DBMs of the racing track surface described in the disclosure of Annex 1. That method, when applied to the two track-limit white lines on each edge of a racetrack, creates a pair of closed polylines representing the outer edges of the two white lines to an accuracy <10mm. It also establishes a ring of control stations around the racing track comprising one master control station (MCS) and a set of secondary control stations (SCSs), each up to approximately 80 metres from its nearest neighbours and spread across both sides of the track where possible (Figure 1). The 3D position of each SCS is known relative to the MCS datum position to high accuracy (<10mm compared to ground truth) and the geolocation of the MCS is known to high accuracy. The method of the disclosure of Annex 1 also models other track-marked features such as starting grid position lines, the start-finish line and pit-lane entry / exit lines, to comparable accuracies.

[0037] In embodiments of the present disclosure, systems and methods take advantage of the highly accurate 3D relative positioning of each of the control stations and the geolocation of the MCS in order to enhance the accuracy of the DTM. In particular, each of the control stations may be established as a ground control point (GCP) used in orthorectification when creating the DTM. This may be achieved by fixing a unique aerially-visible marker, such as a simple black and white barcode or QR code, temporarily but securely on the ground next to each control station and capture a photographic image of each, so that the data recording at the position of each control station can be associated with the unique marker in aerially-captured photogrammetric images. Whilst barcodes and QR codes have been mentioned, it is to be appreciated that any suitable form of marker may be used which enables the required functionality of being able to uniquely associate the data recording at the position of each control station with the marker in photogrammetric images.

[0038] This then forms a network of GCPs spread over the survey site (i.e. an area to be surveyed including the race track and surrounding areas to be included in the created DTM), but closely following the trajectory of the motor racing circuit, with separations comparable with or better than the optimal density (Ref [4]), A drone-based aerial photogrammetric survey can then be performed in the standard manner, but with a flight plan tailored to concentrate on low-level flying over the marked GCPs (i.e. the control stations) and hence the racing track surface, thus yielding the most accurate DTM of the track surface, in both planimetric and elevation dimensions. The corners of critical track markings such as the starting grid lines and red-white stripe (raised) kerbs and any other clear features can be used as additional GCPs such that the photogrammetric images are optimised and matched precisely to the DBM in those areas using standard software packages, such as Pix4D®. Furthermore, the photogrammetric images can be further orthorectified by fitting the curves of the continuous plain white track limit lines as captured in the images to their determined DBM representations (thereby directly utilising the data from the created DBM as discussed in Annex 1). In this way, a highly accurate (<10mm in x, y, z) integrated model of the track surface and its boundary limit markings plus any imperfections in the track surface (slight humps, bumps, depressions, colour changes, etc) is achieved in the form of a fine network or mesh or point cloud of 3D surveyed points combined with high-resolution images superimposed with equally high accuracy.

[0039] Flat or sloping driving features immediately adjoining the track surface such as raised kerbs and gravel run-off areas will also be captured precisely at this stage, as will proximate non-driving surfaces such as grass verges. Features that are immediately adjoining the track surface, but which have significant vertical aspects, such as Armco crash barriers, will be present in the model in terms of the elevation of their top surfaces, but the photogrammetric images will be from above so will not capture their vertical faces. The DTM can be enhanced into a 3D-DSM by adding segments to the survey flight paths designed to orient the camera towards the vertical surfaces (namely tilt the camera from a downward facing position (0 degrees to the vertical) to an angle <90 degrees to the vertical) --creating ‘fagade’ image enhancements. A typical flight path may involve flying over one track edge at an altitude equal to the track width whilst orienting the camera at approximately 45° from vertical towards the opposite track edge.

[0040] In areas of the photogrammetric images that are increasingly further away from the track surface, the accuracy of the model may decrease since the orthorectification aspect is focussed on the track surface, but as will be seen below, this is not of significance for all the applications of the 3D-DSM in mapping a motor racing circuit / track.

[0041] In some embodiments of the present disclosure, the required level of detail and accuracy of the 3D-DSM decreases at positions which are located at increasing distances from the track. For example, in instances where the 3D-DSM is to be used in order to provide a user with full control of a virtual camera to view a live or recorded motorsport event which is occurring on the track, it may only be of critical importance that the track, areas immediately surrounding the track, and optionally any vehicles or objects on the track be represented in the 3D-DSM to a high level of detail and accuracy (since these will be the areas of interest to the user). At greater distances from the track, the accuracy and fidelity requirements of the 3D-DSM may be lower since these are generally less likely to be viewed in detail by a user and are provided primarily to enhance the overall visual aesthetic of the 3D-DSM.

[0042] As a result, more economic, high-level drone flying can be carried out to cover the totality of the area of interest within and outside the racing circuit with reducing photogrammetric accuracies (up to 50cm) compared to ground truth, including facade imaging of the vertical surfaces of buildings etc. Orthorectification of this imagery can again be performed with respect to the same set of GCPs and this has the beneficial effect of ‘stitching together’ or integrating all of the photogrammetric images into a single 3D-DSM which incorporates the DBM in a completely seamless manner. This digital model is referred to as the static element of the digitised racing circuit, a combination of a fine 3D mesh and high-fidelity images of all static surfaces.

[0043] There are a number of beneficial uses for this purely static model, particularly for racing circuit owners and operators. Firstly, it forms an accurate record at a point in time of the state of the whole property, from which subsequent inspections (visual or photogrammetric) can assess changes, particular wear and tear, for the purpose of maintenance planning. Secondly, it provides a powerful design tool for planning additions, changes and improvements to the property. The design options for such changes can be created as highly accurate prototypes in the model. In this way, the model will lead the real-world changes and when these changes are actually implemented in the real world, the prototype features of the model can be replaced using an additional localised drone survey, as described above, but of limited scope. Thirdly, the model may be used to incorporate data from other sources, such as accurate surveys of underground services, to become the master record set for the whole facility.

[0044] A further use o f the static digital circuit model in respect of this disclosure is to act as the basis for the detailed design and realisation of a vehicle tracking system embodying the claims of WO 2021 / 214496A, or an alternative accurate vehicle tracking system, to create a fully digitised racing circuit (DRC) model that comprises both static and dynamic data that can be used in many beneficial ways, including the consumer-controlled viewing of motorsports events. [004S] The present disclosure also extends to the design of the tracking system as an optional enhancement to the static element of the DRC model (as described in Annex 2), This is a second aspect of the present disclosure which can use the first aspect, but is independent of the first aspect. By way of an example, a tracking system of WO 2021 / 214496A will be described, although any accurate vehicle tracking system can be used. In the case of the tracking system of WO 2021 / 214496A, infrared sensors on masts located around the circuit are used to track vehicles fitted with two infrared (IR) markers (IR emitters). Prior to use of the system, all vehicles participating in the motorsports event are first surveyed to measure and record the precise 3D position of the IR markers in relation to the vehicle wheelbase and then scanned using standard photogrammetric or LIDAR based methods, or both, or any alternative method that creates the same output data, to create a high-fidelity 3D data model of the vehicle in the same car-body centred reference frame as the IR markers and wheelbase.

[0046] Specific embodiments are now described with reference to the appended figures.

[0047] Turning firstly to Figure 1, there is shown an arrangement of an envisaged system 100 for the creation of a static 3D-DSM in accordance with embodiments described herein. Specifically, the arrangement illustrated in Figure 1 is provided in order to enable the creation of a high-accuracy, high-precision 3D-DSM of features of a motor racing track 101 and areas proximate to the track 101 surface in accordance with the precision requirements described in embodiments herein. It is to be appreciated that whilst the system 100 is described in the context, of this purpose, the system 100 may equally be applied to other use scenarios, such as for the creation of an accurate 3D-DSM of any suitable area. Similarly, the system 100 described herein may also be used in combination with other suitable systems which are used to create 3D-DSMs of lower precision or accuracy. For instance, such systems may be used to obtain portions of a 3D-DSM which are located more remotely from a motor racing track 101 and which require less precision, and these may be combined with the high precision 3D-DSM data created by the system 100 of Figure 1 to create a 3D-DSM of a large region, with varying precision in relation to ground truth in dependence upon proximity to the motor racing track 101.

[0048] The system 100 comprises a processing system 102 which is configured to create a 3D-DSM in accordance with embodiments described herein. The processing system 102 is configured to create the 3D-DSM on the basis of data received by the processing system 102. In particular, the data received by the processing system 102 may comprise photographic images (and in some instances, photogrammetric images) of the area for which the 3D-DSM is to be created. In the case of Figure 1, this area comprises the motor racing track 101 and areas immediately surrounding the track 101. For example a racing circuit is typically 6km in length, so a circle of 2km diameter and area of 3km2. Adding an area outside the circle would give for example a total area of larger than say 4km2. Such images may be obtained by use of a drone 104 or other suitably configured airborne vehicle provided with appropriately configured imaging devices (e.g. cameras) which is able to obtain the photographic images of the motor racing track 101 for the purposes of creating the 3D-DSM (such as an alternative manned or unmanned airborne vehicle). The photographic images / image data may then be provided from the drone 104 (or other configured airborne vehicle) to the processing system 102 by way of a wireless communications network 106. The areas immediately surrounding the track 101 for which images are to be captured and received may comprise any required area for which a high precision and accuracy 3D-DSM is deemed necessary e.g. areas that typically will be more likely to be viewed by a user for the purposes described above (e.g. following the path of a particular vehicle, reviewing trackside events etc). The process of 3D-DSM creation based upon captured imagery will be well known to the skilled person and therefore will not be discussed in further detail.

[0049] In some embodiments, the image data which is received by the processing system 102 may have been previously captured and stored in a data store 132. Upon determination that the processing system 102 is to create a new 3D-DSM, the processing system 102 may be configured to retrieve the relevant images from the data store 132.

[0050] In addition, the processing system 102 is also configured to receive data relating to a created digital boundary model (DBM) of the track 101 which establishes boundary lines of the track 101 in a virtual space. Typically, these boundary lines will include the two track-limit white lines on each edge of the track 101, representing the boundaries of the driving surface of the track 101. The DBM may be created in accordance with the systems and methods described in Annex 1 In particular, a series of control stations 108 may be established around the track 101 comprising at least one master control station (MCS) 1Q8A and a set of secondary control stations (SCSs) 108B, each up to approximately 80 metres from its nearest neighbours and spread across both sides of the track where possible. The 3D position of each SCS is known relative to the MCS datum position to high accuracy (<10mm compared to ground truth) and the geolocation of the MCS is known to high accuracy. Each of the control stations 108 is configured to establish a relative position with one or more points on the boundary lines of the track 101 with high accuracy. These points may then be virtually connected to create a set of polylines which represent the outer edges of the two boundary lines in a virtual space. In accordance with the techniques described in Annex 1, the position of these boundary lines relative to the control stations (and the MCS in particular) is established to an accuracy of <10mm. The DBM therefore creates a highly precise reference point for visible features of the track 101 (i.e. the white boundary lines of the track) relative to specific geographical positions (i.e.; the control stations 108, and in particular the MCS 108A). This DBM can therefore be used in the process of orthorectifying photogrammetric images received of the track 101 by a drone 104 or other suitable airborne vehicle by providing data which can relate a virtual model of the track 101 to visible images provided by the drone 104 or other suitable airborne vehicle, in accordance with embodiments described herein. The received data will typically include the positions of the created polylines relative to each control station 108, as well as the geolocation of each control station 108. The control stations 108 will each typically be labelled with a unique identifier within the data such that the positions of the polylines relative to each control station 108 may be readily defined and the control station 108 may be readily associated with its geolocation.

[0051] In some embodiments, the processing system 102 may be configured to receive raw data from one or more of the control stations 108 located at the track, and may further be configured to generate the set of polylines of the DBM in accordance with methods described in Annex 1. In other embodiments, the raw data may be received indirectly from another source (e.g. a separate system to which the raw data is transmitted to from the control stations 108). In cases where raw data is provided, the system may comprise a curve fitting engine as is described later with reference to Figure 3. In some embodiments, the processing system 102 may be configured to simply receive the created DBM data from a separate processing system which is configured to generate the polyline data of the DBM.

[0052] The system 100 is also provided with a plurality of markers 110, where each marker 110 is located proximate to each of the control stations 108 positioned at the track 101. The markers 110 may be permanently affixed at the proximate location or may be configured to be removably affixed at the location such that it may be removed once the functionality described herein is achieved. The location of each marker 110 is such that the marker may be uniquely and unambiguously associated with a specific control station 108. Such association is enabled by each of the markers 110 being configured to be a unique readable marker where the marker 110 encapsulates data and can be read in captured imagery in order to identify the unique identity of the control station 108 (i.e. the encapsulated data relates to a unique identifier of the related control station 108, such as a serial number or any other numbering scheme of control stations 108 assigned by the system 100). Such markers may include a black and white barcode or QR code, or any other marker which is configured to enable unique identification of its associated control station 108 from captured images of the track 101 and the surrounding areas. In so doing, each of the control stations 108 may be established as a ground control point used in orthorectification when creating the 3D-DSM using the processing system 102. Furthermore, the polylines of the created DBM may be similar used to increase the accuracy of the orthorectification when creating the 3D-DSM using the processing system 102.

[0053] The processing system 102 may be configured to identify from received imagery of the track 101 and the surrounding areas, the markers 110 associated with the control stations 108. Similarly, the processing system 102 may be configured to identify from received imagery of the track 101 and the surrounding areas, the control stations 108 themselves. In each instance, this may be achieved using appropriately configured image recognition software 120 (see Figure 2) provided within the processing system 102. The processing system 102 may further be configured to extract the data encapsulated in the markers 110, again using the appropriately configured image recognition software 120. For example, the image recognition software 120 may be configured to identify and read the data encapsulated within a QR code or barcode provided as the marker 110. Similarly, processing system 102 may be configured to identify from received imagery of the track 101 and the surrounding areas, positions of the control stations 108 within the image. This may be achieved through standard image recognition techniques.

[0054] In some embodiments, the processing system 102 may instead be provided with user input in order to identify the markers within the received images and / or to identify the positions of the control stations 108 within the image. Additionally or alternatively, the processing system 102 may also be provided with user input which provides the data encapsulated within the markers 110 identified in the image. This may be performed in instances where the processing system 102 does not contain image recognition software configured to perform these tasks. In further embodiments, each of these tasks may be performed by a separate image recognition system (not shown) and the data may be provided to the processing system 102 via these alternative systems.

[0055] The processing system 102 is configured to utilise the received information relating to the markers 110 identified in the image, the data encapsulated within the markers 110, and the positions of the control stations 108 to uniquely identify the control stations within the image 108. Specifically, the processing system 102 may be configured to associate the data encapsulated within an identified marker 110 (i.e. a unique identifier of a particular control station 108) with a control station 108 which is positioned closest to the marker 110 within the image (which may be achieved through image recognition techniques). In so doing, the processing system 102 may therefore be able to associate each of the control stations 108 in the image with its established geolocation (in accordance with embodiments described above). In this manner, the identified control stations 108 in the image may be used for the purposes of orthorectification techniques to increase the precision and / or accuracy of a 3D-DSM created by the processing system 102.

[0056] in addition, in some embodiments when creating the 3D-DSM, the received (or created) data relating to the DBM (i.e. the created polylines) may also be utilised to enhance the precision and accuracy of the 3D-DSM. Due to the inherent accuracy of the created polylines and the fact that the position of these polylines with respect to the control stations 108 is also known to a high precision, the polylines themselves can act as further ground control points (GCPs) for the purposes of orthorectification of the captured image data. In particular, the polylines of the DBM are virtual representations of the positions of the outer edges of the two boundary lines of the track 101 which will be visible in the captured image data. The data provided regarding the position of the polylines of the DBM relative to the control stations 108 (whose geolocation is precisely known) in effect enables the polylines to be used as additional continuous ground control points when orthorectifying the captured image data. This enables the accuracy and precision of the created 3D-DSM to be further enhanced.

[0057] In some embodiments, the corners of critical track markings such as the starting grid lines and red-white stripe kerbs and any other clear features can be used as additional ground control points for the purposes of orthorectification. This may be achieved through use of the image recognition software 120 in the processing system 102 or through identification of these critical track markings by user input (e.g. a user may review images provided to the processing system 102 and highlight the positions of the relevant markings).

[0058] Referring now to Figure 2, there is shown in greater detail a schematic block diagram of the processing system 102 of Figure 1. The processing system 102 firstly comprises a receiver 130 configured to receive incoming data in accordance with embodiments described herein. In particular, the receiver 130 is configured to at least receive image data 122 relating to an area for which a 3D-DSM is to be created. In the case of embodiments discussed above, this area comprises the motor racing track 101 and areas immediately surrounding the track 101, although it is to be understood than any area suitable for modelling via 3D-DSM may be used, in some embodiments, the receiver 130 is configured to receive this image data 122 directly from one or more drones 104 (equipped with a camera to capture images) or other airborne vehicles through a wireless communications network 106 in accordance with embodiments described above. Such wireless communication may be enabled via radio frequency communication. Alternatively, the receiver 130 may receive this image data 122 using any other suitable form of communication, which enables the image data 122 to be received in the required format to enable functionality described herein. In some embodiments, the receiver 130 is configured to receive the relevant image data 122 from a separate system (not shown) to which the one or more drones 104 are configured to provide captured image data .

[0059] The receiver 130 may also be configured to receive the DBM data 124 relating to a created DBM of the track 101 which establishes boundary lines of the track 101 for which image data 124 has been captured in a virtual space, in accordance with embodiments described above. In some embodiments, the receiver 130 is configured to receive this DBM data 124 directly from one or more control stations 108 either through a wireless communications network 106 or an appropriately configured wired configuration in accordance with embodiments described above. Such wireless communication may be enabled via radio frequency communication. Alternatively, the receiver 130 may receive this DBM data 124 using any other suitable form of communication, which enables the data to be received in the required format to enable functionality described herein. In some embodiments, the receiver 130 is configured to receive the relevant DBM data 124 relating to a created DBM from a separate system (not shown) to which the one or more control stations 108 are configured to provide the data to.

[0060] In some embodiments, the image data 122 and / or the DBM data 124 relating to the created DBM, may be received incrementally (i.e. sequential images or data from control stations 108 may be received separately). In alternate or additional embodiments, all required image data 122 and / or all the DBM data 124 may be received simultaneously.

[0061] In some embodiments, the processing system 102 may be configured to perform the creation of the DBM based on raw DBM data received by the receiver 130 in accordance with the methods detailed in Annex 1. In such embodiments, the raw DBM data relating to a plurality of points on a preestablished boundary of the track 101 for the creation of a single DBM may be received incrementally (i.e. a series of data points may be received separately). In alternate or additional embodiments, each of the data points relating to a single DBM may be received simultaneously following ail data points having been previously measured.

[8062] The processing system 108 further comprises the data store 132 and a 3D-DSM creation processor 134, which is communicably coupled to the data store 132 and to the receiver 130. The 3D-DSM creation processor 134 may be configured to receive data from the receiver 130 and store that data in the data store 132. in particular, this received data may be in accordance with embodiments described above. In some embodiments, when data is received by the receiver 130, that data may be directly stored in the data store 132 after being received either prior to, or simultaneously with the data being provided to the 3D-DSM creation processor 134. The 3D-DSM creation processor 134 may also be configured to retrieve any required data previously stored in the data store 132 via a communicable coupling between the two.

[0063] The data store 132 may be configured to store image recognition software 120 to recognise and identify features within the received image data. In particular, the image recognition software 120 may be configured to identify the control stations 108 and markers 110 within the image data. The image recognition software 120 may also be configured to read the data stored within the markers 110 identified within the image data 122 (e.g. to read the data encapsulated within a QR code or a barcode). Similarly, the data store 132 may be configured to store appropriate reading software (not shown) for reading the data encapsulated within the markers 110 (such as QR code or barcode interpretation software).

[0064] The 3D-DSM creation processor 134 is configured to create 3D-DSMs 128 in accordance with embodiments described herein utilising the received image data 122 and received DBM data 124. in particular, the 3D-DSM creation processor 134 is configured to create an initial 3D-DSM by utilising the received image data 122 in order to create a virtual image of the area encapsulated within the image data, which in the embodiment illustrated in Figure 1, comprises the track 101 and surrounding areas of the track, where the surrounding areas encapsulate at least the control stations 108 and the markers 110 located proximate to the surface of the track 101. The initial 3D-DSM is created in accordance with standard techniques.

[0065] The 3D-DSM creation processor 134 is also configured to orthorectify the initial 3D-DSM on the basis of the received image data 122 and received DBM data 124.

[0066] In some embodiments, the orthorectification is performed at least partially on the basis of the position of the control stations 108 and the markers 110 within the image. In particular, the 3D-DSM creation processor 134 may be configured to first identify any control stations 108 or markers 110 which are present in the image data which has been received. This identification may be performed upon receipt of the image data 122 via the receiver 130. Alternatively, in embodiments where the image data 122 is stored in the data store 132, the 3D-DSM creation processor 134 may be configured to retrieve the image data 122 from the data store 132 to perform this identification process. The 3D-DSM creation processor 134 may also be configured to utilise any image recognition software 120 stored in the data store 132 (where applicable) to perform the identification process. Such image recognition techniques are well known to the skilled person and will not be discussed further here. In some embodiments, the features to be identified within the image data 122 may be previously identified within the image data prior to receipt at the receiver 130. This may be achieved via the use of external systems or through a process of manual identification by a user. In such instances, when the image data is provided to the receiver 130, position data (not shown) identifying the positions of the one or more control stations 108 and the markers 110 will also be provided to enable the 3D-DSM creation processor 134 to locate the relevant features within the image data 122. In other embodiments, the 3D-DSM creation processor 134 may be configured to present image data to a user when creating the 3D-DSM, and receive user input identifying the one or more control stations 108 and associated markers 110. In such embodiments, the processing system 102 may be configured to output the image data on a display (not shown) and may be configured to receive user input (position data) via the receiver 130 which highlights the position of the relevant features in the image data 122 in a format which is usable by the 3D-DSM creation processor 134.

[0067] The 3D-DSM creation processor 134 may also be configured to read information stored within the identified markers 110 identified within the image data 122 (e.g, to read the information encapsulated within a QR code or a barcode). In particular, the 3D-DSM creation processor 134 may be configured to implement marker-reading software (not shown) stored within the data store 132 which is configured to read the data encapsulated within the marker (e.g. QR code or barcode interpretation software). Alternatively, and similarly to embodiments described above, this process may be performed by an external system or via manual user identification in the case of some forms of marker (e.g. where the marker is plain text identifiable from images), and may be provided to the receiver 130 in conjunction with the received image data 122. In other embodiments, the 3D-DSM creation processor 134 may be configured to present image data to a user when creating the 3D-DSM in order to receive user input indicating the data encapsulated by the markers 110. In such embodiments, the processing system 102 may be configured to output the image data on a display (not shown) and may be configured to receive user input via the receiver 130 which provides the data encapsulated by the identified markers 110 in the image data in a format which is usable by the 3D-DSM creation processor 134.

[0068] The 3D-DSM creation processor 134 may be further configured to utilise the identified data encapsulated by the one or more markers 110 in order to associate the identified one or more control stations 108 with a unique identifier for a particular control station 108. The 3D-DSM creation processor 134 may be configured to associate an identified control station 108 within the image data 122 with an identified marker 110 within the image data 122 based on the relative positions of the two within the image data (e.g. the marker 110 closest to a particular control station 108 may be associated with that control station 108). Based upon this association, the 3D-DSM creation processor 134 may then be configured to retrieve relevant control station data regarding the uniquely identified control station 108 from the data store 132. This data may include relevant information regarding the geolocation of the control station 108 (which may be provided with DBM data 124 relating to the created DBM). The 3D-DSM creation processor 134 may then be configured to perform orthorectification of the initially created 3D-DSM based upon the geolocation of the identified control stations 108 within the image. This is performed in accordance with standard orthorectification techniques. The effect of this is to create a 3D-DSM 128 which is more accurate than prior art 3D-DSMs to the ground truth position of the track 101 and surrounding areas.

[0069] In additional embodiments, the 3D-DSM creation processor 134 may be configured to perform further orthorectification based upon the DBM data 124. As discussed above, due to the accuracy of the DBM created using the techniques described in Annex 1 and the known relative positioning between the DBM and the control stations 108, the boundary lines themselves can act as further GCPs within the image data 122. In such embodiments, the 3D-DSM creation processor 134 may be configured to use orthorectification techniques using the boundary lines of the tracks (for which the relative position to the geolocation of the control stations 108 is accurately established) within the image data, again using conventional orthorectification techniques. This may be achieved through the use of a curve fitting engine (not shown) provided within the 3D-DSM creation processor 134.

[0070] In yet additional embodiments, the 3D-DSM creation processor 134 may be configured to further orthorectify the initial 3D-DSM using the corners of critical track markings such as the starting grid lines and red-white stripe kerbs and any other clear features. This may preferably be achieved through use of the image recognition software 120 stored in the data store 132 or even through identification of these critical track markings by user input in a manner similar to identification of other features in the image data 122 as described above.

[0071] Following the creation of the orthorectified 3D-DSM 128, the 3D-DSM creation processor 134 may be configured to generate a visual representation of the 3D-DSM. The visual representation may then be stored in the data store 132 for later retrieval by the 3D-DSM creation processor 134. When required, the virtual representation may be retrieved by the 3D-DSM creation processor 134 and then transmitted via the transmitter 136 to an appropriately configured display screen (not shown). In some embodiments, the visual representation may be generated by systems external to the processing system 102. In such embodiments, the data relating to the 3D-DSM 128 may be provided to the relevant external system via the transmitter 136 in accordance with embodiments described above. This created data may then be used either to generate the visual representation of the 3D-DSM, or to be used in any further suitable manner (e.g. to perform an analysis of the data for veracity, enhancement of the created 3D-DSM, external storage etc).

[0072] In some embodiments, the processing system 102 may be configured to create a 3D-DSM 128 which extends beyond the track 101 and the immediate areas surrounding the track 101 described in embodiments discussed above. In such embodiments, the processing system 102 may receive, via the receiver 130, additional data regarding the areas extending beyond the areas immediately surrounding the track 101. The areas of the 3D-DSM which represent these more remote areas will typically be of lower accuracy and precision than the areas orthorectified using the DBM model data as described above, however they will be of sufficient accuracy for the purposes described herein. The 3D-DSM creation processor 134 may be configured to create the 3D-DSM in these areas in accordance with standard techniques. In some instances, the positions of the control stations 108 may still be used for the purposes of orthorectification in dependence upon whether these stations 108 (and their associated markers 110) are visible in the relevant image data. Such orthorectification and association may be performed as described above. Additionally, where applicable, the DBM data may also be used for orthorectification (in a manner as described above) if the distinct track boundaries are visible in the relevant image data 122.

[0073] In the above discussion, the embodiments describe focus on the surface level aspect of the 3D-DSM to be created (i.e. elements related to the digital terrain model (DTM) and digital surface model (DSM)). In order to introduce a three-dimensional aspect to the 3D-DSM, the processing system 102 may be configured to receive data relating to the “fagade” elements of the track 101 and the surrounding areas, where the fagade elements represent vertical, near vertical, or otherwise steep elements in the area to be modelled which may not be accurately represented by the image data used to create the DSM and DTM elements. Such facade elements may comprise image data which contains representations of the vertical or near vertical elements of the track 101 and the relevant surrounding areas to be modelled. These elements are typically orthogonal or near orthogonal to the surface of the DSM. This image data may be captured by drones 104 who are configured to follow a flight path which passes the relevant features and which has a camera oriented towards the vertical surfaces (namely tilt the camera from a downward facing position (0 degrees to the vertical) to an angle <90 degrees to the vertical). A typical flight path may involve flying over one track 101 edge at an altitude equal to the track width whilst orienting the camera at approximately 45° from vertical towards the opposite track edge. The captured image data may then be provided to the receiver 130 of the processing system 102 directly by the relevant drones 104 and utilised by the 3D-DSM creation processor 134 when creating the 3D-DSM. In some embodiments, the facade image data which is received by the processing system 102 may have been previously captured and stored in a memory (not shown) or the data store 132. Upon determination that the processing system 102 is to create a new 3D-DSM, the processing system 102 may be configured to retrieve the relevant images from the memory or data store 132.

[0074] In embodiments of the present disclosure, the processing system 102 may be configured to receive additional inputs via the receiver 130 in order to enhance the realism and interactivity of the created 3D-DSM, as well as to improve the image quality of the 3D-DSM. These additional inputs may also be provided in order to include features which may not be present or easily viewable in the received image data. Such data may include data models of peripheral features which one may typically find in situ at a race track 101 or other area to be modelled. Such peripheral features may include (but are not limited to) flags, advertising boards, wildlife etc. In some embodiments, such data models may be static (i.e. the features do not move in the created 3D-DSM). In other embodiments, the data models may be dynamic, and may result in the insertion of moving imagery within the 3D-DSM. These data models (not shown) will typically be stored within a database (not shown) and may be provided to the processing system 102 when a 3D-DSM is to be created. Alternatively, any data models (not shown) which may be required when creating the 3D-DSM may be stored in the data store 132 of the processing system 102 and retrieved by the 3D-DSM creation processor 134 when creating the 3D-DSM.

[0075] The created 3D-DSM data will typically be configured to enable the creation of a 3D area of the area to be modelled which can be freely manipulated to allow a user to generate a view from any origin position within the model. In some embodiments, the view may originate at any point within the model and the user may adjust the angle of the view through a complete sphere (i.e. through a solid angle of 4k steradians). Providing such view manipulation in a 3D model image may be achieved through a variety of techniques known to the skilled person and therefore such techniques will not be further discussed here.

[0076] In some embodiments of the present disclosure, the 3D-DSM creation processor 134 may be split into separate elements; one which focusses on the created on the surface components (e,g, the DTM and DSM) and one which enhances the created DSM by introducing additional fagade elements and other static or dynamic embellishments (as described above) to create the 3D-DSM.

[0077] The created 3D-DSM data 128 may be used for a plurality of potential purposes. As discussed above, a visual representation of the track 101 and the surrounding areas may be created using the 3D-DSM and may be used to create a user controllable image which enables the user to view the model from a plurality of views using a virtual camera. The virtual camera may be positioned at any origin point within the 3D-DSM model. In some embodiments, the 3D-DSM may be further augmented by the positioning of vehicles within the model. In particular, the 3D-DSM may be configured to include three-dimensional models of vehicles which are tracked as they traverse the track 101 that the 3D-DSM represents. The vehicles may be remodelled on the 3D-DSM based upon their tracked position on the real-life track 101. In some instances, the vehicles' locations may be tracked over a period of time (e.g. the duration of a race event) and models of the corresponding vehicles may be reproduced on the 3D-DSM based upon their position at a particular time frame (i.e. the entirety of the race event may be reproduced virtually using the 3D-DSM and the tracking data). In some cases, such a race event may be monitored and tracked and then subsequently modelled virtually using the 3D-DSM at a later time. In other instances, the reproduction of the vehicles on the (previously created) 3D-DSM may be configured to occur substantially in real-time, i.e. as the race event is proceeding. In effect, the reproduced vehicles on the 3D-DSM may be used to recreate virtually an event which is occurring in real time, namely as the event is occurring. The tracking system used to achieve this real time tracking and subsequent reproduction is in accordance with the system described in Annex 2.

[0078] There are a plurality of use cases for tracked vehicles included on a created 3D-DSM 128. A first use case is viewing of a race event which may be in part or partially controlled by the user in a virtual environment, in existing systems, typically a user viewing a real-life race event (captured by cameras) will only be able to watch a particular view as determined by a broadcaster. In some cases, multiple views may be selected by a user, however these are still limited by the views which are made available by cameras positioned at a track 101. A reproduced event using the 3D-DSM and reproduced vehicles (detailed above) allows a user to fully control the view which they use to watch the event by reproducing the event accurately in a virtual space and then allowing the user control of the position of the camera. Since the track and the vehicles are reproduced accurately in real time, the virtual reproduction of the race event is effectively indistinguishable from a version of the race which is captured by a video camera. In addition, since an individual version of the virtual race event is provided to each user, each user may personalise their view in a manner which does not affect other users.

[0079] A second use case is to enable race steward to better review incidents which occur on a track to enable appropriate decision to be made. Typically, race stewards can only make decisions based on captured footage and therefore may not be able to make correct decisions where minimal footage of an incident exists. Such incidents may include crashes or where a vehicle may transgress over a boundary line. Using the approach detailed above, race stewards are able to control the view and go back to where a potential incident occurred in order to make a more informed decision.

[0080] Another use case is for the purposes of driver training. It is often advantageous for a new or existing driver to review previous data (both video footage and telemetry data) in order to improve their future performance. Again, using previously existing techniques, the video footage is limited to that which has been captured by cameras. Using present embodiments, drivers are able to control the view during an event and focus on the performance of a particular vehicle (either their own or a competitor) in order to improve their future performance.

[0081] It is to be appreciated that the uses noted above are provided for the purposes of example only and are not intended to be limiting.

[0082] Referring now to Figure 3, there is shown an example of an arrangement of a 3D-DSM Creation Processor 134 in addition to example inputs and outputs into the 3D-DSM Creation Processor 134, in accordance with embodiments described above. In particular, this arrangement is intended for the creation of a 3D-DSM 128 of a race track 101. In the arrangement of Figure 3, the 3D-DSM Creation Processor 134 is shown as comprising two separate elements, a track DTM generator 150, and a 3D model combiner 152, where the two elements are separated by function. In particular, the track DTM generator 150 is configured to receive inputs in order to generate the elements of the 3D-DSM relating to surface level features (i.e., the DTM and DSM). The 3D model combiner 152 is configured to receive inputs related to fagade elements of the track 101 and surrounding elements and any additional features, in addition to the DTM and DSM data created by the track DTM generator 150 in order to create the final 3D model (i.e., the 3D-DSM). It is to be appreciated that the two separate elements 150, 152 may either be carried out by a single appropriately configured processor or may be carried out by two separate processors, with each being configured to perform the functions as described above. It is also to be appreciated that in embodiments of the present invention, the 3D-DSM 128 may be created either by first creating a two-dimensional DSM and then modifying this to create the final three-dimensional model with fagade elements and embellishments as described herein, or the two processes may be carried out substantially simultaneously.

[0083] Figure 3 also illustrates examples of two different types of inputs 154, 156 into the 3D-DSM Creator Processor 134. In particular, DTM inputs 154 represent inputs which are used to create the two-dimensional DSM aspect of the model of the track 101. These DTM inputs 154 include photographic surface digital images 154a of the track 101, as well as digital boundary models 154b of the track 101, which have been created in accordance with embodiments described herein and in Annex 1. In embodiments in which the 3D-DSM Creator Processor 134 comprises two separate elements 150, 152, these DTM inputs 154 are provided to the track DTM generator 150. Enhancement inputs 156 represent inputs provided to the 3D Model Combiner 152 of the 3D-DSM Creator Processor 134 used to enhance a created DTM or DSM in order to create the final three-dimensional DSM 128. These enhancement inputs 156 may include static data models 156a and dynamic data models 156b of peripheral features (such as flags), as well as fagade image enhancements 156c and proximate data (such as image data) 156d relating to the surroundings of the track 101 not initially created in the two-dimensional DTM / DSM. These are provided in accordance with embodiments described above.

[0084] The Track DTM Generator 150 comprises an Orthorectification Processor 170, a Curve Fitting Engine 172 and an Additional Feature Processor 174. The Orthorectification Processor 170 is configured to determine GCPs from the DBM. As described above, each of the control stations 108 (identified by its associated marker 110A / B) may be established as a GCP and used in orthorectification when creating the 3D-DSM 128. However, in addition to the identified control stations being used as GCPs, due to the accuracy of the DBM 154b and the known relative positioning between the DBM 154b and the control stations 110, the established continuous boundary lines of the racetrack 101 can act as further GCPs within the image data 154a. In this way, the Orthorectification Processor 170 utilises the polylines of the DBM 154b to increase the accuracy of the orthorectification when creating the 3D-DSM 128.

[0085] The Curve Fitting Engine 172 is also configured to assist in orthorectification process when the data supplied to the Track DTM Generator 134 is in the form of discrete points which lie on the boundary of the racetrack 101. When the DBM 154b is provided in this format, the Curve Fitting Engine 172 can be used to create continuous boundary lines of the racetrack 101 by fitting curves of continuous lines to the DBM representation 154b. These fitted curves of track limits can then be used as GCPs to help further orthorectify the photogrammetric images 154a so as to improve the accuracy ofthe3D-DSM 128.

[0086] The Additional Feature Processor 174 is configured to determine additional GCPs from clear features within images 154a which have accurate relative positional information provided in the DSM. For example, the comers of critical track markings, such as starting lines and red-white stripe kerbs of the racetrack or any other clear features of the race-track surface can be used as additional GCPs. The corresponding feature is found in the captured programmatic images and given their position is known to a high accuracy from the DSM 154b, they can be used as GCPs in orthorectification. These elements of the Track DTM Generator 150 ensure that the photogrammetric images 154a are optimised and can be matched precisely to the DBM 154b.

[0087] Figure 3 also provides examples of outputs of the 3D-DSM Creator Processor 134 for use of the created 3D-DSM 128. In particular, illustrated is the output 128 from the 3D-DSM Creator Processor 134 to a digitised racing circuit creator 158, which is configured to utilise the created 3D- DSM 128 in conjunction with other inputs 160 in order to create a reconstruction of race events which occur on a track 101, either in real-time or from previous data. In particular, the inputs may include tracking data from a dynamic vehicle tracking system 160a (in accordance with embodiments described in Annex 2 which enables tracking of vehicles using a low amount of data transfer, and 3D-digitaily scanned images of vehicles 160b which may be used in conjunction with the tracking data to provide a representative virtual image of the relevant vehicle to its corresponding tracked position in the 3D-DSM 128. As discussed above, there are a plurality of uses 162 for the combined 3D-DSM and tracked vehicles, including (but not limited to) control of views of a race event by a user, judging whether a vehicle has transgressed a boundary of the track by a race steward, and training of drivers. The 3D-DSM may also be implemented within video game technology in order to provide a virtual reconstruction of a track 101 which a user may virtually race on in a virtual vehicle (either against other users or in some instances, against virtually reconstructed vehicles which have been tracked as they traverse the race track 101 and reproduced in the 3D-DSM).

[0088] Referring now to Figure 4, there is shown a method of operation 300 of the above system 100. In particular, the method 300 is for generating a 3D-DSM 128 representing a race track 101 and its surroundings. It is to be appreciated that the 3D-DSM may represent any suitable real-world environment where appropriate. For ease of reading, the following description will assume that the area to be modelled is the race track 101. [00S9] The method 300 proceeds by the processing system 102 receiving, at Step 302, image data relating to the track 101 and surrounding areas that are to be modelled in the 3d-DSM by the 3D-DSM creator processor 134. This may be received via the receiver 130 of the processing system 102. The image data may be received directly from the device capturing the image data (e.g., the one or more drones 104) in accordance with embodiments described above. Alternatively, the receiver 130 may receive this data using any other suitable form of communication, which enables the data to be received in the required format to enable functionality described herein. In some embodiments, the receiver 130 is configured to receive the relevant image data from an intermediary system which initially receives the image data from the one or drones 104 (or alternative image data source). At this stage of receipt, the image data may be used to create an initial digital terrain model and / or a digital surface model. Such a model at this stage will typically not be orthorectified and therefore may not be accurate to ground truth. The received image data will also typically include other image elements which can be used for creation of the 3D-DSM in accordance with embodiments described above (e.g., fapade elements 156c, data models of peripheral features 156a, distant surroundings imagery 156d etc).

[0090] Following receipt of the position data, the method 300 proceeds by the processing system 102 receiving, at Step 304, DBM data 154b relating to a created digital boundary model (DBM) of the track 101 which establishes boundary lines of the track 101 in a virtual space. The created DBM includes data in accordance with embodiments described above. In particular, the received data will typically include the geolocation of each control station 108 and the positions of the created polylines relative to each control station 108. The control stations 108 will each typically be labelled with a unique identifier within the data such that the positions of the polylines relative to each control station 108 may be readily defined and the control station 108 may be readily associated with its geolocation.

[0091] In some embodiments, the processing system 102 may be configured to receive raw data from one or more of the control stations 108 located at the track, and may further be configured to generate the set of polylines of the DBM in accordance with methods described in Annex 1. These methods will not be described in further detail here. In other embodiments, the raw data may be received from another source (e.g. a separate system to which the raw data is transmitted to from the control stations 108). In some embodiments, the processing system 102 may be configured to simply receive the created DBM data from a separate processing system which is configured to generate the polyline data of the DBM.

[0092] It is to be appreciated that the receipt of data in Steps 302 and 304 may occur in the opposite order or substantially simultaneously In some embodiments.

[0093] Following this, the processing system 102 identifies, at Step 306, markers 110 and control stations 108 which are present in the image data received by the processing system 102. This identification may occur either through use of appropriate image recognition software 120 or through input provided by a user in accordance with embodiments described above. The result of this stage is that the position of the relevant features of interest within the image are then known.

[0094] The processing system 102 then reads, at Step 308, the data encapsulated by the markers 110 identified in the image data. In particular, the 3D-DSM creation processor 134 may be configured to run software stored within the data store 132 which is configured to read the data encapsulated within the marker (e.g. QR code or barcode interpretation software). Alternatively, and similarly to embodiments described above, this process may be performed by an external system or via manual user identification in the case of some forms of marker (e.g, where the marker is plain text identifiable from images), and may be provided to the receiver 130 in conjunction with the received image data. In other embodiments, the 3D-DSM creation processor 134 may be configured to present image data to a user when creating the 3D-DSM in order to receive user input indicating the data encapsulated by the markers 110. In such embodiments, the processing system 102 may be configured to output the image data on a display (not shown) and may be configured to receive user input via the receiver 130 which provides the data encapsulated by the identified markers 110 in the image data in a format which is usable by the 3D-DSM creation processor 134. The data encapsulated by the markers will typically include at least an indication of identification data of a control station 108 (typically the control station 108 located proximate to the relevant marker 110) and its associated geolocation. [009S] The processing system 102 then associates, at Step 310, the data encapsulated by an identified marker 110 with an identified control station 108 and assigns the relevant data to the associated control station 108. In particular, the association will occur by associating a particular marker 110 with the control station 108 most proximate to the marker 110 within the image. Then the data previously read from the marker 110 (i.e., including the identification data of a control station and its geolocation) will be assigned to the control station 108 in the image. In effect, this stage establishes ground control points (GCPs) within the image data for the purposes of orthorectification. Once this stage has been completed, the processing system 102 orthorectifies, at Step 312, the image data based upon the received DBM data and the unique identification of the control stations 108 within the image. Specifically, the received DBM data 154b will include geolocations of each unique control station 108 in the physical world. Since these have now been identified within the image data and associated to data accordingly, they can be used to orthorectify the image to create a two-dimensional DSM accordingly. In some embodiments, the orthorectification may also include using created polyline data (which in effect provides a continuous set of ground control points as described above) included as part of the received DBM data. Further, in some embodiments, the corners of critical track markings such as the starting grid lines and red-white stripe kerbs and any other clear features can be used as additional ground control points for the purposes of orthorectification in accordance with embodiments described above.

[0096] The processing system 102 then creates, at Step 314, the 3D-DSM 128 based on the previously created two-dimensional DSM and the other image elements received in Step 302. In particular, the processing system 102 may be configured to include the facade elements 156c in order to introduce three-dimensional elements to the created DSM, and / or to include peripheral features (e.g. flags, advertising boards) 15Sa to the model to enhance the realism of the model. The result of this step is the creation of a 3D-DSM 128 which may be exported for use in accordance with embodiments described herein.

[0097] Following this, the method 300 continues by exporting, at Step 316 the 3D-DSM 128 via the transmitter 136, to an external system for use in accordance with embodiments described above. This may comprise providing a 3D-DSM data set to an external system which produces a virtual model representing the data set which may then be utilised in accordance with embodiments described above. This may also comprise the transmitter 136 exporting a virtual model representing the data set which is created by the processing system 102. The method 300 then ends at Step 318.

[0098] As described, the system of the present application can be configured to accurately map a racetrack 101 and vehicles moving on the track 101 into a digital environment. To achieve this, elements of the sensor tracking system must be arranged appropriately. Such an exemplary racetrack is displayed in Figure 5. Figure 5 shows a photograph of an exemplary racetrack to which has been added a set of 15m high mast, around 50m apart, each with two or more sensors 52 at the top each viewing a different portion of the racetrack (these masts and sensors (cameras) are shown as overlaid markings). The initial and approximate placements of the masts 51 shown in Figure 5 have been chosen based on a 60 degree square field of view (FOV) of the sensors 52 and a rough geometrical calculation. The precise relationship between each proposed real sensor’s 52 FOV and the real racing circuit can be modelled by combining the sensor’s modelled position, pose and field of view with the racing circuit 3D-DSM 128.

[0999] Figure 6 illustrates the precise sensing footprint across the terrain of all the sensors 52 around the racetrack 101. Each quadrilateral (almost a trapezoid) 54 shown in Figure 6 represents the FOV of a particular sensor 52. As can be seen, some of the sensors 52 have partially overlapping FOVs in order to cover the entire surface of the racetrack 101 and preferably proximate adjacent regions where a vehicle may inadvertently leave the track. The placements of the 15m masts 51 and the sensor 52 poses can be refined in the model to ensure that the required tracking coverage is achieved -- including allowance for factors such as the maximum height of any of the vehicle markers above the ground, obscuration of sensor-to-marker lines of sight, etc. For practical implementation, the final mast placements need to be approved by the relevant safety authorities. The result of the described exemplary set up and modelling is the creation of a digital corridor model of known dimensions in all areas where vehicle tracking is required. The final mast 51 placement positions and camera poses can be extracted from the model and used to construct the circuit tracking system in the real world, the model becoming a digital twin of the real-world circuit with its new tracking system.

[0190] Figure 7 illustrates in greater detail a mast 51 of the exemplary racetrack of Figure S. The sensors 52 attached at the top of the mast 51 are shown. In addition to the tracking sensors, Figure 7 shows the tracking system may include a pole compute module (PCM) 56 (which acts as a localised edge computer for each tracking sensor pair) on each mast 51 to which the sensors 52 are connected. The PCM 56 of each mast 51 is connected to a central race control system (RCS) via fibre cables 58 in this embodiment, where the accurate tracking data of all vehicles on the circuit is aggregated and displayed. This information can also be distributed by the RCS more widely to other servers, such as gaming servers via the internet.

[0101] In some embodiments. The RCS of the exemplary race track described in Figures 5 to 7, is connected to a cloud event hub (CEH) (a central computing hub). Referring to Figure 8, live or recorded data can be distributed around the world into various application markets (possibly involving other computing hubs and end-user devices) through the CEH and the CEH may also receive data from other motor racing circuits with tracking systems as shown in Figure 8.

[0192] Each PCM 56 contains the relevant portion of the static DCM (relevant to its location on the track) so that the vehicle tracking data the PCM 56 creates locally is in the DRC model coordinate system. The RCS contains the complete static DCM so that it aggregates all vehicle tracking data in the DRC model coordinate system. All PCMs and the RCS maintain a common network time clock such that all tracking data can be accurately synchronised. The sensors 52 on the masts 51 operate at a measurement frequency suitable for the applications described below, typically 90-100Hz, and all computations in the PCMs operate in real time, typically completing in 10-11ms (the corresponding time window for the measurement frequency of 90-100Hz). All network connections between PCMs 56 and the RCS have low transmission latencies such that the RCS and then the CEH can use or transmit tracking data in close to real time.

[0103] This completes the basic description of the systems and methods involved in a fully digitised racing circuit (DRC). There now follows, by way of illustrative example, a primary use of the system involving the consumer-controlled viewing of motorsports events. This aspect of the disclosure utilises the disclosure of Annex 2 to provide greatly reduced transmission bandwidths and reduced data storage requirements for tracking data to enable more efficient viewing of live or historic events with the ability of the viewer to control the point of points of view as is explained below. Three illustrative types of consumer are also given as examples. The first is a race viewer, the second a race adjudicator, and the third a racing team that is practicing.

[0104] The primary example is a motor racing fan that wishes to use a digital streaming media device such as television, phone, tablet, or computer, connected to the internet, together with a live or recorded streaming service such as YouTube, to display high-fidelity video streams of the event from viewpoints which they can define and control in real time. As the consumer requires to control viewpoints, the use of a video streaming service like You Tube is replaced by the service provided by the system from the CEH as described above. Having requested a viewing session by accessing the CEH, and probably paying for the service, the CEH first downloads to their device the static DCM, the digital models of ail the vehicles, and a software package that interacts with the DRC to create the high fidelity displays on the consumer’s device and provides all the control functions for the user to select viewpoints.

[0105] This software package is based on graphics pipeline software technology similar to that used by computer games, simulators, etc in that it uses a 3D model of a motorsport event captured in industry standard formats, such as in industry standard packages such as Blender®, to provide 2D viewpoints on screens or immersive VR devices, in this present embodiment of the disclosure, however, the movement of vehicle models is being determined directly and live (or recorded) from vehicle tracking data produced by the DRC and streamed by the CEH. As the static DCM employs high-fidelity photogrammetric imaging the quality of the viewing experience is very similar to viewing live streams generated by video or tv cameras, but is doing so by receiving very efficient vehicle tracking data streams compared to video (typically Kb / s compared to 10s of Mb / s). In some cases, depending on the fidelity of the photogrammetric imagery and the graphics processing capability of the consumer’s device, it may be necessary to decimate the static DCM data, reducing its volume at the expense of some degree of visual fidelity. Again, this is standard practice in computer games, simulators etc and can be applied without detracting from the viewing experience.

[0106] There are refinements to the above-described method that can further increase the reality of the viewing experience. The most important is the addition of weather and lighting condition modelling. Standardised weather and lighting data can be measured at the real motorsport event at regular intervals, day or night, and updates provided to the consumer device as supplementary data from the CEH. The software package described above is able to accommodate the alterations to the 2D imagery required by a wide range of weather and lighting conditions, including the display of shadows which are dependent upon and change with the time of day.

[0107] Furthermore, as mentioned above, augmentations (from dynamic data models) can be added to the 2D displays either as constant augmentations, static or dynamic, loaded as part of the static DCM, examples include static advertising hoardings, fluttering flags, etc, or can be refreshed or changed intermittently via the data stream from the CEH. As the CEH can have information about the consumer then advertising can be tailored to countries, regions or even individual users. Furthermore, graphic overlays showing timing data, race positions etc can be added. Furthermore, augmented views can be created even within the structure of a vehicle, perhaps showing fuel tank levels, brake temperatures, etc in textual, numeric or graphic additions. Again, all accessible to consumers via selections and functions in the software package.

[0108] A secondary example is a race controller, marshal or adjudicator at a motor racing event who requires accurate and instant replays of events within the race, such as a vehicle possibly transgressing a driving boundary limit. Using the system and methods described above this can easily be achieved as the CEH stores the vehicle tracking data in a very efficient and hence easily retrievable manner. The consumer can request a series of replays of the same event from a series of differing viewpoints (angles, ranges, etc), even from underneath the vehicle. In this latter case it is to be appreciated that the model has the full car representation, including its underside, so placing the virtual camera between the underside of the vehicle and the road, moving with the vehicle, is as straightforward as any other virtual camera position / behaviour.

[0109] The final, illustrative example of a consumer, is a racing team of drivers and engineers practicing; attempting to improve the driver’s skills or to improve the car’s performance. This often involves repeated runs to tweak technique or tweak a car’s set-up. In this case, viewing very accurate replays from varying viewpoints and combining the accurately timed vehicle tracking data with telemetry data from the vehicle is a very powerful tool. The system described above can also be easily integrated into a high-fidelity racing simulator to provide an improved capability for both car and driver development.

[0110] Finally, it is to be emphasised that these are just illustrative uses of the systems and methods detailed in this disclosure.

[0111] Having described several exemplary embodiments of the present embodiments and the implementation of different functions of the device in detail, it is to be appreciated that the skilled addressee will readily be able to adapt the basic configuration of the system to carry out described functionality without requiring detailed explanation of how this would be achieved. Therefore, in the present specification, several functions of the system have been described in different places without an explanation of the required detailed implementation as this is not necessary given the abilities of the skilled addressee to implement functionality into the system. The scope of the present disclosure is only limited by the spirit and scope of the present claims.

[0112] Furthermore, it will be understood that features, advantages, and functionality of the different embodiments described herein may be combined where context allows. REFERENCES [1] Elaksher, A.; All, T.; Alharthy, A. A Quantitative Assessment of LIDAR Data Accuracy. RemoteSens. 2023,15, 442. [2] Rogers, S.R,; Manning, I; Livingstone, W. Comparing the Spatial Accuracy of Digital Surface Models from Four Unoccupied Aerial Systems: Photogrammetry Versus LIDAR. RemoteSens. 2020, 12, 2806. [3] Jimenez-Jimenez,S.I.; Ojeda-Bustamante, W.; de Jesus Marcial-Pablo, Ml.; Enciso, J. Digital Terrain Models Generated with Low-Cost UAV Photogrammetry: Methodology and Accuracy. ISPRS Int.J.Geo-Inf. 2021,10, 285. [4] Gindraux, S.; Boesch, R.; Farinotti, D. Accuracy Assessment of Digital Surface Models from Unmanned Aerial Vehicles' Imagery on Glaciers. Remote Sens. 2017, 9,186

Claims

1. A method of creating a digital terrain model (DTM) of a racetrack, the method comprising:obtaining digital boundary model (DBM) of the racetrack, wherein the DBM comprises a plurality of reference points used in the creation of the DBM, the plurality of reference points including locations of a master control station (MCS) and a set of secondary control stations (SCS), and each reference point being determined with respect to the locations of other reference points and the boundary of the racetrack;receiving:a plurality of aerially visible markers, each marker being positioned proximate a location of a reference point of the DBM and respectively at a predetermined position from each reference point; andcaptured ariel images of the racetrack including the visual markers;using the predetermined positions of the visual markers from the respective reference points to specify the locations of the reference points as a plurality of ground control points (GCPs) of the DTM; andorthorectifying the captured images with the DBM of the racetrack using the plurality of reference points of the DBM as GCPs for the DTM.

2. The method of Claim 1, wherein the receiving step comprises receiving captured ariel images of the racetrack including a boundary of the racetrack; and the using step further comprises adding the boundary of the racetrack as determined by the DBM, to the plurality of GCPs used in the orthorectifying step.

3. The method of Claim 1 or 2, wherein the obtaining step further comprises obtaining the DBM of the racetrack which includes one or more marked features of the racetrack and the receiving step further comprises receiving captured ariel images of the racetrack including the one or more marked features; and the using step comprises adding the one or more marked features as determined by the DBM, to the plurality of GCPs used in the orthorectifying step.

4. The method of Claim 3, wherein the one or more marked features include a raised kerb of the race track and / or a starting grid line.

5. The method of any of Claims 1 to 4, wherein the visual markers comprises visible QR codes or barcodes.

6. The method of any of Claims 1 to 5, further comprising receiving dynamic data tracking an object moving on the racetrack and combining the dynamic data with static data of the DTM to create a fully digitised racing circuit (DRC) model that comprises both static and dynamic data.

7. The method of Claim 6, wherein the dynamic data comprises data captured from a fixed location relative to a reference point.

8. The method of any of Claims 1 to 7, wherein the obtaining step comprises obtaining a DBM having an accuracy of <10mm at the boundary of the race track.

9. The method of any of Claims 1 to 6, further comprising capturing the ariel images of the racetrack including the visual markers.

10. The method of Claim 9, wherein the capturing step comprises capturing photogrammic aerial images from an ariel drone flying at an altitude equal to the width of the racetrack above one of two boundaries of the racetrack.

11. The method of Ciaim 10, the capturing step comprises capturing photogrammic aerial images from a camera angled at 45 degrees from the racetrack surface towards the other one of the two boundaries of the race track.

12. The method of any of Claims 1 to 11, wherein the DBM of the racetrack comprises a plurality of discrete data points representing different adjacent points along an boundary of the racetrack and the method further comprises fitting a curve to a plurality of discrete data points to determine a continuous digital representation of the boundary of the racetrack.

13. The method of any of Claims 1 to 11, further comprising creating a visual representation of the 3D-DSM.

14. The method of any of Claims 1 to 12, further comprising transmitting the DTM or a visual representation of the DTM to a third party server using a wide area network.

15. A method of creating a three dimensional (3D) Digital Surface Model DSM of racetrack, the method comprising:creating a DTM as described in any preceding claim;supplying elevation images of features in areas adjacent the race track; andcombining information derived from the elevation images into the DTM to create the 3D-DSM.

16. The method of Claim 15, further comprising acquiring feature information regarding peripheral objects In geographical areas adjacent the racetrack, and combining the feature Information with the DTM.

17. The method of Claim 16, wherein the acquiring step comprises acquiring predetermined static digital models of the peripheral features of objects.

18. The method of Claim 16 or 17, wherein the acquiring step comprises acquiring predetermined dynamic digital models of the peripheral features of objects which are modelled in movement.

19. The method of any of Claims 16 to 18, further comprising getting feature information regarding peripheral landscape in geographical areas adjacent the racetrack, and combining the feature information with the DTM.

20. The method of any of Claims 16 to 19, wherein the obtaining step comprises obtaining a DBM having an accuracy of <10mm at the boundary of the race track and the acquiring or getting step comprises acquiring or getting features information having an accuracy of <50mm in the geographical areas adjacent the racetrack.

21. A non-transitory, computer-readable media storing instructions that, when executed by the one or more processors, cause the one or more processors to carry out a method according to any of Claims 1 to 20.

22. A system for creating a digital terrain model (DTM) of a racetrack, the system comprising:a 3D-DSM creator processor configured to:obtain digital boundary model (DBM) of the racetrack, wherein the DBM comprises a plurality of reference points used in the creation of the DBM, the plurality of reference points including locations of a master control station (MCS) and a set of secondary control stations (SCS), and each reference point being determined with respect to the locations of other reference points and the boundary of the racetrack;receive a plurality of aerially visible markers, each marker being positioned proximate a location of a reference point of the DBM and respectively at a predetermined position from each reference point; and captured ariel images of the racetrack including the visual markers:use the predetermined positions of the visual markers from the respective reference points to specify the iocations of the reference points as a plurality of ground coniro! points (GCPs) of the DTM; andorthorectify the captured aerial images with the DBM of the racetrack using the plurality of reference points of the DBM as GCPs for the DTM.A

Images