AU2004234414A1

AU2004234414A1 - Feature set transformation

Info

Publication number: AU2004234414A1
Application number: AU2004234414A
Authority: AU
Inventors: Michael Hele Elfick; Michael James Fletcher
Original assignee: Geodata Information Systems Pty Ltd
Current assignee: GEODATA AUSTRALIA Pty Ltd
Priority date: 2003-04-30
Filing date: 2004-04-07
Publication date: 2004-11-11
Anticipated expiration: 2024-04-07

Description

WO 2004/097339 PCTIAU2004/000452 FEATURE SET TRANSFORMATION Background of the Invention The present invention relates to a method and apparatus for determining a transformation 5 for transforming a feature set defined with respect to a first network to a second network. In particular the invention relates to determining a transformation for mapping a cadastral data set to a new reference cadastre. The present invention also relates to a database for storing cadastral data, and in particular, 10 a method and apparatus for storing, retrieving and amending data within the database. Description of the Prior Art The reference to any prior art in this specification is not, and should not be taken as, an acknowledgement or any form of suggestion that the prior art forms part of the common 15 general knowledge. Real property (land) is a fundamental natural resource for any economic activity. Accordingly, it is desirable to be able to construct geographic information systems (GIS) and land information systems (LIS) in which areas of land can be uniquely identified, 20 allowing the ownership and value of land to be monitored accurately. This is usually achieved using cadastral systems, which are designed to provide land and real estate owners and interested parties with security of tenure while also providing an efficient means for the transfer of ownership and interests in the real estate market. 25 Most countries currently use dimension based cadastral systems in which areas of land are subdivided into a large number of individual parcels, each of which has dimensioned boundaries (the metes) and information showing the relationship of each boundary with physical objects and adjacent parcels (the bounds). 30 Whilst current methods of boundary definition have been adequate for a long period of time, because the techniques use dimensions, this provides a very indirect way of defining a point and disputes arise when different surveyors use different data to determine the WO 2004/097339 PCT/AU2004/000452 -2 location of a boundary. In particular the cadastral systems are based on data that has been collected for over 100 years and its quality and accuracy of each record is generally a function of the date of the 5 record itself and the surveying technology available at the time. Thus, for example, prior to 1880, survey transits were not commonly in use and angles were measured to the nearest half-degree. Over the years, developments in angle measurement technology in survey transits and theodolites have improved this accuracy and many of the modem instruments in use have an accuracy of close to five seconds of arc. Similarly, the technology for 10 manually measuring distances steadily improved until about 1970 when electronic distance measurement was introduced. Modem instruments can now measure distances of 1000 meters to a precision of better than five millimetres. In addition to this, the completeness of the record and the standard of presentation of the 15 plans and documents in the record are largely dependent on the authority which manages the record system in each locality. Thus for example, in countries where the state guarantees title, such as in Australia and New Zealand, the records are complete and the documents generally of a good quality. In other countries the records may be held in various locations with little supervision on the quality of presentation of document. 20 In contrast to this, a coordinate can provide a unique and unambiguous definition of a point and global positioning system (GPS) can provide the necessary technology to quickly and accurately locate that point. At present most surveyors compute new subdivisions laying in the corners by coordinates, then preparing "metes and bounds" documents to satisfy the 25 legal requirements for land transfer. Technology has allowed surveyors to adopt a coordinate based approach and this can be used to reform the land ownership record system. Accordingly, in most countries of the world there is a move towards improving the 30 dimensioned metes and bounds cadastre with a coordinated cadastre in which the location of land parcels is uniquely identified using geographical coordinates. In order to achieve this, coordinates have initially been used simply as another dimension to assist in the WO 2004/097339 PCT/AU20041000452 -3 location of parcel corners, but as control becomes more established, coordinates will become the improved and ultimately sole definition. In theory a coordinate cadastre could be determined by resurveying every parcel. 5 However, this would be totally impractical and unnecessary as there is already adequate dimensional data available within the existing survey record system to carry out this job provided that control data is applied where needed. Existing GIS and LIS systems typically utilise data sets formed from a number of layers. 10 Each layer will represent a different type of feature within the overall survey, map or the like. Thus, a typical cadastral data set will include a base layer defining packets or parcels, the boundaries of which are defined in accordance with land ownership. Subsequent layers will then provide data regarding respective features related to the land such as permissible land use, natural features, buildings, telephone poles, roads, rail networks, or the like. In 15 this instance, this allows individuals to select a respective layer from a cadastral data set and thereby observe the positioning of respective features with respect to land boundaries. The relationship between these layers is known as the "associativity". Accordingly, when there is any change or improvement to the cadastral network it is 20 preferable to simply transform the existing cadastral system and associated feature layers to the new, changed or improved parcel network Some existing GIS applications software packages provide ways to create points and lines in relative position to other features. However, in most cases, the relativity data is 25 discarded and only the absolute coordinates of the points are stored for each layer. Accordingly, if the base layer is changed at a later date, then the associated points on other related or associated layers are no longer in their correct relative position to the base layer. For example, telegraph poles may be located by their offset from street boundaries, and at 30 a later date the street boundaries are changed because a better value for their geographic position is available. If corrections are not applied to the telegraph pole layer, they will appear to be incorrectly located with respect to the street boundaries.

WO 2004/097339 PCT/AU20041000452 -4 As there may be many layers dependent on the base cadastral layer, any changes to that layer have to be allowed for in all the associated layers and in the past changing all of the associated layers has been an expensive manual and time consuming process. 5 This problem is exacerbated due to the fact that the most common method of creating the original coordinate cadastre in GIS systems is by digitising administrative documents such as tax maps. The accuracy of this data depends on the maps and the quality of the digitising process. Many of these maps were designed simply to show the relationship of 10 the various attributes to each other rather than being compiled to an accurate coordinate base. Consequently, the accuracy of position varies from place to place and any mistakes in the original map compilation are also carried forward into the digitised records. There are various propriety systems on the market to help manage the cadastral network 15 and associated layers on a database when the base cadastral framework is changed or upgraded. Broadly, these systems can be defined in three groups: * "Rubber Sheeting" techniques where coordinates are changed according to their proximity to selected control points. For example, ESRI fit a triangular mesh with nodes at each control point and then use either linear interpolation or surface fitting 20 techniques within each triangle to determine the correction for the points inside each triangle. In this case, the points are defined manually and therefore only a few may be used. e "Least Squares Adjustment" procedures where observation equations are formed from coordinates of nodes in networks in the associated layer. These are used in a least 25 squares process similar to a geodetic network adjustment to compute corrections to coordinates based on selected control points. By including extra constraints, this technique can maintain features of the adjusted network such as shape and relative scale. * "Measurement Based Adjustment" techniques where a connection between each node 30 and the cadastre is created and stored as a separate layer to be used to restore the associativity when the base cadastral framework is changed or updated.

WO 2004/097339 PCT/AU2004/000452 -5 All of these systems provide some degree of success depending on the nature and size of the changes to the base cadastral framework and the structure of the associated layer. However, a number of drawbacks also exist. 5 For example, the measurement based approach carries significant overhead in the building and maintenance of the "connection layer". This is labour intensive and, therefore results in a very expensive process, with significant set-up costs. The least squares adjustment approach uses the associated network itself as the "control 10 layer" and therefore does not require extra information to be built and maintained. It works best with structures like power line grids etc., where the changes in the base cadastre are relatively uniform and there are no sudden "jumps" between the old and new cadastre. However, it does require considerable computing resources during the processing and some special expertise in the setting up and formulation of the adjustment process. 15 The rubber sheeting systems can provide a general solution to most data types provided that there is adequate control, but they tend to distort the visual appearance of networks such as power lines etc. Additionally this technique will tend to improve positional accuracy near to each control point but overall they do nothing to correct the inaccuracies 20 in the base data, which is exacerbated due to the fact that only a few control points may be easily defined. However, to attempt a shift from existing typically less accurate cadastral data sets based on old survey data to more recent cadastres determined using more accurate technique such 25 as GPS is hampered by the above mentioned drawbacks. In particular, these techniques are unable to handle the size and types of changes likely to occur in a practical situation. Examples of the types and sizes of changes required are as follows: * In urban areas there are corrections of up to eight metres, and in the rural areas the corrections are often considerably larger. 30 e There are big variations in accuracy throughout existing cadastral data sets, This is particularly true where a new subdivision has been inserted in its "correct" position and the cadastre stretched to fit to its external boundaries. When there is a patchwork of WO 2004/097339 PCT/AU20041000452 -6 new subdivisions distributed throughout an older area, this problem is also very evident. * There are physical errors in the data sets such as where pathways are missing at the end of a cul-de-sac or an allotment is in the wrong position relative to its neighbour. These 5 errors probably occurred when the original cadastral maps were drawn and were carried through into the current cadastre when these maps were digitised. These factors pointed to the need to have as many survey control points as possible to identify differences for it is simply not good enough or rigorous enough to have a limited 10 number of control points per data set and rely on some interpolation or modelling procedure. If a large number of survey control points are required per data set, then this precludes the least square technique and complicates the measurement based adjustment (MBA) approach. The "rubber sheeting" systems would work well, but the overheads in working within a triangular mesh would be very significant in any large data set. 15 Accordingly, an improved solution to this problem is required. An additional problem with existing systems is that once parcel networks have been created, they must be stored in a manner which allows the information to be used and 20 updated as required. This is usually achieved by storing the underlying cadastral data in a cadastral database. Typically however cadastral databases can be very large and are subject to continual change as well as having a high level of enquiries. For example NSW has about four 25 million parcels made up of about 25 million lines. Each day between 40 and 50 new plans are lodged to create 150 to 200 new titles. There are also over 1000 search requests for data from the register. Accordingly, a large amount of data that must be considered when the database is queried. 30 As a result, the examination process for each new parcel can take from one hour to several days depending on the complexity of the data and many of the processes are quite intensive WO 2004/097339 PCT/AU2004/000452 -7 mathematically. This poses a problem in the database because if they are carried out directly, the response time for the users can become unacceptable. Summary of the Present Invention 5 In a first broad form the present invention provides a method for determining a transformation for transforming a feature set defined with respect to a first network to a second network, the method including: a) Comparing the first network to the second network; b) Defining a number of pairs of common points, each pair of common points 10 including corresponding points in the first and second networks; and, c) Determining a transformation for transforming the first network to the second network in accordance with the determined common points. Typically the first and second networks are formed from respective parcel networks, the 15 method including: a) Selecting a feature in the first network; b) Determining a corresponding feature in the second network; and, c) Defining the common points in accordance with the determined corresponding features. 20 Each feature can have a unique identifier, in which case the method can include: a) Comparing the identifiers of features in the first and second networks; and, b) Determining the corresponding features in accordance with the results of the comparison. 25 The method may include: a) Comparing the shape of features in the first and second networks; and, b) Determining the corresponding features in accordance with the results of the comparison. 30 The method of defining the common points may include: a) Selecting a point in the first network; and, WO 2004/097339 PCT/AU2004/000452 -8 b) Selecting a corresponding point in the second network. The feature is typically a parcel, such as a land boundary parcel. However, any features common to the reference and feature network, such as buildings, telegraph poles, or other 5 features can be used. In any event, the identifier can be a parcel centroid, or other identifier. In the case of the features being parcels, the points can be located on a parcel boundary, 10 and are typically the parcel boundary vertices. The first and second networks may include a number of points, the method including determining common points for a predetermined number of the points. 15 The method can include: a) Considering the points in the first and second networks; b) Rejecting selected ones of the points in accordance with predetermined criteria; and, c) Determining common points for a predetermined number of the remaining points. 20 The criteria may include that the points form part of a curved parcel boundary. The predetermined number of points can include at least 10% of the number of points, at least 50% of the number of points, or substantially all of the points. 25 The method may further include: a) Determining a displacement for each pair of common points, the displacement representing the difference in position of the common points within the pair; b) Modifying at least one of the pairs of common points in accordance with the 30 determined displacement; and, c) Determining the transformation in accordance with the modified common points.

WO 2004/097339 PCT/AU2004/000452 -9 The method typically includes: a) Generating a contour map in accordance with the displacements in a first direction; b) Determining at least one incorrectly defined pair of common points in accordance with the contour map; and, 5 c) Correcting any incorrectly defined pairs of common points by selecting a new point in the first or second networks. The method may also include: a) Generating a contour map in accordance with the displacements in a second 10 direction, the second direction being orthogonal to the first; b) Determining at least one incorrectly defined pair of common points in accordance with the contour map; and, c) Correcting any incorrectly defined pairs of common points by selecting a new point in the first or second networks. 15 The method can include determining an incorrectly defined pair of common points by locating a predetermined structure in the contour map. The predetermined effect may be a whirlpool effect surrounding the incorrect common 20 points. The transformation is typically in the form of a transformation grid. The method may include determining the transformation in accordance with the 25 displacements, which can be displacement vectors. The method may include-. a) Generating a modified first network by applying the transformation to the first network; 30 b) Comparing the modified first network to the second network; and, c) Determining the success of the transformation in accordance with the results of the comparison.

WO 2004/097339 PCT/AU2004/000452 -10 The method can further include determining a new transformation if the transformation is deemed to be unsuccessful. 5 The method can be performed using a processing system having a store and a processor, in which case the method typically includes causing the processor to: a) Obtain feature data representing the feature set; b) Compare the first network to the second network; c) Define a number of pairs of common points, each pair of common points including 10 corresponding points in the first and second networks; d) Determine a transformation for transforming the first network to the second network in accordance with the determined common points; and, e) Apply the determined transformation to each network in the feature set, to thereby transform the feature set to the second network. 15 The processing system can include a display and an input for receiving input commands for a user, the method including causing the processor to: a) Generate a contour map; b) Display the contour map to the user on the display; and, 20 c) In accordance with input commands from the user: i) Determine at least one incorrectly defined pair of common points ; and, ii) Correct any incorrectly defined pairs of common points by selecting a new point in the first or second networks. 25 The feature set may be a cadastral data set or some other fundamental data set for GIS. In this case, the first network usually represents an existing cadastral base layer, or other layer, in the GIS, with the second network being a reference layer of new, changed or improved cadastre. 30 The parcel may be a land boundary parcel, or may be any other feature in the respective network.

WO 2004/097339 PCT/AU2004/000452 - 11 In a second broad form the present invention provides a method of transforming a feature data set defined with respect- to a first network to a reference data set defined with respect to a second network, the method including: a) Determining a transformation in accordance with the method of the first broad form 5 of the invention; and, b) Applying the transformation to each of the networks in the feature set. In a third broad form the present invention provides a computer program product for determining a transformation for transforming a feature set defined with respect to a first 10 network to a reference data set defined with respect to a second network, the computer program product including computer executable code which when executed in a suitable processing system causes the processing system to perform the method of first broad form of the present invention. 15 In a fourth broad form the present invention provides apparatus for determining a transformation for transforming a feature set defined with respect to a first network to a reference data set defined with respect to a second network, the apparatus including a processing system adapted to: a) Compare the first network to the second network; 20 b) Define a number of pairs of common points, each pair of common points including corresponding points in the first and second networks; c) Determine a transformation for transforming the first network to the second network in accordance with the determined common points; and, d) Apply the determined transformation to each network in the feature set, to thereby 25 transform the feature set to the second network. Typically the processing system includes: a) A store for storing: i) Feature data representing the feature set; and, 30 ii) Reference data representing the second network, iii) Transition data representing the transformation data between the feature network and the reference network, and WO 2004/097339 PCT/AU20041000452 -12 b) A processor adapted to: i) Perform the method of the first broad form of the invention; and, ii) Store transition data representing the transformation in the store. . 5 In a fifth broad form the present invention provides a method of storing cadastral data in a database defining a parcel network, the method including: a) Receiving cadastral data defining a packet having a number of parcels; b) Comparing the packet to predetermined criteria; c) Modifying the parcels in accordance with the results of the comparison; and 10 d) Adding the data representing the modified parcels to the database. The predetermined criteria typically include the coordinate system of the parcel network, the method including: a) Comparing the coordinate system of the received packet to the coordinate system 15 of the parcel network; and, b) Transforming the coordinate system of the packet in response to an unsuccessful comparison. The predetermined criteria may also include at least one of: 20 a) The closure accuracy of each parcel; b) The stated area of each parcel compared to the computed area of the parcel; c) The precision with which each corner fits with the corresponding corner in the parcel network; and, d) A "least squares adjustment" to determine the effect of adding the new parcels on 25 the parcel network. The method typically includes: a) Receiving cadastral data defining a packet having a number of parcels; b) Generating a representation of the parcels; 30 c) Joining the parcels to form a new parcel network; and, d) Adding the new parcel network to the parcel network.

WO 2004/097339 PCT/AU2004/000452 - 13 The method is preferably performed using a processing system coupled to the database. In this case, the method can include modifying the parcels in accordance with the first broad form of the invention. 5 In a sixth broad form the present invention provides apparatus for storing cadastral data in a database defining a parcel network, the apparatus including a processing system coupled to the database, the processing system including a processor adapted to: a) Receive the cadastral data defining a packet having a number of parcels; 10 b) Compare the packet to predetermined criteria; c) Modify the parcels in accordance with the results of the comparison; and d) Add the data representing the modified parcels to the database. The processing system is typically adapted to perform the method of the fifth broad form 15 of the invention. The processor is typically adapted to modify the cadastral data in accordance with input commands received from the user. 20 In a seventh broad form the present invention provides a method of updating cadastral data in a database defining a parcel network, the method including: a) Retrieving cadastral data defining a packet having a number of parcels; b) Altering the parcels within the packet; c) Comparing the modified packet to predetermined criteria; 25 d) Further modifying the parcels in accordance with the results of the comparison; and e) Updating the database in accordance with the modified parcels. The predetermined criteria typically including provision of details of the alterations, including information for identifying: 30 a) The parcels altered; b) The date and nature of the alterations; c) The purpose of the alteration; WO 2004/097339 PCT/AU2004/000452 - 14 d) The individual making the alteration; and, The predetermined criteria typically include at least one of: a) The closure accuracy of each parcel; 5 b) The stated area of each parcel compared to the computed area of the parcel; c) The precision with which each corner fits with the corresponding corner in the parcel network; d) A "least squares adjustment" to determine the effect of adding the new parcels on the parcel network. 10 e) The presence of database id numbers which match those in the database; and, f) That edge points of the packet are not allowed to change in co-ordinate position. The method may also include: a) Selecting a number of parcels defining a packet; 15 b) Creating a copy of the cadastral data defining the packet to thereby retrieve the cadastral data; and, c) Locking the cadastral data in the database corresponding to the retrieved packet, to thereby prevent further modification of the packet. 20 The method may also further include updating the database in accordance with the modified parcels by: a) Creating a copy of the cadastral data representing the unaltered parcels; and, b) Adding the cadastral data representing the altered parcels to the database. 25 The method is typically performed using a processing system coupled to the database. In a eighth broad form the present invention provides apparatus for updating cadastral data in a database defining a parcel network, the apparatus including a processing system coupled to the database, the processing system including a processor adapted to: 30 a) Retrieve cadastral data defining a packet having a number of parcels; b) Alter the parcels within the packet; c) Compare the modified packet to predetermined criteria; WO 2004/097339 PCT/AU2004/000452 - 15 d) Further modify the parcels in accordance with the results of the comparison; and e) Update the database in accordance with the modified parcels. The processing system is typically adapted to perform the method of the seventh broad 5 form of the invention. The processor being is typical adapted to modify the cadastral data in accordance with input commands received from the user. 10 Brief Description of the Drawings An example of the present invention will now be described with reference to the accompanying drawings, in which: Figure 1 is a flow chart of the process of the present invention; 15 Figure 2 is a schematic diagram of an example of a system for implementing the present invention; Figure 3A and 3B are a flow chart of the process implemented by the system of Figure 1; Figure 4 is schematic diagram of an example of a reference network and a target network as presented to a user of the processing system of Figure 2; 20 Figure 5 is schematic diagram of a close up of a portion of Figure 4; Figure 6 is schematic diagram of an example of a contour map as presented to a user of the processing system of Figure 2; Figure 7 is schematic diagram of a close up of a portion of Figure 6; Figure 8 is schematic diagram of an example of the reassignment of one of the common 25 points; Figure 9 is schematic diagram of the contour map of Figure 7 as revised following the reassignment of the common point; Figure 10 is schematic diagram of the contour map of Figure 6 as revised following the reassignment of the common point; 30 Figure 11 is schematic diagram of the corrected target network; Figure 12 is schematic diagram showing a corrected target network of Figure 11 compared to the reference network; WO 2004/097339 PCT/AU2004/000452 - 16 Figure 13 is a schematic diagram of a second example of a system for implementing the present invention; Figure 14 is a schematic diagram of an example of one of the end stations of Figure 13; Figure 15 is a flow chart outlining the process of storing new parcels in a cadastral 5 database; Figure 16 is a flow chart of one example of a method of storing new parcels in a cadastral database; and, Figure 17 is a flow chart of one example of a method of altering parcels in a cadastral database. 10 Detailed Description of the Preferred Embodiments Feature Set Transformation An example of the process of determining a transformation to map layers from an existing feature data set to a new reference network will now be described with reference to Figure 15 1. In particular, as shown in Figure 1, at step 100 the existing cadastral data set to be updated is obtained, with the new reference network, which typically forms a reference cadastre, being obtained at step 110. 20 The feature data may be received in any one of a number of forms and this will typically depend on the data source. Thus, for example, the cadastral data set may be a data set belonging to an entity which has previously performed their own surveys on local datum and which now wishes to map their existing surveys to a reference network, such as a 25 "world based" coordinate system. This may be performed for example to transfer an existing cadastre to a GPS generated reference network as will be appreciated by a person skilled in the art. Thus, for example, a geodetic datum is a defined spheroid used as the computing surface 30 for geodetic calculations. Typically it is an ellipse in rotation fitted to a number of points around the country set at a particular height. The AGD is a spheroid fitted to a number of tide gauges which defined mean sea level around Australia in 1966. Its semi major axis is WO 2004/097339 PCT/AU2004/000452 - 17 6378160 metres and flattening is 1/298.25. In contrast, the GDA is very close to a world based datum (WGS84). Accordingly, as GPS technology develops, it is desirable to be able to change coordinate systems from a local datum such as AGD to a world based datum GDA. 5 In the case of the reference network, this may again be received from an external source, or be predetermined. At step 120 it is necessary to compare common points in the existing feature data set to the 10 new reference network. This may be achieved in a number of ways depending on the implementation of the invention and the nature of the feature data and the new reference network. Typically however the reference network is formed from a parcel network having parcels relating to land ownership, or the like. In this case, the determination of common points will be achieved by matching parcels within a layer of the cadastral data 15 set having an equivalent network (hereinafter referred to as the "target" network) with corresponding parcels in the new reference network, as will be described in more detail below. Once common points have been determined in the base layer and the new reference 20 network, a transformation is generated. When applied to the target network, the transformation, which is typically in the form of a correction grid, will map points in the target network to corresponding points in the new reference network. Accordingly, it will be appreciated that this allows the transformation to be applied to the 25 target network at step 140, thereby mapping the target network onto the reference network. This can be used to make sure that the transformation is acceptable. In addition to this, the transformation can also be applied to all remaining layers in the feature data set at step 150, thereby allowing the entire feature data set to be mapped to the new reference network. As these final stages may be performed separately, or by a different entity, these 30 steps are shown in dotted lines. A person skilled in the art will appreciate that whilst the above outlined procedure may be WO 2004/097339 PCT/AU2004/000452 - 18 performed manually, typically it will require a substantial degree of processing that will be beyond the scope of an individual. Accordingly, this process is typically performed utilising a processing system that is adapted to receive the feature data set and generate the transformation as required. An example of apparatus suitable for performing this 5 procedure will now be described with reference to Figure 2. In particular, in this example, the apparatus is formed from a processing system 10 having a processor 20, a memory 21, an input/output (I/O) device 22 and an optional external interface 23 coupled together via a bus 24 as shown. 10 The feature data set may be provided to the processing system either via the input device 22 or the external interface 23, and the manner in which this is achieved will depend on the nature of the feature data set. Thus, for example, the cadastral data set will typically have a high data volume, and may therefore be supplied directly via the external interface from 15 another processing system, or the like. In any event, once the feature data set has been received, the processing system 10 is adapted to execute appropriate applications software stored in the memory 21, to allow the transformation to be determined. The processing system is also adapted to apply the 20 transformation to the feature data set to thereby allow the layers therein to be defined with reference to the new reference network. Accordingly, it will be appreciated that the processing system may be any form of processing system suitably programmed to perform the analysis. The processing system 25 may therefore be a suitably programmed computer, lap-top, palm computer, or the like. Alternatively, specialised hardware or the like may be used. This process will be described in more detail with reference to Figures 3A and 3B. 30 In particular, the processing system 10 obtains the existing feature data set to be updated at step 200, with the new reference network being obtained at step 210. Again, the manner in which this is achieved will depend on the respective implementation of the invention and WO 2004/097339 PCT/AU2004/000452 - 19 the nature of the feature data set and the reference cadastre. In particular, the feature data set may be received from an external source via the external interface 23. The reference network may be obtained in a similar manner, or may be pre-stored in the memory 21. 5 In any event, at this point the processing system will operate to store data in the memory 21 representing both the reference network and the feature data set. At step 220 the processing system selects a target network from the layers in the existing feature data set. As mentioned this target network will need to correspond to the new 10 reference network, and in particular will need to include corresponding information thereon. Thus, for example, if the new reference network includes an indication of the location of telegraph poles, it will be necessary to select the target network to be a layer which also includes information regarding the location of telephone poles. 15 The reason for this is that during processing the target network and the reference network are typically defined as a number of polygons, with polygons in the target network being mapped to the polygons in the base cadastre. This is therefore typically achieved by generating a parcel network as will be appreciated by a person skilled in the art. 20 Thus, if the reference network includes information relating to land boundaries then it will be necessary to select a target network including land boundaries from the existing feature data set to ensure that similar polygons are present in the target network and the reference network. Otherwise, if a different layer from the existing feature data set were used, such as a layer relating to road positions, the layer would include different polygon structures in 25 the target network to those in the reference network, thereby preventing a successful transformation being performed. Accordingly, whilst the parcels may be based on any form of information however, it is essential that the parcels are based on the same information. In this example, as most 30 layers in a cadastral data set are defined with respect to a layer containing information regarding land boundaries, with the parcels themselves representing the land boundaries, the remainder of the example will focus on the use of parcel networks representing land WO 2004/097339 PCT/AU2004/000452 - 20 boundaries. At this stage, the processing system 10 will store details of points in the target and reference networks in the form of point data in the memory 21. These points generally 5 form nodes within the network, such as boundary vertices, or the like, as will be appreciated by persons skilled in the art. At step 230 having determined the target network, the processing system 10 matches parcels in the target network and in the new reference network. This is carried out on a 10 polygon by polygon basis such that each polygon in the current system is mapped to a corresponding polygon in the target system by performing an appropriate search. This may be performed in a number of ways. Thus for example, each polygon in the reference network and the target network may be 15 provided with a unique identifier that will allow the polygons to be mapped to each other directly. This may be achieved for example in accordance with land ownership, plot number, or any other suitable identifiers which allow corresponding polygons to be identified. 20 If suitable identifiers do not exist, or following preliminary matching using the identifiers, the processing system 10 then determines or confirms the matches by examining the shape, area and position of the polygons. In particular, this is usually achieved by having the processing system 10 examine polygons at a similar locations in the target and reference networks, and then comparing the area and centroid position of these polygons until a 25 likely match is found. It will be appreciated that whilst differences will exist between the reference network and the target network, mainly due to differences in the survey techniques used to create the target network and the reference network, these differences will be fairly uniform. This is 30 highlighted with respect to Figure 4, which shows an example of a graphical user interface (GUI) presented to a user by the processing system 10. In this example, the GUI shows a reference network, represented by the boundaries R, and a target network represented by WO 2004/097339 PCT/AU2004/000452 -21 the boundaries T. Thus, as shown, the differences in the shape and position of the polygons are relatively minor, allowing corresponding polygons to be matched relatively easily. 5 When performing this operation, some polygon pairs may have a different size and shape due to differences in the survey process or parcel definition. A typical example of this is when the parcels include a curved boundary, as shown in Figure 5, which is a close up view of the region 40 in Figure 4. 10 The reason for this is that curved boundaries have up until recently been considered as a number of short line segments. Thus, as shown the curved boundary on the target network will typically consist of a number of straight lines 41, whereas the corresponding boundary in the reference network will include a curved line 42. Whilst it is still possible for the processing system 10 to match the polygons in the reference network to the polygons in the 15 target network, this will have implications in the subsequent processing steps. Accordingly, at step 240, the processing system 10 is adapted to perform a search of the network for any line patterns of this form and then eliminate these from the subsequent processing steps. This will typically be achieved by removing the corresponding points 20 from the point data stored in the memory 21. At this point a search for line boundaries satisfying other criteria may also be eliminated as required. This will typically include irregular boundaries, such as those encountered at creeks, or the like. 25 Having performed this, the processing system 10 operates to select pairs of common points in the reference network and the network layer at step 250. An example of this is shown in Figure 5 by the points TI, T2, T3, T4 on the target network T and the corresponding points R1, R2, R3, R4 on the reference network R. 30 At step 260 the processing system 10 determines displacement factors between the common points in each pair. In particular, the processing system 10 will determine WO 2004/097339 PCT/AU2004/000452 - 22 displacements in orthogonal directions typically along X and Y axes, and then stores the displacements as a displacement vector in the memory 21. The displacement vectors can be stored as attributes of the points in the reference or target network, and can therefore be associated with the point data, as will be appreciated by a person skilled in the art. 5 As step 270 the processing system 10 generates a contour map representing displacements between corresponding points. An example of a contour map is shown in Figure 6, with a close up of the region 43 being shown in Figure 7, with the contours shown at 45. 10 In particular, a respective contour map may be generated for displacements in the X direction and for displacements in the Y direction so that one or two respective contour maps may be shown. In general, as the displacements between the common points should vary fairly uniformly between the target network and the reference network, the contours should be reasonably uniform in distribution across the contour map. Where significant 15 differences exist between the locations of the corresponding points in a respective pair of common points, this will be represented by a steep gradient and hence a large number of contours on the contour map. In particular, pairs of common points that have been incorrectly defined will be highlighted 20 by whirlpool areas of close contours on the contour map. An example of this is shown at area 43 in Figure 6. A close up of this is shown in Figure 7. In particular, as shown in Figure 7 the point A on the reference network has been mapped to the point B on the target network. However, in actual fact point A should have been mapped to the point C and accordingly, a contour whirlpool exists around the point B indicating that point B is 25 incorrectly assigned. Accordingly at step 280 the user modifies corresponding points in accordance with the contour map. Thus as shown in Figure 8 the user alters the pair of common points so that the common points are now points A and C as required. This is typically achieved by 30 having the user select the incorrect common point on the contour map presented on the GUI using an input device such as a mouse, and then select a new common point in the same way. Alternatively the user may simply enter new coordinates at which the common WO 2004/097339 PCT/AU2004/000452 - 23 point should occur. In any event, at step 290 the processing system 10 will operate to recalculate the displacement vector, which will replace the existing common point displacement vector 5 stored in the memory 21 for the respective common points. The processing system can then regenerate the contour map for checking. It will be appreciated that a number of incorrectly defined common points may b corrected before the contour map is redrawn. 10 An example of the regenerated contour map is shown in Figure 10, with a close up of the corrected common points in the region 44 shown in Figure 9. As shown the contours in the corrected region include far smoother contour lines without the whirlpool effect indicating that the common point selection for this respective pair of common points is 15 now successful. Once this has been completed the processing system 10 determines a correction grid in accordance with the selected displacement factors. In general the correction grid is fonned using a triangular network having vertices located at the common points, with the 20 deformation of the triangular network under the determined displacements setting the transformation that will be required for points falling within the triangular area. This correction grid can then be applied to the target network, as shown for example in Figure 11. 25 The success of the correction grid can be checked by applying the correction grid to the target network and then superimposing the corrected target network on the reference network as shown in Figure 12. In this example, the original target network is indicated by the boundaries T, with the modified target network represented by the boundaries M. As 30 shown, the modified target network M is substantially identical to the reference network. Assuming that the correction grid is acceptable, the correction grid can be stored in the WO 2004/097339 PCT/AU20041000452 - 24 memory 21 and then subsequently applied to remaining layers in the GIS related to the existing feature data set to map these to the new base cadastre as will be appreciated by a person skilled in the art. Alternatively, the correction grid may be output to an external processing system, for example via the external interface 23, to allow the correction of the 5 cadastral data set to be performed remotely. From this it will be appreciated that when the feature data set is received at step 200, it is only necessary to receive the layer containing the target network. 10 A number of additional features that can be implemented in the present invention will now be described with reference to a second example of the present invention. In this example the process operates to provide a means to warp an existing network of polygons to fit with a new network of the same polygons which have been spatially 15 displaced in various amounts according to their position in the network. Again, the new network is referred to as the reference network, with the old network being referred to as the target network. The processing system 10 receives a data set representing the target network and stores this 20 in the memory 21. This may be achieved in a number of ways, but is typically achieved by storing a target point having a target coordinate value for each node or point in the network. Similarly, reference coordinate values are stored for the nodes or points in the reference network. 25 The processing system 10, under guidance of the operator follows the following stages: * Identify those points in the network which are common to both systems. * Compute the local displacement in X and Y at each common point. * Display the displacements as isolines (contours) and vectors. * Edit the data by selecting or de-selecting of common points. 30 * Compute a "correction grid" from the displacements. o Apply the correction grid on the target network to verify the quality of the correction grid.

WO 2004/097339 PCT/AU2004/000452 -25 Apply the correction grid to all data sets associated to the target network to bring them into association to the reference network. This is carried out on a polygon basis. For each polygon in the reference system the 5 processing system 10 searches the polygons in the target layer to find a corresponding polygon in the reference cadastre. If there is a unique identifier for each polygon, then this can be used to select likely polygon pairs. If not, likely pairs can be found by looking for those which have a similar area and whose centroid positions are close to each other. 10 Polygon pairs may have a different number of sides and be slightly different in shape, however there will be some line segments which will correspond very closely and the points on the ends of these line segments can be considered as "common points". If either the reference network or the target network has been generated by digitising from 15 a plan, the curved boundaries and irregular boundaries (creeks, rivers etc) will often be represented as a series of short segments of similar length. A search is made for this type of line pattern and these lines are excluded from the process. For the residual lines, the processing system 10 selects the longest line in the reference 20 network polygon, then computes the vector from its polygon centroid to the mid point of the selected line. On the target polygon in the target layer, the processing system 10 finds a line of similar length with a similar vector from its centroid. These lines can then be considered as common lines. The processing system 10 checks the 25 direction of the lines as they are in their polygon to determine whether each polygon line sequence is in either a clockwise or anti clockwise direction, and then computes the transformation parameters to rotate, scale and shift the end points on the target line to match the end points of the reference line. 30 Then, for each point in the reference polygon (excluding points on arcs and irregular lines) the processing system 10 finds a corresponding point in the target polygon to form a pair of common points. In order to achieve this the processing system 10 takes the coordinate WO 2004/097339 PCT/AU2004/000452 - 26 for each point in the target polygon (excluding points on arcs and irregular lines) is transformed using the transformation parameters computed form the initial reference line, and a number of closest points are selected as the potential corresponding points. 5 The processing system 10 then tests the points more rigorously by performing the following steps: * Computing transformation parameters (rotation, scale, shift in X and shift in Y) between the coordinates of the target points and the reference point coordinates using a least squares procedure. 10 * Using the parameters to transform each target point and calculate the residuals in X and Y between the transformed values and the coordinates of the corresponding reference point. * Computing the mean and standard deviation of the set. e If the standard deviation exceeds a given value: 15 * Excluding the target point with the biggest residual and re-computing the transformation parameters and the mean and standard deviation of the remainder of the set. e Transforming the excluded point and computing its residuals and if these are greater then three times the standard deviation, then the target point is discarded, 20 otherwise it should be included. * If a target point has been discarded, repeat the whole sequence with the remaining target points and continue to do this until either no point is discarded or the number of points is equal to three. * If only three points are remaining and the standard deviation of the set is large, reject 25 all three and determine that a corresponding point cannot be established for the respective reference point. " If more than three points remain or if the standard deviation of the remaining three is within the tolerance expected for the data set, then those points in the polygon are accepted as being "common points". 30 When a corresponding target point is found, the target point and reference point form a pair of common points. The processing system 10 then computes target coordinate value WO 2004/097339 PCT/AU2004/000452 -27 and the reference coordinate value. These values are stored as attributes of the reference point in the memory 21. The processing system 10 generates a triangular network (TIN) for the common points and 5 this is used to generate a contour map for either the displacements in X or displacements in Y between the common points. The displacement vector at each common point may also displayed to the user at this time. This allows the user to review the area of interest to examine the selected common points. 10 By using a 200mm contour interval, the overall trends are clearly evident and any unsuitable points cause a "whirlpool" of contours around them. The TIN can then be edited by adding or excluding points matched points. A point is excluded by selecting it with a mouse and a point is included by selecting the point pair with a mouse. After each editing sequence the TIN and contours are recomputed and the results displayed. 15 Editing ceases when the operator considers that the matched points correctly show the warping needed to bring the target network to correspond with the reference network. The processing system then generates a "correction grid" from the TIN. 20 The correction grid is formed as a table of corrections with the intersection of each row and column corresponding to a location on the surface of the earth. The value for each point on the correction grid is interpolated from the three values at the apex points of the surrounding triangle in the TIN. 25 Accordingly, this allows the correction grid can be applied to any layer in a processing system which is associated with the target network, as will be appreciated by persons skilled in the art. It will be appreciated that the general techniques outlined in the document may be applied 30 to any target and reference networks, and these do not have to be formed from cadastral data.

WO 2004/097339 PCT/AU2004/000452 -28 Furthermore, whilst the above description has focused on the comparison of parcels, the techniques can be used for any features common in the feature data set and the reference network. 5 Thus, for example, if the reference network includes an indication of the location of telegraph poles, then the position of telegraph poles in the reference network and the feature data set can be compared in a manner similar to that defined above with respect to the parcels. In this case, it will be appreciated that the location of each telegraph pole will be a point, with the common points being selected to be identical telegraph poles. These 10 will be determined in a manner similar to that described above. However, any suitable feature may be used, such as buildings, or the like. Furthermore, whilst the above description has focused on the comparison of parcels, the 15 techniques can be used for any features common in the feature data set and the reference network. Thus, for example, if the reference network includes an indication of the location of telegraph poles, then the position of telegraph poles in the reference network and the 20 feature data set can be compared in a manner similar to that defined above with respect to the parcels. In this case, it will be appreciated that the location of each telegraph pole will be a point, with the common points being selected to be identical telegraph poles. These will be determined in a manner similar to that described above. 25 However, any suitable feature may be used, such as buildings, or the like. Architectures It will be appreciated that the above method may be achieved in a number of different manners. Thus, for example, a respective processing system 10 may be provided for each 30 individual that is to use the system. This could be achieved by supplying respective applications software for an individual's computer system, or the like, for example on a transportable media, or via download.

WO 2004/097339 PCT/AU2004/000452 - 29 In this case, if modifications to the processing techniques are determined, these could be made available through program updates or the like, which again may be made available in a number of manners. However, alternative architectures, such as distributed architectures, 5 or the like, may also be implemented. An example of this is shown in Figure 13 in which the processing system 10 is coupled to a database 11, provided at a base station 1. The base station 1 is coupled to a number of end stations 3 via a communications network 2, such as the Internet, and/or via 10 communications networks 4, such as local area networks (LANs) 4. Thus it will be appreciated that the LANs 4 may form an internal network at a company, medical institution, or the like, depending on the nature of the system and the types of sequences being distinguished. 15 In use the end stations 3 must be adapted to communicate with the processing system 10 positioned at the base station 1. It will be appreciated that this allows a number of different forms of end station 3 may be used. An example of a suitable end station is shown in Figure 14. As shown the end station 3 20 includes a processor 30, a memory 31 and an input/output (110) device 32 such as a keyboard and monitor and an external interface 33, coupled together via a bus 34, as shown. In use, the internal interface 35 is provided to allow the end station to be coupled to one of the communications networks 2, 4. 25 In use, the processor 30 is adapted to communicate with the processing system 10 provided in the base station 1 via the communications networks 2, 4 to allow the above described process to be implemented. Accordingly, it will be appreciated that if the communications network 2 is the Internet, this will typically be achieved by having the base station 1 present web pages to users of the end stations 3. 30 Accordingly, it will be appreciated that the end stations 3 may be formed from any suitable processing system, such as a suitably programmed PC, Internet terminal, lap-top, hand- WO 2004/097339 PCT/AU2004/000452 - 30 held PC, or the like, which is typically operating applications software to enable data transfer and in some cases web-browsing. In this case, the feature data set to be updated may be supplied to the processing system 10 5 via the end station 3, allowing the processing system 10 to determine the transformation or correction grid as described above. The correction grid can then be applied to the feature data set, or be transferred to the end station 3 for subsequent use. In any event, it will be appreciated that access to the process may be controlled using a 10 subscription system or the like, which requires the payment of a fee to access a web site hosting the process. This may be achieved using a password system or the like, as will be appreciated by persons skilled in the art. Cadastral Database 15 An example of the manner in which parcel data may be stored in a database and subsequently modified will now be described with reference to Figures 15 to 17. In particular, Figure 15 describes an overview of the process of maintaining a database with a parcel data. In this example, the process is achieved using a processing system 20 similar to the processing system 10 shown in Figure 2, which is coupled to a cadastral database containing cadastral data via the external interface 23. However it will be appreciated by persons skilled in the art that the techniques outlined below could be achieved using any suitable architecture. 25 In any event, in this example, at step 400 a packet representing the parcels to be added to the database are defined. This is typically achieved by either importing the data to the processing system 10 from external data source, such as a separate database, from survey data or the like, or by entering the data manually. Once the parcels have been defined, the parcels are compared to predetermined criteria to determine if they are acceptable for 30 inclusion within the database at step 410.

WO 2004/097339 PCT/AU20041000452 -31 As set out in more detail below, this will typically involve at least checking the coordinate system of the defined parcels to ensure that they can be correctly integrated into database. For example, the coordinate system of the database is typically the world coordinate system based on the coordinates of latitude and longitude. Accordingly, it will be 5 appreciated that as many surveys are performed on the basis of planar Cartesian coordinate systems, transformation of the coordinates is necessary in order for the parcels to correctly map to the database coordinate system. This can be achieved using the processes outlined above. 10 Additional criteria may also be set depending on the circumstances and the respective implementation. Once the parcels have been modified as required at step 420, for example, by transforming the parcels into an alternative coordinate system, the modified parcels are added to the 15 database at step 430. In this regard, it will be appreciated that the database may include any one of the number of schemas and the respective schema will depend and effect the manner in which this is achieved, as will be described in more detail below. In any event, once the cadastral database has been updated, the parcels defined therein can 20 be subsequently modified as required, for example if more accurate survey data is determined, or if changes are made to parcel boundaries. The manner in which this can be achieved will now be outlined. In particular, at step 440 the user can select the parcels to be updated. At step 450 having selected the parcels, the 25 parcels are extracted from the database as a respective packet. The user can then modify the selected parcels as required at step 460 before updating the database in accordance with the modified parcels at step 470. As will be appreciated by persons skilled in the art, the result of this is that the user can 30 operate to work only on the parcels contained in the packet. By extracting data corresponding to these parcels from the database, this ensures that the user is working on WO 2004/097339 PCT/AU2004/000452 - 32 only a small volume of data thereby making the modification operation computationally inexpensive and quick. In general, the modified parcels must also satisfied predetermined criteria, as outlined 5 above with respect to new parcels, to ensure that they can be correctly integrated into the existing parcel network defined in the database. A further feature of this technique, is that the extracted packet can be created as a copy of the cadastral data in the database, such that the original data remains in the database while 10 the modifications are being made. Thus, the parcels contained within the extracted packet are flagged on the database to warn any enquirers that a change is in progress and they are also "locked" to prevent a further change taking place until the current transaction for the packet has been completed. This 15 also allows other users to interact with the parcels within the database without being effected by the changes caused by the modification of the parcels. With this type of operation, there is a single enquiry on the database for the extraction of a packet, and another for the insertion after the update. However it requires the extraction 20 and insertion software to know all about the geometry and the rules for managing that geometry. This process of adding new parcels to a database will now be described in more detail below with respect to Figure 16. In particular at step 500 the user inputs data defining the 25 parcels contained within a packet to be added to the database. This will typically initially consist of inputting boundary data, vertices, coordinates or the like. At step 510 the processing system 10 typically displays a representation of the defined parcels to the user. This allows the user to join the parcels together at step 520 to form a 30 parcel network. Thus, if all of the parcels are already interconnected in the received packet, steps 510 and 520 may not be required. Typically however it is necessary to adjust parcels from different surveys so that the parcels correctly fit together to define a network.

WO 2004/097339 PCT/AU2004/000452 -33 Once this has been completed, the geometry of the parcels are checked to determine if they can be integrated into the parcel network defined by the cadastral database at step 530. Thus, for example, this will include ensuring the coordinate system of the parcel network 5 and the coordinate system of the cadastral data in the database are the same. In the event that the geometry is deemed not to be acceptable at step 540, then the packet is considered to determine if the geometry can be modified at step 550. If not, then it will be appreciated that the parcels defined by the packet cannot be written into the database. 10 However, typically the geometry can be modified and this is therefore performed at step 570, before the check of the geometry is repeated at step 530. It will be appreciated that if the geometry modification required is to update the coordinate system utilised by the parcel network in the packet, then this can be achieved using the 15 techniques and apparatus described above with respect to Figures 1 to 14. Once this has been completed, the parcels of the packet are checked to determine if they will fit within the parcel network defined by the cadastral database at step 580. Thus, for example, this will ensure that the parcels can be joined to existing parcels in the database, 20 and that there is no overlap with existing parcels, or the like. In the event that it is determined that the parcels will not fit at step 590, then the packet is considered to determine if the parcels can be modified at step 600, for example, by the addition of linking parcels, or the like. If not, then it will be appreciated that the parcels 25 defined by the packet cannot be written into the database as set out at step 560. Otherwise, the parcels are modified at step 610, before the fit is rechecked at step 580. Once this has been completed, the effect of adding the parcels in the packet to the parcel network defined by the database is checked at step 620. Thus, for example, this will 30 ensure that addition of the parcels will not result in undue deformation of existing parcels, if modification of the existing parcels is required to allow the existing parcels to correctly join to the parcels in the packet to be added.

WO 2004/097339 PCT/AU2004/000452 -34 In the event that it is determined that the packet will have an undue effect at step 630, then the packet is considered to determine if the parcels can be modified at step 640, for example, by the alteration of parcel boundaries, or the like. If not, then it will be 5 appreciated that the parcels defined by the packet cannot be written into the database as set out at step 660. Otherwise, the parcels are modified at step 650, before the check is repeated at step 620. Finally, assuming this is completed successfully, the processing system 10 updates the 10 database by writing new cadastral data defining the parcels in the packet into the database in accordance with the database schema, at step 660. One example of the database schema is as set out below in Appendix A, which as will be appreciated by persons skilled in the art, ensures that sufficient information is provided for the parcels to be of use in practical situations. 15 The process outlined above therefore operates to ensure that the parcels included in the packet are suitable for inclusion within the database by checking the parcel geometry, as well as the fit with existing parcels. However, it will be appreciated by persons skilled in the art that checks of the data against additional criteria may also be performed in a similar 20 manner. Thus for example, it may be desirable to ensure certain degrees of accuracy for the parcels before they are added into the database. Other checks may include checking: * The closure accuracy of each parcel; * The stated area of each parcel against the computed area of the parcel; * The precision with which each corner fits with the corresponding corner in the existing 25 packet network; * A "least squares adjustment" to determine the effect on the existing boundary network from the inclusion of the new data. However, it will be appreciated that any such other criteria may be checked using similar 30 techniques.

WO 2004/097339 PCT/AU2004/000452 - 35 Thus, in general the processing system 10 will execute applications software which causes specific criteria to be checked automatically or in accordance with manual intervention, before the parcels can be added to the database. 5 The process of allowing the user to manipulate parcels contained within the database will now be described with reference to Figure 17. In particular, whilst users could manipulate parcels directly within the database, this is typically a time consuming process due to the length of time taken to modify a database 10 containing such a large number of records. Accordingly, at step 700 the user selects a region of parcels covered by the database. This may include for example a town, an area within a town such as a suburb, or the like. In any case, at step 710 the processing system 10 typically generates a representation of the 15 selected region. This allows the user to select a number of parcels from the representation at step 720, which may be achieved by having the user highlight the parcels on the representation. It will be appreciated that the selection of parcels may be achieved in a number of manners depending on the database and its implementation and that this is not important for the purposes of this invention. 20 In any event, at step 730 the processing system 10 generates a query based on selected parcels. The query is an SQL query used for extracting cadastral data from the database in the usual way. 25 At step 740 the processing system 10 extracts a packet containing the selected parcels. In particular, the packet contains a copy of the cadastral data for each of the parcels, with the cadastral data of the parcels also being retained within the database. At this point, the parcels in the database are locked so that they may not be simultaneously edited by other third parties also using the database. This is achieved by setting a flag in the cadastral data 30 stored in the database which indicates to other users that the data is currently undergoing modification and which prevents the data being extracted for modification by any other individuals.

WO 2004/097339 PCT/AU20041000452 - 36 This means that other parties can continue to use the database and in particular view the parcels that are currently undergoing modification, thereby allowing the database to be updated whilst in use. In any event, in this case, when a user accesses a parcel that is 5 currently being modified, they will generally be-provided with an indication, such as by having the parcel displayed in a different colour or the like. In any event, at step 750 the user modifies the parcels and the packet at required. The manner in which this is performed will depend on the modifications occurring but will 10 typically revolve modifying one or more parcels, adding parcels, removing parcels, moving parcel boundaries or vertices, or the like. This may be achieved utilising appropriate parcel editing software as will be appreciated by a person skilled in the art. In any event, the user only works on parcels contained within the packet extracted from the 15 database, and this is achieved by having the processing system act directly on the cadastral data contained within the packet. As this includes only a limited volume of data, there is no delay in having the processing system 10 having to access and modify large quantities of data within the database. 20 Once the modification is completed at step 760 the processing system 10 compares the modified packet to predetermined criteria. The predetermined criteria are used to ensure consistency of parcel modification, and this will typically involve checking: * The closure accuracy of each parcel; * The stated area of each parcel against the computed area of the parcel; 25 e The precision with which each corner fits with the corresponding corner in the existing packet network; and, * A "least squares adjustment" may be carried out on the updated packet to determine the effect on the existing boundary network from the inclusion of the new data. 30 The following checks will also typically be done on the data packet compared to the data stored in the database.

WO 2004/097339 PCT/AU2004/000452 - 37 * All Parcels, lines and points with database id numbers must currently exist in the database; * Edge points of the packet are not allowed to change in co-ordinate position * The processing system 10 will check for close points. 5 Additional checks may include checking the coordinate system employed by the modified parcels, as well as checking that information regarding the update is provided. This will include details, such as details of the individual modifying the parcels, the time and date on which the modifications were made, the reasons for the modifications, and the like. This 10 will also typically include the requirement that any modifications to the database are signed off by an examiner and resubmitted back to the database. It will be appreciated that satisfaction of the criteria may be achieved in a manner similar to that described with respect to Figure 16. 15 At step 770 the processing system determines if the criteria are satisfied, and if not, the user is informed that further changes must be made to the packet at step 750. Otherwise, the processing system 10 adds the modified data packets to the database at step 780. In this regard, the processing system 10 also creates the record of the changes made and 20 store a copy of the parcels prior to the amendment thereby allowing the original status of the database to be retrieved should this become necessary. Thus, the system will retain all historical data including point, parcel and plan definitions as parcels are retired. This data will be flagged as 'Historical' and will not show as the 'Current' cadastral network definition. This historical plan/parcel data will be able to be seen using the viewer when 25 enabled, allowing updates to the parcel network to be reviewed at a later date. This allows the status of a parcel network at any date to be reviewed. Accordingly, no cadastral data will be deleted from the database, it will always be archived. When a parcel, line or point is updated with new data, the old copy is not erased, 30 it is stored in the archive data tables.

WO 2004/097339 PCT/AU20041000452 - 38 To achieve this, the database will contain a duplicate set of the primary data tables to hold archive data. These data tables may be used in the event of errors in the data to trace when the data values were changed and by whom. It may also be used, by professional IT staff, to reconstruct a 'snapshot' of the Cadastral database at an instant in time. 5 Accordingly, the above described process allows areas of interest within a database to be extracted from the database as a "packet" of cadastral information. The parcels within the packet are flagged on the database to warn any enquirers that a change is in progress and they are also "locked" to prevent a further change taking place until the current transaction 10 for the packet has been completed. The packet can then be processed, signed off by an examiner and resubmitted back to the database. If it is accepted, the area of interest is updated and the area "unlocked". With this type of operation, there is a single enquiry on the database for the extraction of a 15 packet, and another for the insertion after the update. However it requires the extraction and insertion software to know all about the geometry and the rules for managing that geometry. These techniques are therefore complementary to the system described above for 20 performing feature set transformation, and other cadastral software which can be used to assemble and check cadastral boundary data and maintain the integrity of the data within the database. This also allows "web based" routines to be used for enquiries on the database, which can use SQL type enquiries on the tables to allow the information contained within the database to be accessed. 25 It will be appreciated from this that the system can be implemented using a suitable processing system 10, or end station 3 coupled to the database via a suitable network 2, with the database being implemented as an SQL database. It will be appreciated however that other enquiry and reporting systems can be used. 30 In any event, by utilising "parcel based" as opposed to "line based" database schemas, this allows parcels to be selected and extracted as packets for ease of modification.

WO 2004/097339 PCT/AU2004/000452 - 39 Furthermore, by ensuring that update of the packets are suitably monitored this ensures integrity of the database whilst allowing the database to be updated whilst the database is in use. 5 It will be appreciated that the techniques outlined above may also be applied to maintaining a spatial database of easements, or the like. Persons skilled in the art will appreciate that numerous variations and modifications will become apparent. All such variations and modifications which become apparent to 10 persons skilled in the art, should be considered to fall within the spirit and scope that the invention broadly appearing before described.

WO 2004/097339 PCT/AU2004/000452 - 40 Appendix A An example of the database schema for a cadastral Database is shown below: 5 Plans Plan Name Date of survey Units of Survey, metres, feet, links Surveyor Name 10 Surveyor Company Date Lodged Date approved Bearings - true mid bearings 15 Parcels LAIS code number Plan name Lot number Centroid 20 Type - a numeric parcel type Area - legal area of the parcel as a text field Date and operator for last update An accuracy indicator for coordinates Whether the parcel is historical (superseded) 25 Whether the parcel is in a proclaimed area Lines Start point/co-ordinate End point/co-ordinate 30 Bearing for line, from plan Type of bearing (true mid or plane) Distance of line, from plan Line type, numeric tag WO 2004/097339 PCT/AU2004/000452 - 41 Radius of line if curved Centre point/co-ordinate Line Points 5 Start point/co-ordinate of line point record End point/co-ordinate of line point record Line Point/co-ordinate of each line point Points 10 latitude of point longitude of point height of point code - text field to indicate the type of mark type (corner, control point, reference mark etc.) 15 mark name (if it is also a control point) xy accuracy indicator z accuracy indicator Date and operator for last update 20

Claims

THE CLAIMS DEFINING THE INVENTION ARE AS FOLLOWS:

1) A method determining a transformation for transforming a feature set defined with respect to a first network to a second network, the method including: a) Comparing the first network to the second network; b) Defining a number of pairs of common points, each pair of common points including corresponding points in the first and second networks; and, c) Determining a transformation for transforming the first network to the second network in accordance with the determined common points.

2) A method according to claim 1, the first and second networks being formed from respective parcel networks, the method including: a) Selecting a feature in the first network; b) Determining a corresponding feature in the second network; and, c) Defining the common points in accordance with the determined corresponding features. 3) A method according to claim 2, each feature having a unique identifier, the method including: a) Comparing the identifiers of features in the first and second networks; and, b) Determining the corresponding features in accordance with the results of the comparison. 4) A method according to claim 2 or claim 3, the method including: a) Comparing the shape of features in the first and second networks; and, b) Determining the corresponding features in accordance with the results of the comparison.

5) A method according to any one of the claims 1 to 4, the method of defining the common points including: a) Selecting a point in the first network; and, b) Selecting a corresponding point in the second network.

6) A method according to any one of the claims 2 to 5, the feature being a parcel.

7) A method according to claim 6, the identifier being a parcel centroid. 8) A method according to claim 5 and claim 6, the points being located on a parcel boundary. 9) A method according to claim 8, the points being parcel boundary vertices. 10) A method according to any one of the claims 1 to 9, the first and second networks including a number of points, the method including determining common points for a predetermined number of the points.

11) A method according to any one of the claim 1 to 10, the method including: a) Considering the points in the first and second networks; b) Rejecting selected ones of the points in accordance with predetermined criteria; and, c) Determining common points for a predetermined number of the remaining points.

12) A method according to claim 11, the criteria including that the points form part of a curved parcel boundary.

13) A method according to claim 12, the predetermined number of points including at least 10% of the number of points.

14) A method according to claim 13, the predetermined number of points including at least 50% of the number of points. 15) A method according to claim 13, the predetermined number of points including at substantially all of the points.

16) A method according to any one of the claims 1 to 15, the method further including: a) Determining a displacement for each pair of common points, the displacement representing the difference in position of the common points within the pair; b) Modifying at least one of the pairs of common points in accordance with the determined displacement; and, c) Determining the transformation in accordance with the modified common points.

17) A method according to claim 16, the method including: a) Generating a contour map in accordance with the displacements in a first direction; b) Determining at least one incorrectly defined pair of common points in accordance with the contour map; and, c) Correcting any incorrectly defined pairs of common points by selecting a new point in the first or second networks.

18) A method according to claim 17, the method including: a) Generating a contour map in accordance with the displacements in a second direction, the second direction being orthogonal to the first; b) Determining at least one incorrectly defined pair of common points in accordance with the contour map; and, c) Correcting any incorrectly defined pairs of common points by selecting a new point in the first or second networks. 19) A method according to claim 17 or claim 18, the method including determining an incorrectly defined pair of common points by locating a predetermined structure in the contour map. 20) A method according to claim 19, the predetermined effect being a whirlpool effect surrounding the incorrect common points. 21) A method according to any one of the claims 1 to 20, the transformation being in the form of a transformation grid.

22) A method according to any one of the claims 1 to 21, the method further including determining the transformation in accordance with the displacements.

23) A method according to any one of the claims 1 to 22, the displacements being displacement vectors.

24) A method according to any one of the claims 1 to 23, the method including: a) Generating a modified first network by applying the transformation to the first network; b) Comparing the modified first network to the second network; and, c) Determining the success of the transformation in accordance with the results of the comparison.

25) A method according to claim 24, the method further including determining a new transformation if the transformation is deemed to be unsuccessful.

26) A method according to any one of the claims 1 to 25, the method being performed using a processing system having a store and a processor, the method including causing the processor to: a) Obtain feature data representing the feature set; b) Compare the first network to the second network; c) Define a number of pairs of common points, each pair of common points including corresponding points in the first and second networks; d) Determine a transformation for transforming the first network to the second network in accordance with the determined common points; and, e) Apply the determined transformation to each network in the feature set, to thereby transform the feature set to the second network.

27) A method according to claim 26, the processing system including a display and an input for receiving input commands from a user, the method including causing the processor to: a) Generate a contour map; b) Display the contour map to the user on the display; and, c) In accordance with input commands from the user: i) Determine at least one incorrectly defined pair of common points ; and, ii) Correct any incorrectly defined pairs of common points by selecting a new point in the first or second networks.

28) A method according to any one of the claims 1 to 27, the feature set being a cadastral data set.

29) A method according to claim 28, the first network representing a base layer in the cadastral data set.

30) A method according to claim 28 or claim 29, the second network being a reference layer, or a reference cadastre.

31) A method of transforming a feature set defined with respect to a first network to a reference data set defined with respect to a second network, the method including: a) Determining a transformation in accordance with the method of any one of the claims 1 to 30; and, b) Applying the transformation to each of the networks in the feature set.

32) A computer program product for determining a transformation for transforming a feature set defined with respect to a first network to a second network, the computer program product including computer executable code which when executed on a suitable processing system causes the processing system to perform the method of any one of the claims 1 to 30.

33) Apparatus for determining a transformation for transforming a feature set defined with respect to a first network to a reference data set defined with respect to a second network, the apparatus including a processing system adapted to: a) Compare the first network to the second network; b) Define a number of pairs of common points, each pair of common points including corresponding points in the first and second networks; c) Determine a transformation for transforming the first network to the second network in accordance with the determined common points; and, d) Apply the determined transformation to each network in the feature set, to thereby transform the feature set to the second network.

34) Apparatus according to claim 33, the processing system including: a) A store for storing: i) Feature data representing the feature set; and, ii) Reference data representing the second network, iii) Transition data representing the transformation data between the feature network and the reference network, and b) A processor adapted to: i) Perform the method of any one of the claims 1 to 30, to thereby determine a transformation; and, ii) Store transition data representing the transformation in the store.

35) A method of storing cadastral data in a database defining a parcel network, the method including: a) Receiving cadastral data defining a packet having a number of parcels; b) Comparing the packet to predetermined criteria; c) Modifying the parcels in accordance with the results of the comparison; and d) Adding the data representing the modified parcels to the database.

36) A method according to claim 35, the predetermined criteria including the coordinate system of the parcel network, the method including: a) Comparing the coordinate system of the received packet to the coordinate system of the parcel network; and, b) Transforming the coordinate system of the packet in response to an unsuccessful comparison.

37) A method according to claim 35 or claim 36, the predetermined criteria including at least one of: a) The closure accuracy of each parcel; b) The stated area of each parcel compared to the computed area of the parcel; c) The precision with which each comer fits with the corresponding comer in the parcel network; and, d) A "least squares adjustment" to determine the effect of adding the new parcels on the parcel network. 38) A method according to any one of the claims 35 to 37, the method including: a) Receiving cadastral data defining a packet having a number of parcels; b) Generating a representation of the parcels; c) Joining the parcels to form a new parcel network; and, d) Adding the new parcel network to the parcel network. 39) A method according to any one of the claims 35 to 38, the method being performed using a processing system coupled to the database.

40) A method according to any one of the claims 35 to 39, the method including modifying the parcels in accordance with the method of any of the claims 1 to 31.

41) Apparatus for storing cadastral data in a database defining a parcel network, the apparatus including a processing system coupled to the database, the processing system including a processor adapted to: a) Receive the cadastral data defining a packet having a number of parcels; b) Compare the packet to predetermined criteria; c) Modify the parcels in accordance with the results of the comparison; and d) Add the data representing the modified parcels to the database.

42) Apparatus according to claim 41, the processing system being adapted to perform the method of any one of the claims 35 to 40.

43) Apparatus according to claim 42, the processor being adapted to modify the cadastral data in accordance with input commands received from the user. 44) A method of updating cadastral data in a database defining a parcel network, the method including: a) Retrieving cadastral data defining a packet having a number of parcels; b) Altering the parcels within the packet; c) Comparing the modified packet to predetermined criteria; d) Further modifying the parcels in accordance with the results of the comparison; and e) Updating the database in accordance with the modified parcels. 45) A method according to claim 44, the predetermined criteria including provision of details of the alterations, including information for identifying: a) The parcels altered; b) The date and nature of the alterations; c) The purpose of the alteration; d) The individual making the alteration; and,

46) A method according to claim 44 or claim 45, the predetermined criteria including at least one of: a) The closure accuracy of each parcel; b) The stated area of each parcel compared to the computed area of the parcel; c) The precision with which each comer fits with the corresponding comer in the parcel network; d) A "least squares adjustment" to determine the effect of adding the new parcels on the parcel network. e) The presence of database id numbers which match those in the database; and, f) That edge points of the packet are not allowed to change in co-ordinate position.

47) A method according to any one of the claims 44 to 46, the method including: a) Selecting a number of parcels defining a packet; b) Creating a copy of the cadastral data defining the packet to thereby retrieve the cadastral data; and, c) Locking the cadastral data in the database corresponding to the retrieved packet, to thereby prevent further modification of the packet.

48) A method according to any one of the claims 44 to 47, the method further including updating the database in accordance with the modified parcels by: a) Creating a copy of the cadastral data representing the unaltered parcels; and, b) Adding the cadastral data representing the altered parcels to the database.

49) A method according to any one of the claims 44 to 38, the method being performed using a processing system coupled to the database.

50) Apparatus for updating cadastral data in a database defining a parcel network, the apparatus including a processing system coupled to the database, the processing system including a processor adapted to: a) Retrieve cadastral data defining a packet having a number of parcels; b) Alter the parcels within the packet; c) Compare the modified packet to predetermined criteria; d) Further modify the parcels in accordance with the results of the comparison; and e) Update the database in accordance with the modified parcels.

51) Apparatus according to claim 50, the processing system being adapted to perform the method of any one of the claims 44 to 49.

52) Apparatus according to claim 51, the processor being adapted to modify the cadastral data in accordance with input commands received from the user.