CH712678B1 - Method of displaying cartographic data. - Google Patents

Abstract

The present invention relates to a method for displaying cartographic data, comprising the following steps: -capture using a multimedia electronic device (3) comprising a camera (5) of image data representing a model of a geographical relief; - relief identification, by comparison with at least one geographic relief reference model; -access to additional data; -display on a display device (7) of said device of an image obtained from said image data and said superimposed additional data.

Description

Technical area

The present invention relates to the field of cartographic models, as well as a computer program to provide additional information to observers of such a model.

State of the art

[0002] In the state of the art, models of cartographic relief are known which represent, for example, a country, a mountain range, a town, etc. in miniature and in three dimensions. By cartographic relief, we mean a relief corresponding to an extended geographical area, such as a continent, a country, a city, a district, a region, a village, a park, a mountain, a valley, covering for example an area of 5 square kilometers at least.

[0003] Such models are often made of wood or plastic and adorn, for example, hotel lobbies, chalets, tourist office receptions, and the like. They are used in particular by travelers to plan or relate an excursion or a hike by being much more aware of the relief than by observing a simple paper or electronic map.

[0004] The manufacturing techniques used for the manufacture of such relief models, however, limit their resolution as well as the quantity of information which may appear therein. We know, for example, wooden or plastic models of Switzerland or alpine art in which the minimum resolution corresponds to 1 kilometer. These models allow you to observe the main mountains well, but not to plan an excursion in detail.

[0005] MA NAN ET AL: „Virtual conservation and interaction with our cultural heritage: Framework for multidimensional model based interface“, 2013 DIGITAL HERITAGE INTERNATIONAL CONGRESS (DIGITALHERITAGE), IEEE, vol. 1, October 28, 2013 (2013-10-28), pages 323-330 describe museographic city models, as well as a process for converting them into three-dimensional virtual models, on which additional information can be added.

[0006] This process makes it possible to explore a virtual model of the city even when one is at a distance from the model. The virtual reality process makes it possible to recover the work of the model makers to more quickly generate a virtual environment in which one can navigate even when moving away from the model. However, it does not provide additional information to the visitor observing the physical model. It is therefore virtual reality (VR), but not augmented reality. The 3D model obtained from the mockup can be visited, but it is apparently never aligned with an image of the physical mockup.

[0007] Navigating in an entirely virtual environment is always less intuitive than observing a model. Many users get lost or lose their sense of direction in a three-dimensional simulation, whereas mock-ups are easier to grasp. This is due in particular to the limited resolution of smartphone and even computer screens, which do not display as much information as a model, and therefore require constant scrolling or zooming to go from one overview to a detailed view. For example, an ultra-high resolution computer screen can display about 8 million three-dimensional pixels, while a cartographic model can be machined from a cloud of 100 million or even a billion points, in three dimensions.

[0008] Projections of additional information directly on the model are also described in this document, but it is still not augmented reality.

[0009] The content of the models is however static and cannot be personalized according to the information that interests the user. An object of the present invention is to provide a method to increase the usefulness of such models, and to provide additional information to users admiring them.

Disclosure of Invention

[0010] According to one aspect of the invention, this object is achieved by means of a method for displaying additional data on a model of a geographical relief, comprising the following steps: -capture using an electronic device multimedia comprising a video image data camera representing the model of a geographical relief; -identification of the relief, by comparison with at least one reference model of geographical relief, so as to determine whether the captured image corresponds to a known model, and to which portion of this model it corresponds; - identification of features in said image data; - access to additional data relating to the identified relief; - alignment of captured image data with additional data, based on said features; -display on a display device of said device of an image obtained by superposition in real time of said additional data on said image data.

According to one aspect, the invention therefore relates to a method for applying augmented reality to the world of geographical relief models.

[0012] Augmented reality is naturally known as such. In general, it relates to the superposition of additional information, for example texts, images, videos, textures, etc., on a video image stream. We speak of augmented reality in real time when the shift introduced by the process of calculating the information to be superimposed is sufficiently small to be imperceptible, and the user has the impression of seeing on the screen of his electronic device a modified reality. . Augmented reality therefore differs from virtual reality, in which the user moves through an entirely virtual environment, generally modeled in advance.

[0013] Augmented reality has already been used in the world of cartography. By way of example, document US2011/199479 describes an augmented reality mapping system and a method in which a user films and watches a real scene using a camera and a display screen on a portable device. of pocket. The geographical position of the user is determined by GPS, and the mapping data is overlaid on the moving image of the scene captured by the camera, so as to provide directions or the like to the user, and to highlight highlights points of interest.

[0014] Augmented reality is also known for cartographic applications. For example, the augmented reality map application UCLive (see https://ucliveproject.wordpress.com/, archived at https://web.archive.org/web/2.014081403052.7/http://ucliveproject.wordpress .com/) works as follows. A user takes an image of a standard campus map, and a pre-generated 3D model of the map, including other information layers, is displayed on the handheld device screen. The cropping of the 3D model and the angle of view correspond to the image captured by the camera, so that the user can explore the 2D map in 3D through the handheld device.

[0015] However, these applications focus on enhancing 2D information with 3D modeling, or enhancing the real world with a map or direction information. As such, they are limited in the level of detail they can provide.

[0016] The application of augmented reality to geographical relief models has never been suggested. One of the reasons probably comes from the low resolution and precision of conventional models. These defects make the alignment of additional data on the mock-up image extremely difficult. The resolution of the paper maps and other sources of additional data available is indeed typically several orders of magnitude higher than that of the best models. It is therefore difficult or even impossible to correctly superimpose useful additional data on the image of a model, rendering any attempt at augmented reality in this field useless. Recently, advances in both geographic relief modeling and model-making techniques have opened up hope for geographic relief models of a resolution and accuracy deemed impossible until recently. . Such high-resolution models made it possible for the first time, within the framework of the invention, to imagine extremely useful augmented reality applications, as will be seen later. It is for example possible to manufacture a model from a substrate (which can be the base of the model or a mold for the model) by a process incorporating a machining step in which the movements of a cutting tool having a point are expressed as a path of the tool along the perpendicular axes X, Y and Z in a grid in an XY plane calculated on the basis of a cloud of points generated by the measurement of the relief, said cloud of points comprising a plurality of points which are normalized in a grid of normalized points through which the cutting tool tip is to pass, said tool path being generated by sequentially following said normalized grid of points, i.e. following the grid lines one by one in the same directions or in alternating directions.

[0017] Therefore, a highly detailed rendering of the model can be virtually augmented by providing information via the multimedia electronic device. Due to the high-precision machining process of the model or its mould, the characteristic elements (i.e. the areas of interest, more commonly referred to by the English term of features), for example the edges in relief are reproduced with extreme precision, which is often not the case with conventional calculation and machining techniques. As it is the detection of these features which is used as a reference for the identification of the relief reproduced by the model, then for the alignment of the image of this model with the additional data to be superimposed, the precision of the identification and overlay is greatly improved, which opens the door to entirely new applications and the overlay of additional data that would not have been possible to overlay accurately on a conventional mock-up.

[0018] Further, the additional data does not need to be marked directly on the artwork, and a greater amount and variety of information can be provided compared to prior art methods. Furthermore, as the model is in three dimensions, the user experience is improved compared to a "flat" augmented reality which improves for example a flat map with 3D virtual information: in the present method, a 3D structure is augmented with virtual information (whether 2D or 3D), and the experience is therefore much more tactile and immediate for the user.

In variants, the model can represent an animal or a plant, or another natural object that no model can reproduce with a resolution equivalent to the original (as opposed to an industrial object for example).

[0020] Advantageously, the model identification step is carried out by at least one of the following steps: capturing an image and identifying a code provided on or adjacent to the artwork; analysis of the topographic characteristics of the model; communication with a wireless identification device provided in or adjacent to the breadboard, such as an RFID chip, Bluetooth or WiFi transponder or the like; determination of the geographical location of the multimedia electronic device and comparison of this geographical position with a database comprising a register of the geographical locations of said models.

[0021] Advantageously, the method further comprises a step consisting in determining whether the rendering model is authorized by means of at least one of the following steps: analysis of markers provided on or adjacent to the model; communication with a wireless identification device provided in or adjacent to the model; detection and analysis of an intentional error provided on the model.

[0022] This step can be integrated with, or separated from, the model identification step, and allows the detection of “false” models.

[0023] Advantageously, said point cloud is normalized by scaling said point cloud to predetermined dimensions, namely the dimensions of the substrate to be machined or a part thereof. In the event that the point cloud is not already in the form of a grid, normalizing said point cloud also comprises a step of transforming said point cloud into a uniformly spaced grid having a desired resolution by modifying the X and Y coordinates of each point in order to conform to said grid.

[0024] Advantageously, the normalization of said cloud of points can also comprise the filling of the missing points on the said grid and/or the elimination of the excess points.

[0025] Advantageously, the edges are identified by analyzing the shadows present in said image, which can be accentuated by illuminating the model laterally with a source of artificial light.

[0026] Advantageously, the additional data extracted from the database and projected onto the image of the model represent at least one of the following elements: two-dimensional map information; three-dimensional map information; weather information; routes; annotations; points of interest; animations relating to the object; characteristics of the object; direct links to websites.

These data can be in two or three dimensions.

[0028] They may be available in the form of a two- or three-dimensional image.

[0029] They can be projected onto a plane perpendicular to the shooting direction.

[0030] They can be projected onto a surface corresponding to the portion of relief captured. It is possible to take occlusions into account.

The invention also relates to a product comprising a computer-readable medium, and computer-executable instructions on the computer-readable medium for bringing a multimedia electronic device comprising a processor, a camera and a display device to perform the above process.

Brief description of the drawings

Other details of the invention will appear more clearly in the following description with reference to the appended figures, which show: Fig. 1: a system according to the invention; Fig. 2: an example of using the system from a user's point of view; Fig. 3: another example of using the system from a user's perspective; Fig. 4: a flowchart illustrating the method of the invention; Fig. 5a and 5b: two examples of cartographic information to project onto a view of a geographical relief model; Figure 6: a flowchart illustrating a method of manufacturing a model for use in the invention; Fig. 7: a representation of a point cloud dataset used as the basis for making a model; Fig. 8a and 8b: representations of the limitations on the diameter of the point of the machining tool; Fig. 9a and 9b: two-dimensional illustrations of a finishing tool and the corresponding finishing tool path; Fig. 10a: a view of a substrate, a roughing tool and the corresponding roughing tool path; Fig. 10b: a view of a substrate after roughing; Fig. 10c: a three-dimensional view of part of a substrate after finish machining, with a finish toolpath; Fig. 11: a view of a model equipped with anti-counterfeiting measures; Fig. 12 a-d: a schematic representation of how a direct post processor works to generate a tool path from a point cloud Fig. 13: a schematic representation illustrating the process for processing a cloud of points before a machining step according to the invention, Fig. 14: an embodiment of a system according to the invention.

Embodiment of the invention

Figure 1 illustrates a system 1 according to the invention. The system 1 comprises a multimedia electronic device 3, for example a "smartphone", a tablet or the like, provided with a camera 5 (shown schematically in FIG. 1 by dotted lines representing its field of vision) on a first side thereof, and a display screen 7 on a second side thereof.

As is generally known, such a multimedia electronic device 3 further comprises a processor and a computer-readable storage medium contained therein. These last two elements are contained as is conventional in the device 3 and are therefore not illustrated. The computer-readable storage medium contains, stored thereon, a computer program product (colloquially referred to as an "application"), which includes computer-executable instructions for performing the method described below. In a variant, part or all of this method is implemented by a remote server accessible by the electronic device 3. The multimedia electronic device 3 would typically be provided by a user himself, but this does not necessarily have to be the case.

The system 1 further comprises a model 9 (“relief rendering”) representing a geographical relief characterized by a cloud of points. In the case of FIG. 1, the subject of model 9 is the topography of Switzerland, and is therefore a relief map. The model 9 is manufactured by machining by means of a high-resolution digital control (CNC) machine tool of a plastic, metal, wood or similar substrate. A particularly suitable method of such CNC machining with sufficient resolution is described in more detail below.

[0036] Alternatively, CNC machining of an injection molding or die casting mold is also possible, followed by injection molding, casting molding, blow molding in plastic or lightweight materials (aluminum, ceramic, etc.) or heavy (concrete, etc.) to produce the model 9 is also possible, the substrate 45 then being the base for the mold in this case.

As mentioned above, in Figure 1, the model 9 represents the topography of Switzerland. It can of course represent any different geographical region at an appropriate scale (a country, a region, a state, a municipality, a street, or even a park).

The model 9 can be left colorless, or can be painted, coated, or have information already provided on it. Typically, however, the model 9 should not have data directly on it, so as to encourage the use of the whole system.

It may be advantageous, although not essential, to install the model 9 in an enclosure 11 with controlled lighting, such as an artificial light source, such as a lamp 13 for illuminating the model 9 from on one side (for example from the top in the present figure), so as to cast shadows and therefore improve the contrast of the image received by the camera 5. Such increased contrast is useful for simplifying the processing required during digital manipulation of the image of the model 9 so as to detect the cropping, the position, the angle, the orientation and so on.

The system 1 also includes a database 15, which can either be stored locally in the electronic device 3 or remotely and accessible via a direct or indirect wireless connection such as WLAN, WiFi, Bluetooth, 3G, 4G or similar. The database 15 includes additional data relating to the model portion 9 of which an image has just been captured, such as surface textures, cartographic information (for example place names, level curves, altitudes) , points of interest, weather information, didactic information, routes, for example hiking, biking, cycling routes, recommended or previously traveled by the user and recorded via a satellite geolocation receiver, for example a GPS receiver.

The additional data can include annotations, for example texts, which can be placed more or less freely on the captured image in relation to an element of this image. They may include textures (for example portions of geographical maps) or images (for example satellite images, photos, synthetic images, meteorological images, etc.) which must be superimposed precisely on the captured image, by projecting onto the relief. The data may include still images, for example photographs or graphics. They may include animated images. They may include sounds.

The additional data may include three-dimensional images, for example maps, satellite images, topographic surveys, videos, etc., intended to be projected onto the two-dimensional image of the model captured with the camera. They can also include three-dimensional images, for example a synthetic model or an image captured with a three-dimensional camera or a stereo or distance measurement camera.

[0043] The additional data may include identical generic data which is identical for all the users of the system having captured the same image, and/or personal data, for example data corresponding to a user path.

The user can select the type of additional data that he wishes to display if several different types are provided. The database 15 can include information relating to several models 9, one of which is selected automatically, for example by recognizing a relief on the image, or using a bar code, a QR- code on the image, or by manually selecting a model 9. This point is discussed in more detail below in the context of the method of the invention.

[0045] Figure 2 illustrates the system according to the invention seen from the point of view of a user. In the present case, the user visualizes on a display device 7 of his apparatus an image obtained from the captured image data and on which at least part of the additional data is superimposed. In this example, the image data corresponds to a shot of a model of Switzerland in relief, while the additional data includes cartographic information which is combined on the image of this relief, and correctly aligned with respect to to this relief.

The image data can for example be the image directly taken by the image sensor and displayed on the screen of the device. Pre-processing is also possible, for example to correct brightness, white balance, contrast, or to reduce shadows and reflections, for example. It is also possible to reconstruct an image different from those taken by the camera, for example by combining several successive frames to generate a panorama, or to generate an image taken from a virtual point of view different from the real point of view.

[0047] Figure 3 illustrates the display of a model 9 of part of a three-dimensional model of plant leaf with the multimedia electronic device 3. To look at the small details, an optical adapter 17 can be provided on the camera 5 of the electronic device 3 so as to increase the magnification and thus achieve sufficient zoom and sufficient resolution for the operation of the method of the invention in the case where the model and the corresponding additional data are of very high resolution.

Figure 4 illustrates the method of the invention in its simplest form, in the form of a diagram. At step 21, the elements of system 1 as described above are provided. Typically, the model 9 would be provided in a public place or in a public or private building, in the form of a three-dimensional model in wood or in another material. The electronic device 3 can be supplied by the user himself, although this need not be the case. The computer program product can be pre-loaded on the electronic device 3, provided on a portable data medium, or can be downloaded from a wireless network (Internet, local Wi-Fi, Bluetooth or similar). To this end, the model 9 or its surroundings may indicate how to download the computer program product, for example, by scanning a QR code (Quick Response Code) or a bar code, by connecting with a wireless identification device 53 (see fig. 11) such as, but not limited to, a Radio Frequency Identification tag (RFID), WiFi or Bluetooth modem, or by visiting a listed Internet address. In step 22, the user takes an image of the model 9, which would typically be a video image (that is to say a certain number of frames per second).

The device then generates image data corresponding to the captured model portion. This image data may undergo optional pre-processing, for example to change brightness, contrast, white balance, or viewpoint.

[0050] In step 23, the model 9 is identified. This can be done by various methods.

[0051] The computer program product may contain an algorithm which identifies the model 9 directly from the captured image, and by comparing it with one or more reference models in order to determine whether the captured image corresponds to a model known, and to which portion of this model it corresponds. This identification can implement a classifier, for example a classifier based on a neural network or the like. The reference model can be a two-dimensional model or advantageously a three-dimensional model of the relief.

This identification can in particular be based on areas of interest (characteristic elements, called “features” in image recognition) of the captured image, for example characteristic points of the captured image and of the model of reference. In an advantageous embodiment, the image recognition software uses element edges (edges, corners, depressions and so on), in particular those accentuated by shadows. In the case where the database 15 includes information relating to a single model 9, the latter is identified, insofar as the computer program product determines that the image received is indeed that of the model 9 and not that of a wall, floor, or other object. In addition, the exact part of the imaged model 9 is identified, with the detection of the point of view. The latter can be determined solely on the basis of the image or of successive frames captured by the camera 5; however, it can also use information relating to the orientation in space of the electronic device 3, determined by suitable sensors incorporated in the electronic device 3, for example a compass, an accelerometer, a gyroscope, an inclinometer, etc.

[0053] In the case where the database includes information relating to multiple models 9, the algorithm can identify which model 9 is captured by the camera 5, as well as the exact part of the model 9 imaged and the point of view as described above.

The model 9 can also be identified by taking an image and identifying a logo, one or more words, numbers or symbols, a bar code or a QR code, connecting with a Bluetooth or WiFi transducer. In addition, this identification can be performed by comparing the GPS position of the electronic device 3, as determined by a GPS transducer incorporated therein, with a table of known locations of the models 9 integrated into the database 15. In this case, step 23 can be performed before step 22, as indicated by the double arrow in Figure 4.

The model 9 captured can also be identified by means of a manual selection by the user, for example from a list.

The identification of the model 9 can also include a step of determining the authenticity of the relief, for example by checking whether the means of identification mentioned above are contained in a database of approved models. Figure 11 shows other possibilities, namely: constitution of one or more deliberate errors 50 (for example errors of scale in a part of the model, errors of fact, such as a lake or a mountain or another geographical element added or deleted) and their detection; incorporation of codes, website addresses 51 or similar around the model 9; innocuous aesthetic features of the surrounding of the model 9, such as the bars 52; a wireless identification device 53 such as an RFID chip, a WiFi or Bluetooth transponder or the like.

[0057] In the event that the model 9 is determined to be a counterfeit on the basis of one of the measurements mentioned above, the computer program product can stop the process and provide an indication to the user that model 9 is not authentic.

In step 24, the database 15 is consulted so as to extract additional data relating to the model 9 and/or to a part of the latter being imaged by the camera 5. This information is relevant to the particular model 9. In the case of a model of a geographical relief, the additional data may be two-dimensional or three-dimensional cartographic images, may include annotations of points of interest, geological information, historical information, meteorological renderings (e.g., cloud or weather overlay), day-night rendering, planned or traveled route overlay, or the like. In all cases, the additional data extracted from the database 15 is complementary to any information provided directly on the model 9. In addition, audio and video information can also be provided.

In step 25, these additional data are projected onto the video image taken by the camera 5 so as to create a combined video image integrating both the image data taken by the camera 5, fixed or animated, raw or preprocessed, and also the additional data, fixed or animated, extracted from the database 15. This calculation is carried out in real time, that is to say without any noticeable delay for the user, in order to restore to him in real time on the screen of his device a video image modified and enriched with respect to the captured image. The user thus sees a modified image of the model, or of a portion of the model, which he has before his eyes. He can change his point of view, or zoom in on a detail, simply by moving his device in relation to the model.

[0060] The projection may comprise a step of geometric transformation of the additional data in a plane perpendicular to the shooting direction. For example, in the case of projections of a geographical map or other cartographic data onto the image of a topographical model, the projection may include a step of calculating the image of this map from the point of view of the camera, in order to superimpose images taken from the same point of view.

[0061] In the case of an image corresponding to a model comprising significant relief, for example mountain relief, and/or reference data corresponding to a three-dimensional model, the projection may include a step of calculating the value of color of several points in a three-dimensional space, from image data captured in one or more frames and additional data in 2 or 3 dimensions. A two-dimensional image of this space from the point of view of the camera is then calculated by projection, then returned.

[0062] The projection includes an alignment of the captured image data with the additional data. This alignment can be based on the previously identified features, in particular the (sharp) edges of the relief (ridges, depressions, corners and so on, which are particularly accentuated by shadows), and on a correspondence of these features with features of reference in the reference information in the database 15.

As shown in Figure 5a, the superposition can therefore take place in two dimensions, that is to say by projecting "flat" information in two dimensions 19 on the image of the model in three dimensions. Alternatively, and as shown in Figure 5b, the additional data 19 in two or three dimensions can be "wrapped" on the image of the model 9 in three dimensions in such a way that the superimposed data follow the profile of the relief. , taking into account occlusions, viewpoint, shadows etc. The latter naturally requires much more computing power, but it offers many more possibilities to provide extremely detailed information, especially in cartographic applications. In particular, 3D information allows the detection and representation of more vertices, more shadows and so on, and for example makes it possible to visualize or annotate a characteristic hidden by portions of the model 9 such as a road passing behind a mountain in the view of the model 9. By moving the point of view you can also make certain data invisible when the corresponding elements pass out of view. Furthermore, the additional data superimposed may be different depending on the point of view of the multimedia electronic device 3.

[0064] In step 26, the combined image generated in step 25 is displayed on the screen 7 of the electronic device 3 so that the user can display it.

In order for the electronic device to be able to discern sufficient details of the model 9 so as to correctly analyze the position, the angle and the cropping of the image of the model 9 so that the method can be correctly carried out, the model 9 (or its mold in the case of a molded model 9) must be machined to a sufficient degree of detail, as will be described below, with reference to Figures 6-10 and 12a-d. In particular, the way of machining (and in particular of calculating the trajectories of the tool) ensures an exact and precise representation of the edges which are used for the recognition of the angle and the orientation of the model 9 as taken in image by the electronic device 3. The object of the model 9 (that is to say the topography of the illustrated geographical region) is modeled or sampled by a cloud of points 40 representing points on the surface of the object. These point clouds are not always provided in a normalized grid, and are typically first converted into a network of geometrically defined surfaces using cubic spline functions or any other suitable method. From this network of surfaces, tool movement instructions are generated, taking into account the shape of the tool tip. However, these approaches can lead to artifacts such as unnatural transitions between arbitrarily derived surfaces, and joining small partial renders together also leads to unacceptable artifacts at the junction between these partial renders. Furthermore, such approaches become increasingly difficult to compute as the desired degree of detail increases, requiring exponentially more computational power. The process as described below eliminates these artifacts, and is infinitely scalable, the only limitations being available machine time and tool wear.

[0066] Such a cloud of points 40 is provided at step 30, and an example of it is shown in Figure 7, which shows a cloud of points of a topographic map of Switzerland, at a given resolution. . Each point 40a of the cloud of points comprises spatial information on the point measured in three perpendicular axes X, Y and Z, X and Y being the axes in the horizontal plane, and Z being the vertical axis. In the case of topographic maps, they are usually expressed in longitude (X), latitude (Y) and altitude (Z) respectively. Points can be sorted in the XY plane by sorting first by ascending X value, then by ascending or descending Y value (or vice versa) so as to arrange them in two dimensions in the XY plane.

In steps 31 and 32, the cloud of points 40 is converted directly into movement instructions along the same axes X, Y and Z, X and Y being again in the horizontal plane and Z being the vertical axis. These movement instructions are simply the coordinates of points in space through which the tip of the finishing tool 42 is to pass.

It is calculated by a so-called direct post-processor, which is part of the software executed on a computer. The operation of this direct post-processor is illustrated in Figures 12a-12d. This direct post-processor first takes a cloud of points 40, as shown in Figure 12a, and which can be grouped together from a plurality of smaller files each representing a part of the subject of the model 9, then scales the data to the required scale depending on the size of the model 9 to be formed, taking into account the imperatives linked to the machining technologies (correlations between the grids and the capacities linked to the tools used), and with reference to a suitable reference point for the CNC machine tool. Any conversion of units (eg longitude/latitude to millimeters or micrometers) can also take place at this time. This is shown in Figure 12b, which shows individual data points 40a superimposed on a grid 40b of the desired resolution in the XY plane. In step 31, data points 40a are normalized to grid 40b, modifying the X and Y values of each data point 40a to correspond to an intersection of grid 40b. Any excess points, such as intermediate points where all grid intersections are already occupied can be deleted (as indicated by an „x“), or can also be merged with another point or points at an intersection of the grid. grid by taking the average of these points, for example. Furthermore, all the unoccupied intersections of the grid can either be filled, for example by interpolation of neighboring points, or simply left empty. The resulting normalized mesh points 40c are shown in Figure 12c, with fill data 40d represented by open circles rather than black dots.

[0069] In step 32, by reading each line of the normalized mesh points 40c according to a forward-backward scan as illustrated in FIG. 12d, a series of machining passes, first in one direction then in the opposite direction for the next line, a tool path 43 can be read directly from the standard grid points 40c without further calculation. Alternatively, a raster scan pattern can be generated by reading data from the grid, with each line being read in the same direction, although this is less machine-time efficient due to the time required to retract the tool 42 at the start of the next cut. In the case of missing data points, if the fill has not been completed, the instructions simply advance to the next filled grid intersection in the sequence. A header block 43a can also be added to initialize the CNC machine tool, and a footer block 43b can be added to terminate machining, as needed.

[0070] In the case where the cloud of points 40 is already squared, the normalization carried out by the direct post-processor can simply scale the cloud of points to the desired resolution.

[0071] Due to the large amount of data required to be processed, parallel processing may be used. The post-processor can also carry out a selection of a subset of the points constituting the point cloud 40, for example to reduce the resolution of the latter to a manageable level, for example, taking every 10th data point in order to reduce the resolution by a factor of 10. Compared to a conventional approach of defining surfaces and calculating the tool movements required to form these surfaces with a given shape of the tool, the post-processor approach direct is best used with a tool whose tip is small enough that machining errors resulting from the shape of the finishing tool 42 are minimized, which is retained at step 33. This principle is illustrated in Figs. 8a and 8b. These figures illustrate machining paths 43 formed by grid points of standardized data 40c. Figure 8a illustrates the resulting profile 43c cut with a finishing tool 42 of too large a diameter.

As can clearly be seen, this profile 43 differs significantly from the trajectory 43 determined by the points of the first set. In particular, the large diameter of the end of the tool 42 causes collisions with the steepest sides of the relief, and prevents the machining of narrow holes or right angles at the foot of vertical cliffs, for example.

The size of the finishing tool must therefore be sufficiently fine to allow machining by passing directly over the points of the first set, without the radius offsets usually calculated by CAM programs. Figure 4b illustrates a finishing tool 42 whose tip diameter is small enough that the finished profile is within an acceptable tolerance relative to the machining path 43.

However, a tool that is too thin risks breaking frequently; on the other hand, the curvature of the end of the tool is too pronounced, leading to the machining of grooves forming ridges and ribs visible on the surface. It is therefore advantageous to use a finishing tool whose diameter at the end is greater than the pitch between two points of the first set of points, so that the grooves overlap while avoiding leaving a rib between two grooves. Optimal results have been obtained with finishing tool diameters between 2-10 times the pitch between the points of the first set of points.

In practice, for a very large model 9 several meters wide, a finishing tool whose tip has a diameter of 1 mm may be sufficient. For smaller renders, a finishing tool tip diameter of 0.05mm maximum has been found to be the maximum diameter to provide a level of detail sufficient in practice for the image manipulation required in the electronic device. multimedia 3 as described above. Furthermore, the level of resolution and therefore of detail provided by a tool of these dimensions is sufficient for a useful degree of zoom to be able to be used on the camera of the multimedia electronic device. In principle, the smaller the tip of the finishing tool, the greater the precision and resolution of the model 9, but the longer the machine time required, the more complex the management of tool wear and the greater the Grid resolution needed is large. At step 32, an appropriate finishing tool is chosen for this purpose.

[0076] The trajectory 43 of the finishing tool is also illustrated in a three-dimensional view in FIG. 10c. Finishing tool 42 is illustrated in Figures 9a and 9b as being a tapered tool. Other tool shapes (cylinder, sphere, toroids...) are of course also possible.

Advantageously, the finishing tool 42 is mounted on a 6-axis robot in order to improve the precision and the resolution of the model 9. In this way, the angle of the finishing tool 42 can be adapted from such that the axis of the finishing tool 42 is always perpendicular to the machining path 43.

[0078] Compared to conventional CNC machining based on the combination of standard shapes and standard movements of the tool (arcs, straight lines, splines and so on), the present invention provides the possibility of having much more details and therefore the more precise reproduction of the form on which the model is based 9.

[0079] In step 34, a roughing tool is chosen so as to effectively make a first cut in the substrate, in order to prevent the finishing tool from having to cut too deep and remove too much material, which which would cause excessive tool wear, excessive chip formation and high stresses on the finishing tool 42 with an equally high risk of tool breakage. However, it is not essential that a rough cut be made.

[0080] Based on the path of the finishing tool 43 and the shape of the finishing tool 42, a roughing tool path 46 (see Figure 10a) can be calculated at step 35. The roughing tool 44 has a larger cutting diameter than that of the finishing tool 42, and serves to sufficiently remove the excess material from the substrate 45 in which the model (or its mold) 9 is machined, to so as to leave enough material to make a high quality finish cut, but not so much as to strain the 42 Finishing Tool.

Subsequently, a substrate 45 is provided in a CNC machine tool at step 36, which is then roughed (step 37 - optional) and subjected to finishing operations (step 38) using the tools respective 42, 44 and the respective tool paths 43, 46.

FIG. 10b illustrates a substrate 45 after roughing, and FIG. 10c illustrates the substrate after the finishing operations along the path of the finishing tool 43 so as to create a model 9 (or its mold in the case of a molded model 9).

[0083] Although the use of the direct post-processor described above can lead to the production of a large number of redundant points. For example, the surface of a lake can be described as a flat surface and can therefore be defined more efficiently with conventional CNC operations. It results in no uneven surface being describable and thus being precision machined.

[0084] Advantageously, with reference to FIG. 13, the cloud of points representing points on the surface of the object, measured by laser 54, scanning electron microscope, confocal, or any appropriate device well known to those skilled in the art. art, is continuously sent and processed by the direct post-processor described above and the processed point cloud is continuously sent to the CNC 55 which continuously converts said point cloud into cutting tip movement instructions. the finishing tool 42 along the axes. Therefore, the system according to the invention operates in a continuous manner to reduce processing time.

Incidentally with reference to Figure 14, the system 1 according to the invention further comprises a light source 56, such as a laser for example, illuminating the model 9 which includes structures generating a holographic or diffractive effect. For example, the model 9 comprises surface diffraction gratings providing at least one diffracted image when the model 9 is illuminated by the laser source 56. The holographic reflection grid microstructures and/or the diffractive effect generation structures are well known to those skilled in the art. Such microstructures are in particular described in documents WO93/18419, WO95/04948, WO95/02200, US4,761,253 and EP0105099.

It is understood that the present invention is in no way limited to the embodiments described above and that many modifications can be made thereto without departing from the scope of the appended claims.

Claims (5)

1. Method for displaying additional data on a model of a geographical relief, comprising the following steps: –capture using a multimedia electronic device (3) comprising a camera (5) of video image data representing the model of a geographical relief; –identification of the relief, by comparison with at least one reference model of geographical relief, so as to determine whether the captured image corresponds to a known model, and to which portion of this model it corresponds; –identification of features in said image data; –access to additional data relating to the identified relief; –alignment of captured image data with additional data, based on said features; –display on a display device (7) of said electronic device of an image obtained by superposition in real time of said additional data on said image data. 2. Method according to claim 1, said additional data comprising a texture, said display comprising a projection of said texture onto said image data. 3. Method according to one of the preceding claims, said display comprising a projection of said additional data on a plane perpendicular to the shooting direction. 4. Method according to one of the preceding claims, said display comprising a projection of said additional data on said relief seen from the point of view of the camera (5). 5. Product comprising a computer-readable medium, and a computer program product comprising computer-executable instructions on the computer-readable medium for bringing a multimedia electronic device (3) comprising a processor, a camera (5) and a device display (7) to carry out the method according to one of Claims 1 to 4.