TW201024664A - Method of generating a geodetic reference database product - Google Patents

Abstract

A method of generating a geodetic reference database product is disclosed The method comprises acquiring mobile mapping data captured by means of digital cameras, range sensors and position determination means including GPS and IMU mounted to a vehicle driving across the earth surface, the mobile mapping data comprising simultaneously captured image data, range data and associated position data in a geographic coordinate system. Linear stationary earth surface features are derived from the mobile mapping data by processing the image data, range data and associated position data. 3D-models are generated for the linear stationary earth surface features in the geographic coordinate system from the image data, range data and associated position data and stored in a database to obtain the geodetic reference database product. A 3D-model could include an image representing the colors of the surface of the 3D model or a set of smaller images representing photo-identifiable objects along the model. The 3D-models could be used to rectify aerial imagery, to correct digital elevation models and to improve the triangulation of digital elevation models.

Description

201024664 IX. INSTRUCTIONS: TECHNICAL FIELD OF THE INVENTION The present invention relates to the field of generating a geodetic reference library product. The present invention further relates to a computer implementation system for generating a geodetic reference library product, a geodetic reference library product, a computer program product, and a processor having the computer program product or the geodetic reference library Readable media. A large geodetic reference library product can be useful when orthrographically correcting different images in the same geographic area. Φ [Prior Art] Ground Control Point (GCP) is used to correct satellite, aerial or aerial survey imaging orthography to standard map projection. A ground control point can be attached to any point on the surface of the Earth, which is identifiable in telemetry images, maps or aerial photographs and can be accurately located on each of the 'one ground control point in a standard reference system Has defined associated coordinates. A ground control point is a point on the surface of the Earth at a known location (i.e., fixed in a defined coordinate system). 0(:1> is used to georeference image data sources (such as telemetry images or scanned maps) and separate survey grids (such as those generated during geophysical surveys). A GCP may be: - a paper map a copy of a portion that displays a selected point and its 'around;' • an image from a scanned map showing a selected point and its surroundings; _ an image from a digital map showing its selection Point and its surroundings; 137051.doc 201024664 - written description or sketch of one of the selected points; - from a sky / satellite or ground to indicate a selected point and its surroundings; or - a specific place of its sin Daniel does not, and his system is designed to be identifiable in an aerial/satellite image or a flat map. A GCP can be used to identify any photo-identifiable feature to identify an associated precision X, YW within a target reference system. a point of coordinates...(10) Description

A clearly identifiable surface feature of the Earth in a satellite or aerial imaging. The most important requirement for CGP is its ability to orthographically correct the visibility within the image. - The GCP should ideally have a size of at least 4 times the size of a pixel in the image to be corrected. The Earth's surface features for the Derogatory GCP can be cultural, linear, and natural.

An image of the photo, its dominant cultural (artificial) features are usually used as the best point of GCP. It covers the intersection: intersections, roads and orbits, and roads and visible biogeographic boundaries. (such as - roads and intersections between - forests and farmland) K points, river bridges, large Short buildings (hangars, industrial buildings, etc.), airports, etc. The 'line feature' can be used in this application when it has a well defined edge in imaging. Under normal circumstances, GCP is selected as the intersection of two line features, and the two line features of the intersection and the point must intersect at an angle greater than 6 degrees. The feature is generally * because of its irregular shape. Not good. However, natural features may be necessary in areas where there is a lack of adapted cultural characteristics. If a natural special 13705I.doc -8 - 201024664 sign has a well-defined edge, it can be used as a GCP. It may be a forest boundary, a forest trail, a forest open space, a river confluence, etc. When selecting such points, certain boundaries must be considered in a timely manner (forest, water). In the event that there are insufficient features, the surveyor may be allowed to establish an observable feature for the purpose of identifying a GCP. For georeferencing or correcting aerial or satellite imaging, a set of GCPs must be selected for each image. A group of these GCPs should be evenly selected in the image. Points near the edge of an image should be selected and preferably evenly distributed within the image. The preferred group of GCPs should also focus on terrain changes within the scene, ie, select points at both the highest and lowest elevations. The GCP can be generated by a person using a position determining component (e.g., a GPS receiver) to enter the scene and collect an image or corresponding description of the GCP and corresponding X, Y, and Z coordinates in a standard reference system. In "Accurate mapping of Ground Control Point for Image Rectification and Holistic Planned Grazing Preparation" (Jed Gregory et al., Idaho State Pocatello) University, GIS Training and Research Center, ID 83209-8130, October 2006), GCP must be established and its exact spatial location must be recorded to ensure accurate geographic correction of the imaging. Ten GCPs are strategically placed throughout the area to be corrected. These GCPs are set using two plastic strips (six inches wide and six feet long) that are placed in a cross (+) shape and placed together. All GCPs point each arm of the cross to four mains. Directions (North, South, East, West) are oriented. After placing each GCP, use a Trimble GeoXT GPS unit to record 137051.doc •9· 201024664 a GPS location. This document explains the collection of accurate GCP The amount of time and effort required is basically two corrections that are made in an orthorectification procedure. The orthorectification system changes a perspective image. An image is replaced, wherein each pixel has a known XY position on the geoid surface indicating the surface of the earth and each pixel is viewed perpendicular to the surface of the earth at the XY position. First, any deviation can be corrected. Shift (translation and rotation error) tilt or scale problem and secondly correct the distortion effect of the elevation change. The main cause of the horizontal distortion of the standard distortion in the current orthorectification procedure applied to the image. This system is illustrated in Figure 1. A camera disposed in an aircraft 1 records a perspective image of the Earth's surface 2 (shown here as a porch). However, only one pixel in the image may represent one of the Earth's surfaces and other pixels. It is represented by an angular view of the Earth's surface. Figure 1 shows a contour of the Earth's surface for a given y coordinate. It is assumed that the horizontal line 3 represents one of the reference surfaces of the Earth for a given y coordinate within a standard reference system. Contours, such as WGC84 or any other geoid that describes the surface of the Earth within a standard reference system. Display a building structure 4 on the surface of the Earth, for example The bridge is known for its xyz position and height on the Earth's surface 2. In addition, the position and orientation within the coordinate reference system of the capture point 5 of the aerial image is known (eg by accurate GPS and/or other Position/orientation determining member. By geometry, the pixels on the upper side of the building structure can be determined and the corresponding X, y position determined. However, if the height of the earth surface relative to the reference surface 3 (ie, the Z coordinate) Unknown 'will introduce a first terrain induced error 6 in the orthorectified image. Similarly, if the height of the structure of the building is also not 137051.doc -10- 201024664 will be corrected in the final orthophotos " into an additional building height (four) in this case, the upper side or the structure of the building can be The ortho-corrected image is projected within a few meters of the correct xy position. Once the structure of the building is a bridge, the information on the bridge will be mistakenly fed if the information on the elevation of the reference surface is not (accurate). Figure 2 illustrates this type of error. The image is not corrected by an ortho-radiation image. - bribe or often referred to as "the earth's bare face" is to remove a digital surface model by exposing the terrain below. 2 All cultural and biological features (4)... Deleted in the coordinates reference system One of the first surface of the earth and the elevation information may be represented as a - grating square, a contour or a contour set or a triangular irregular mesh network. The USGS Coffee National Elevation Data Set (Lion) is available as a cost effective _, but does not allow accurate ortho-correction of bridges, buildings, and elevated structures as shown in Figure 2. Since the height of the bridge is not taken into account, the upper side of the bridge will be offset relative to the true location of the bridge. The real location of the bridges in Figure 2 is indicated by a white line superimposed on the orthorectified image. I display - ortho-corrected image 'which uses - accurate geocoding DSM to orthographically correct the aerial image . It can be seen that the building structures are correctly projected onto the orthorectified image space by using the correct height of the building structures. The building structures are correctly projected (4) when the white lines indicating the shape of the building structures coincide with the visual appearances within the ortho-corrected image.

• 1U 137051.doc 201024664 It should be noted that both DEM and DSM provide only one model of the Earth's surface. It does not include information that is easily identifiable on sensory images, maps, and aerial photographs. If the GCP is not associated with a DEM or DSM, it cannot be used to orthographically correct such images. The accuracy of the GCP used and the number of GCPs (counts) and the distribution/density across the image to be corrected will determine the accuracy of the resulting image or orthorectification procedure. The characteristics of the elevation changes below determine the required distribution/density of Gcp. For example, a straight portion of Kansas requires only a few GCPs at the edge of the straight section. A small bridge on a small river does not require much. A bridge over a large canyon may require a high density to properly account for the edge of the bridge. Similarly, undulating hills will require more than a flat slope. Geographic information systems often combine both digital map information and orthophoto corrected images in one view. Information from the image can be extracted or analyzed to add, correct, or verify digital map information. Similarly, orthorectified images can be used to extract digital map information for use in a navigation device. In both cases, the feature points in the orthorectified image correspond to their true location on the earth. In the first case, due to the incorrect height, the positive image is in the positive state. The position of the road surface inside is not related to the surface of the corresponding road T from the digital map. See, for example, Figure 2. In this case, the navigation device can & measure differently from the location in the map database that is extracted from the poor ortho-corrector and may provide an alert that erroneously informs the user of the navigation device Driving conditions. : The requirement for correct ortho-corrected image generation from - aerial imagery or satellite imagery is that there is enough GCP for 137051.doc 201024664 in the area represented by the orthorectified image. Today, the cost of orthorectification increases linearly with the number of Gcps that are to be captured by humans. The more you need GCP to get the accuracy of an ortho-corrected image, the more manpower is needed. There is a lack of inexpensive, accurate (with known accuracy) and well-distributed ground control points to help control positional navigation and mapping applications. In addition, the Advanced Driver Assistance System (ADAS) requires accurate 3〇 position information about the road to control such systems. This requires a dense Gcp network along the surface of the road to be able to adequately correct air or satellite imaging. For these applications, it is important to position the road surface correctly in the orthorectified image. In order to be able to do so, it is necessary to have information about the elevation of the road surface, especially the elevation information of bridges, banks, elevated highways and road bridges. Current technology ground control products for geospatial imaging calibration and correction are incomplete and inconsistent in almost all regions of the world. The following sources exist for the calibration and correction of geospatial data: a) DEM/DTM data derived from government topographic data sets. However, these materials are often thick and outdated. In addition, its quality has changed dramatically between regions. b) Derived from the DEM/DTM of the airborne/satellite radar platform. These systems are expensive and often cover a large area of interest that many commercial graphics entities may not be interested in. These still require position calibration from an independent accurate source. The satellite platform currently does not provide data that consistently meets the accuracy requirements for ADAS level work; c) High quality survey level GPS ground control points. These systems are expensive at each point and require special permission for access in some countries. In addition, the chances for repeatable 137051.doc -13- 201024664 are minimal; d) Low quality GPS ground control points (specific/non-survey level). These often photos are unrecognizable and may be affected by rapid discards. Geodetic metadata may not be accurate and the definition is unclear. In addition, the location of the point is generally not well planned; the GPS"track line" from the vehicle. These systems are almost unrecognizable. J_ does not provide an accuracy above the width of the lane. First, it is difficult to correlate with other trajectories and will give different positions based on subtle travel patterns (especially at intersections), making associated transport nodes impossible. f) Existing aerial imaging products. These may be used to verify/correct low quality output. However, in the generation of geospatial data, these lines are not suitable. In addition, these suffer from a group of localization errors that are not easily detectable in 2d images; and g) existing government or commercial centerline maps. These maps are abstract modeled specifications or centerline data. The accuracy profiles of such data sets are inconsistent and | they lack quality elevation data. There is a need for a geodetic reference library product that contains sufficient Gcp or ground control information to accurately or accurately correct three-dimensional orthorectification air or satellite imaging to use the product as a reliable source of information for GIS applications, at least in its application to the road When the surface is. SUMMARY OF THE INVENTION The present invention is directed to an alternative method of generating a geodetic reference library product that can be used in many GIS applications, such as: image ortho-correction, basic mapping, location-based systems, 3D vision , terrain 137051.doc • 14- 201024664 mapping, vehicle navigation, smart vehicle systems, Adas, flight simulation, cockpit situation awareness. According to the invention, the method comprises: - obtaining a motion map captured by a digital camera, a distance measuring sensor and a position determining means comprising a GPS and an IMU set to a vehicle traveling on the surface of the earth, the action mapping The data is contained in a geographic coordinate system. The image data, the ranging data and the associated location data are simultaneously captured. • The linear still earth surface is determined from the motion mapping data by processing the image data, the ranging data and the associated location data. Feature; generating, from the image data, ranging data, and associated location data, a 3d model for the linear stationary earth surface features in the geographic coordinate system; - storing the 3D models in a database to obtain the Geodetic reference library products. The present invention is based on the recognition that accurate ortho-corrective sensing of airborne and satellite imagery requires a positionally accurate 3D model of the Earth's surface. In addition, the relationship between the sensed image and the 3D model must be determined. Current 3D models (such as dsm and DEM) illustrate the Earth's surface based on 3D coordinates. The 3D coordinates do not have an associated color value corresponding to the surface of the earth when viewed from above. Therefore, it is impossible to align the 3D models with the sensed images. In addition, the commercial image has a pixel size of 5.0 m and a vertical accuracy RMSE of rsme and _m with a level of 2 〇 m. These resolutions and accuracy limits the orthorectification procedure to produce orthorectified images with - higher accuracy. The action cartographic vehicle captures a digital camera, ranging sensor (such as laser/radar 137051.doc -15- 201024664 = device) that is set to a vehicle traveling on the surface of the earth on the road and includes (4) IMU's position of the shirt component _ trip (four) map information, the action map data contains a place in the Jiang Wu "Golden Advisory", the geographic cake system simultaneously captures the scene d-ray data, laser / radar data and related joints The component enables us to be absolutely accurate at a level of 5. (10): with "=::direct: sex" to determine the position [by the laser/radar sensor group & "definite position information has been determined, Establish a surface model with a relative horizontal accuracy of exactly 50 m per and a (10) m 35 phase

For vertical accuracy. Money is better and better, that is, a faster ranging sensor that provides a denser laser cloud that achieves an accuracy of one 丨 cm. From these images of the action mapping data, the linear stationary earth surface features can be determined. A linear stationary earth surface feature may be a road section, a bridge upper side, a day bridge, and the like. One of the characteristics of a linear stationary earth surface feature according to the present invention is that it has a visually detectable edge and a smooth surface, i.e., a surface that does not have discontinuities, such that the surface can be approximated as one between the edges Flat surface. This allows us to use a 3D model, which is illustrated by two fold lines, which illustrate the linear earth surface features' the fold lines corresponding to the left and right sides of the flat surface of the earth's surface features. The surface model can be used to transform the image data into an orthorectified image of the surface of the earth having a pixel size of 2 cm and a relative horizontal accuracy of 1 〇〇 m 5 〇 cm. Height information from the surface model can be added to each pixel of the orthorectified image to obtain a 3D ortho-corrected image having a relative vertical accuracy of 35 cm per 丨〇〇m. From this image data, a linear stationary earth surface feature or a ground control object Gc〇 (such as a road surface) can be extracted and stored as a 3D model in a database for imaging positive 137051.doc •16· 201024664 shot correction. The 3D model of a stationary earth surface feature is characterized by a shape and size that can be identified and identified within the image to be corrected. Another advantage of the 3D surface model according to the present invention is that the 3D model defines both the surface and the edges of the edges to improve the existing dome and DSM products [use of the edges allows for not limited to μμ A location of a typical grid pattern places a cut line or broken line within the surface model. In the surface model, it is not clear how to triangulate four adjacent survey points to provide the best approximation of reality. There are two possible outcomes to triangulating the four points for each violent definition of a different surface. Delaunay triangulation will select the result of maximizing the minimum angle of all angles of the triangles in the triangulation. However, this result does not determine the best results for the surface in reality. A 3D model of linear stationary earth surface features in accordance with the present invention, i.e., the edges can be used as break lines to control a corner measurement 1: </ RTI> to select the three angle measurements of the four survey points that approximate the best reality. These 3D models can also be used to improve quality and reliability when using this surface model in a GIS application at an additional survey point of an existing ^M" or DSMr or when using the surface model to correct aerial imaging. The position information of the 3D model in a standard reference system is accurately known, and the corresponding portion of the image can be accurately corrected. The present invention enables us to generate one that can be used as Gc in an easy manner and in a short period of time. A huge 3D model. The 3D model surpasses the (10)-f library--the advantage is that a 3D model models one part of the earth's surface, while a Gcp only refers to a χγζ coordinate. When using the GCP-database, it must be estimated The elevation information of the location between Gcp, (9) and 35 causes the drawing to be inaccurate. The method 137051.doc 17 201024664 The helper captures the 31) model of the Earth's surface. These points can be used only by people using standard survey methods. Used to measure and model the surface of the earth for manual collection, thereby correcting the error shown in Figure 2. The best of the following three areas is combined according to the method of the present invention: accurate bit S decision, Resolution laser/fit or ground light data processing and high resolution • Image processing. Both laser/radar data and image data have a higher resolution and accuracy, which is represented by aerial imaging. Data captured at a relatively short distance from the surface of the mark. This allows us to use lower-cost cameras and laser sensors. A linear stationary earth surface feature may be selected from any of the earth's surfaces selected from the group. Physical and visual linear features, the group comprising at least the road surface of the road segment, the waterway, any 3D model from which the motion map data can be derived and which is identifiable in an aerial or satellite imagery Physical features of well-defined edges (such as bridges, bridges, baselines of building structures). In other embodiments, the 3D models corresponding to road segments are linked to obtain a continuous linear control network; And storing the continuous linear geographic network, 35, in the geodetic reference library product. The continuous linear control network provides us with one of the achievements of the earth's surface. Seamless 3〇 model, allowing us to accurately correct the image areas corresponding to the road segments. As the road network extends along most of the world, by this invention, an accurate road elevation model can be generated, It can be used to more accurately correct air and satellite imaging in almost any part of the world. In particular, the continuous linear control network can significantly improve the orthorectification of such roads in imaging. The continuous 137051. -18- The 201024664 linear control network provides an extremely accurate DEM or DSM of the surface of such road and road structures, which has a resolution of 5 times better than commercial DSM or DEM. In a specific embodiment, a linear stationary earth surface feature Corresponding to a linear characteristic of a road segment selected from a group feature, the group feature includes: a road center line, a left road edge, a right road edge, and a road width. These features are used to illustrate the 3D model. The 3D model may be systematic

A road centerline, a left road edge, or a right road edge that can optionally combine the road widths. A 3D model of the road surface may be based on the left and right road edges. The 3D model illustrates a shape of a road that is identifiable in an aerial or satellite imagery. Preferably, the 3〇 model corresponds to a road edge and linear paint that is identifiable in the image. The coordinates such as β hai associated with the 3D model can be used to correct the image. In addition, if the 3D model accurately describes the surface, i.e., the elevation deviation&apos;, the area within the image corresponding to the 3D model can be corrected very accurately. In addition, the 3D model is used for DTM refinement/improvement. In one embodiment of the present invention, the program for determining a linear stationary earth surface feature is included in the image data (4) to measure the surface of the road, extracting the position of the edge of the road surface, and combining the image data, the ranging data, and the associated The location data is associated with the linear paint in the geographic coordinate system and the linear stationary earth surface features are calculated from the location of the road surface. The model may be based on a vector that illustrates the linearity of the size and location of the earth's surface features within the coordinate reference system. This is a method for describing the efficiency of space. In a specific embodiment, the method further comprises: 137051.doc • 19· 201024664 by using, and “image material and ranging data to generate ortho-corrected images for the 3D models; Correction of each pixel-level elevation information in the geographic coordinate system; ., and the ortho-correction image and the elevation information to obtain a 3D ortho-correction image; and storing the 3D ortho-correction image These images are then linked to individual 3D models within the large survey database product. These features enable us to enhance the models using the visible features of the 3D model's earth's surface. These features provide additional accurate ground control information related to specific points within the area of the _3D model. These visible characteristics are also characteristics of the stationary earth's surface. Examples of stationary earth surface features or road segments: road markings such as &quot;Warning of,Give Way,just ahead&quot;;stop lines;leading arrows;walking crossings; highway exits Tapered road edge lines; hatch marks; herringbone marks, etc. These roads • Marks can be used as additional ground control information that can be used to detect the corresponding road markings in the image to be corrected. For example, a longer pen segment represented by a 3D model does not provide positional information about the start, end, and trajectory of the road segment to the 3D ortho-corrected image. Since the road is straight, the 3D model does not provide sufficient road control information along the road segment to verify the position of one of the pixels along the road segment in the image to be corrected. Visualizing the road surface and/or road markings, the 3D ortho-correction images enabling the correction program to match photo-recognizable objects from two sources and using the positional information associated with the ortho-corrected images to image Corresponding area drawing 137051.doc -20- 201024664 In this ortho-correction imaging, it is possible to have a well-distributed ground control point by using the "Y and other ortho-corrective images to correct the image. The network is based on the accuracy of the mapping application. In this specific example, __ 1 n Λι|~h , ' 1 can represent the entire positive shot of the model 31 ^ The color of the field—the image, the ortho-corrected image—the inlay or the: the photo of the model identifies the object-group smaller ortho-corrected image.

This, the Month's other purpose is to provide a method for the computer-implemented system to generate content for storage in the -ground control database. The other purpose of this month is to provide a method for correcting the geographic coordinates of the digital elevation model. Another objective of this month is to provide a method of correcting an aerial or satellite imagery, wherein the method includes - acquiring an aerial or satellite image; - acquiring a geodetic reference library product comprising a 3D model; from the geodetic reference material Retrieving - or a plurality of 3D models and corresponding coordinates in the library; - finding locations in the image where the - or more 3D models match the aerial or satellite imagery; and using the 3D models in the coordinate reference system Location to correct the aerial or satellite imagery. Since the 3D model of the road section, for example, describes the earth's surface in more detail and extends more than a set of ground control points, it enables us to improve the corrective procedure. The method can also be used to verify that the orthorectification corrects the aerial or satellite imagery and corrects (i.e., corrects) the matching portion of the image having 137051.doc -21 - 201024664 different coordinates within the coordinate reference system. [Embodiment] Fig. 4 shows a simplified flow chart in accordance with one of the methods of the present invention. The method begins at act 400 by obtaining swaying cartographic data. The motion mapping data is set by a digital camera, a laser sensor (such as a laser broadcast Jw·, „broom) and a position determining component including GPS and ΙΜϋ, which are set on a vehicle that is traveling on the hard surface of the earth. _, the line (four) is included in the - geographic coordinate system

Simultaneous capture of image data, ranging data and associated locations (4). A vehicle-based system called Motion Mapping System S, which is equipped with a positional component, a laser sensor, and a digital camera for collecting motion mapping data. - Position Determination The component is configured to determine the position of the vehicle within the target reference system and, if desired, the orientation of the vehicle. It should be noted that instead of a laser sensor, any other ranging sensing H (such as laser radar, light and radar) can be used to capture data that can be used to generate a 3D model or 3D image. In principle, any image data and ranging data can be used as long as the data includes accurate associated position and orientation data with 6 degrees of freedom. Figure 6 shows an MMs system in the form of a car. Car buckles Prepare one or more cameras 29(1)’ i = 1,2, 3,...! And one or more laser sweepers 23(j) ’ j = i, 2, 3,... J. The angle of view of the camera or cameras 29(1) may be in either direction relative to the direction of travel of the car 21 and thus may be a forward looking camera, a side view camera or a rear view camera or the like. The (several) view window of the camera 29(1) covers the entire road surface in front of the vehicle. Preferably, the angle between the direction of travel of the automobile 21 and the angle of view of a camera is in the range of -45 degrees to +45 degrees on either side. The car 2 can be driven by the driver along the road of interest 137051.doc -22- 201024664. The car 21 has a higher level. The position determining device includes the car 21 having a plurality of wheels 22. And the exact position determines the device. The following components are shown in Figure 6: • (10) (Global Positioning System) unit, which is connected to the antenna 28 and configured to communicate with and receive from a plurality of satellites SLi(4), 2, 3, ...) The signal of SU is used to calculate a position signal. The (10) unit is connected to one

, processor μΡ. Based on the signals received from the (four) unit, the micro-processing ¥ can determine the appropriate display signal to be displayed on the monitor 24 within the car 1 to inform the driver of the location of the car and possibly the direction in which it is traveling. Instead of a coffee unit, you can use a differential GPS single it. Differential Global Positioning System (DGPS) is an enhancement of the Global Positioning System (Gps) that uses a network of fixed, predominantly reference stations to broadcast the locations indicated by the satellite systems and The difference between known fixed positions is known. These stations are broadcast. Hai et al. measure the difference between the virtual distance of the satellite and the actual (internal calculation) virtual distance, and then the receiver station can correct its virtual distance by the same number. DMI (distance measurement $ instrument). The instrument is an odometer that measures the travel of the car 21 by sensing the number of rotations of the wheel 22 or more, and is also connected to the microprocessor 1 to allow the microprocessor to The distance from the GPS unit is used to calculate the distance measured by the DMI when calculating the display signal. ...u( it n measures a few). This can be implemented as a single gyroscope configuration of three gyroscopes to measure rotational acceleration in three orthogonal directions with a translational acceleration of 137051.doc •23- 201024664. The IMU is also coupled to the microprocessor μ to allow the microprocessor to take into account the measurement of the DMI when calculating the display signal from the output signal from the GPS unit. The IMU can also include a termination inferred sensor. It should be noted that the skilled artisan will be able to discover many combinations of GNSS and built-in inertia and termination estimation systems to provide vehicles and therefore equipment (referring to a known position and orientation of a reference position and orientation of the vehicle) Set) one of the exact locations and orientations.

The system as shown in Fig. 21 is a so-called, action mapping system that collects geographic information by, for example, taking images using one or more cameras 29(1) provided on the car 21. The (equal) camera 29(1) is connected to the microprocessor μΡ. The camera 29(1) in front of the car may be a stereo camera. The camera can be configured to generate a sequence of images in which a predetermined frame rate has been used to capture the images. In an exemplary embodiment, one or more of the cameras are configured to capture an artifact at each predefined displacement or every time interval of the automobile 21. The camera 29(i) transmits the images to the μΡβ. In a specific embodiment, the action cartographic handle includes three cameras, a front view camera and a camera at each side. The camera has a viewing axis in the light forward direction relative to the vehicle in a range of up to 60 degrees and preferably 45 degrees. In this case, the front view camera capture is particularly suitable for detecting images of the road direction on the road surface and the sides; ^ &amp; drag &amp;# Camera capture is particularly suitable for detecting objects along the road (啫Such as the road sign) image. Moreover, the laser scanner obtains a laser sample while resting. These 23(j) are in the car 21 along the road of interest to the laser sample and thus contain information relating to the environment associated with the 13705 丨.d〇c -24- 201024664 road, and may include the road surface , building buildings, trees, traffic signs, parking cars, people, direction signs, roadside and other related information. The laser scanners 23(j) are also coupled to the microprocessor μ and transmit the laser samples to the microprocessor μΡ. It is generally necessary to provide location and azimuth measurements from the three measurement units as accurately as possible: GPS, IMU and DMI. The location and orientation data are measured when the camera 29(i) takes an image and the laser scanner 23(j) takes a laser sample. Both of the images and the laser samples are stored for later use in a suitable memory of the μΡ in conjunction with the corresponding location and orientation data of the car 21 collected while the images were taken. Such images include visual information&apos; such as regarding road surfaces, building buildings, trees, traffic signs, parked cars, people, direction signs, landmarks, and the like. The laser scanners 23(j) provide a laser scanner point cloud that is dense enough to be visualized along a 3D representation of the road information. In one embodiment, the (and) laser scanner 23(j) is configured to produce an output having a minimum of 35 Hz and a degree of resolution to produce a sufficiently dense output for the method. A laser scanner (such as the MODEL LMS291-S05 from SICK) is capable of producing this type of output. The minimum configuration of the laser scanner has a laser scanner that looks down in front of or behind the car 21 that senses the surface of the road on which the car is traveling. An optimal configuration is one or two laser scanners scanned in the area to the left or right of the car 21 and a laser scanner viewed from behind or in front of the car 21. The latter has a rotational scanning axis parallel to the direction of travel of the car 21. Other laser scanners have a rotational axis that is at an angle of 45 degrees to the direction of travel of the car 21. Unpublished International Application No. PCT/NL2007/050541 discloses 137051.doc -25- 201024664 an additional advantage of using two laser scanners to scan the same surface at different time constants. It should be noted that instead of a laser scanner, any other ranging sensor can be used that provides distance information or a dense point cloud. Figure 7 shows the position signals obtained from the three measurement units Gps, dmi and . shown in Figure 6. Figure 7 shows a microprocessor μp configuration to calculate six different parameters, i.e., three distance parameters x, y, ζ and three angle parameters ..., % and ~ relative to an origin within a predetermined coordinate system, φ These angular parameters represent a rotation around the x-axis, the y-axis, and the x-axis, respectively. Preferably, the Z direction is consistent with the direction of the gravity vector. The global 111 [1^ or WGS84 coordinate system can be used as a predetermined coordinate reference system. It should be noted that the method in accordance with the present invention can be used with a local coordinate reference system (such as NAD 83) and other national grid systems. The six different parameters provide 6 (χ, y, elevation, forward, roll, pitch) positioning and orientation of the MMS platform required to track the position and orientation of the vehicle in time. The camera and laser scanner have a fixed position and orientation relative to the car 21. This enables us to accurately determine the position of each laser sample in the coordinate reference system and the position of the camera in the coordinate reference system from the six parameters at the instant of obtaining the φ image or laser sample and Orientation. In act 402, a linear earth surface feature is detected in the image data. A linear stationary earth surface feature may be any physical and visual linear feature within the surface of the Earth, such as the road surface edge of a road segment, any physical feature with a well defined visual edge between the two regions, and From the action mapping material, a 3D model is derived therefrom and any other surface features of the earth that are identifiable by the photo in aerial or satellite imaging. 137051.doc -26- 201024664 In act 404, the position in the coordinate reference system of the selected linear stationary earth surface feature is extracted from the image data, the laser data, and the location data of the MMS data. In act 406, a 3D model is generated for the selected linear earth surface feature and, in act 408, the 3D model is stored in a geodetic reference library product. There are many embodiments that can be used to implement acts 402, 404, and 406. Those skilled in the art should be aware of appropriate methods and algorithms for performing the corresponding actions. A solution may separately process images from the MMS data and extract the 3D location information of the feature by combining the image data, the laser data, and the location data. If the same linear feature extends more than one image, the 3 position information of the corresponding images must be combined to model the linear feature. In accordance with the present invention, the images from the MMS data are processed to obtain an ortho-correction mosaic. An ortho-rectification corrects the sigma of the trajectory along the trajectory of the motion mapping system to visualize one of the orthorectified views of the earth's surface. In most cases, it represents one of the road surface, the sidewalk, and the road side along the road. In order to capture the water edge defined by the structure of a building such as a wharf and a levee, the action mapping system may be a navigational waterway. International application WO08044927 discloses a method for generating orthorectifying blocks and inlays from an action graphics image. The images are projected onto a virtual plane representing one of the road surfaces in front of the motion graphics vehicle. The true surface model of the road surface can be easily derived from the laser material. Those skilled in the art can readily adapt the method disclosed in WO 〇〇 〇 44927 to project the jersey image onto the real surface model obtained by processing the ranging data instead of the virtual plane. Wait for the orthophoto to correct the image. Since the position of the 137051.doc -27· 201024664 real surface model is known, immediately after the χγ position in the geo-reference system, the image data and the laser data can be easily derived for each pixel. Level information and link it to the ortho-correction mosaic. It should be noted that in the present application, an orthorectified image means an image containing metadata defining the xy coordinates within the s-coordinate reference system for each pixel. The xy coordinate is the position on the geoid that defines the 3D model of the Earth. In addition, each pixel value is considered to represent the earth's surface as seen perpendicular to the orientation of the earth's surface at the "position", ie the earth's surface of the earth's surface model. The elevation information is defined in the &quot;real world&quot; The height difference between the height of an xy position and the height assumed at the xy position by the geoid that defines the 3D model of the earth. The ortho-corrected images thus obtained are highly suitable for detecting linearities such as road segments. Characterizing and extracting locations within the coordinate reference system of the linear features. From the ortho-corrected images, a linear reference image can be generated. A linear reference image is an image in which a particular line corresponds to an active cartographic vehicle The trajectory line and each column of pixels represent the surface of the earth along a line perpendicular to the trajectory. The unpublished international application PCT/NL2007/050477 discloses a method for generating a linear reference image from motion mapping data. In the image, the curved road is visualized as a straight road. For a straight road, it is not difficult to determine the characteristics of a road section. Such as a centerline, a left road edge, a right road edge, a road width, and a linear road marking. The unpublished international application PCT/NL2007/050159 discloses a system and method for generating road width and center data from orthophoto correction images and Suitable for use in 13705 丨.doc -28- 201024664. The unpublished international application PCT/NL2〇〇7/〇5〇569 discloses a system and method for generating linear lane information from images, wherein the road is in the image The presence-known orientation 1 towel request allows us to accurately detect linear road markings in linear reference images. The previous paragraph interpretation method can be used to determine the linear features from the action mapping data to determine the corresponding position within the images. And calculate the corresponding XY position in a standard reference system. Combining the link elevation information can be easily

The -3D model of the linear feature is generated. Preferably, the view is a vector-based model. A road segment can be modeled in different ways. Figure 9 shows some examples. The 30 model may illustrate the road segment based on the road centerline 92, the paving edge 90, the legal edge 91 (i.e., the left and right edges of the lane line), or any combination thereof. It should be noted that the road centerline may be at the midpoint between the edge of the road or the painted road marking the centerline of the road. When using the model in an application, it should be understood which definition of the H-body embodiment is used, and the linear features are illustrated by polylines. In computer graphics, a fold line consists of one or more line segments—continuous lines... fold lines are specified by the endpoints of each line segment. In a specific embodiment, a fold line is used to illustrate a road section. Furthermore, the slope of the road width and the direction of the vertical road section in the x, y plane can be added to account for the size/width and shape of the road surface. In another embodiment, a road segment is illustrated as two fold lines corresponding to the paving edges. Further, the shape of the road surface that can be added to the one-way road segment corresponding to the road centerline can be described. The road centerline is defined as «, the edge of the pavement section or the edge I37051.doc -29- 201024664 is defined as a soft round surface. In general, the 'road surface can be approximated at two fold lines (eg left and right edges) The shortest line between the lines. Figure 8 shows by way of example the line corresponding to the road t-line 81 and the edge, line 8 〇 &lt; line. Figure 8 shows a first road section 82, a second road section 83 And a third road section 84. The first road section 82 corresponds to the last part of the highway section up to one of the highways. The second road section 83 is opposite to the two highways. The intersection and the third road segment 84 are wider than the beginning of another highway segment. Since the 3D models include elevation information 'the first road segment 82 will be lower than the third road segment 84 and the second Road section 83 will be from first road section 82 The level gradually changes to the level of the third road segment 84. This is used to correctly correct valuable information for aircraft or satellite imaging. In addition, the slope and curvature of the second road segment provide important information for ADAS applications. The 3D model can also illustrate the surface of the road surface between the edge lines by a DSM derived from the laser data. This DSM will have a denser than the current DSM derived from the onboard bite φ satellite platform. The laser point grid. The DSM thus obtained can be used to locally enrich the DSM/DEM from the airborne or satellite platform with more accurate and dense elevation information » The 3D models of the road segments are stored in the database. In act 41, the 3D models of the road segments are joined together to form a continuous control network. The nodes of the network correspond to contacts and the branches of the network correspond to between contacts Or a road segment connected to the joint. In the action Μ, the continuous linear control network is stored in the georeferenced database. The network provides a component to easily extract a road from the database. 137 051,doc •30- 201024664 Surface-to-job. The characteristics of the network are that the road segments are touched in the continuous and seamless DSM from the road network. You can be sure this / point Since the road segments are derived from the same data source, that is, the phase-in-time mapping period. The primary image data is used to first locate the image on the road surface of the image and by combining them in the images. The location and the laser data determine the location of the road surface within the coordinate reference system. However, the image data can be further used to use the "real world" appearance of the road surface to enhance/enhance the 3D model, display road markings , the texture and color of the road surface, the type of people's lamp, shoulders, etc. In addition, the indicia can form a dense (10) array to accomplish the positioning and/or correction of a road segment. In action ,, an orthorectification image is generated for a linear feature. As explained above, in acts 402 through 406, one of the road surfaces has been formed to orthographically correct the image or mosaic. Thus, act 414 can be limited to selecting corresponding regions or pixels of the orthorectified images to form an orthorectified image for a 3D model. Optionally, in act 416, the elevation information is associated with each pixel of the orthorectified image for a 3 〇 model. If the linear feature is a road segment' then the road surface can be approximated as a flat surface between the edge lines. The elevation information can be derived by interpolation techniques between the edge lines. In act 418, a 3D ortho-corrected image is generated by linking the orthorectification corrected image to the elevation information. In act 42, the 3D orthorectification image is stored in the georeferenced library along with one of the corresponding 3D models. Thus, in one embodiment, a 31) model further includes an orthorectified image of the corresponding road segment as illustrated by the 137051.doc 31·201024664 fold line. The positive image can be accurately derived from the image data, ranging data, and position/location data. The ortho-rectification image created from the procedures described above can be used as a reference image to improve the process of correcting aerial or satellite imagery and even to correct/improve corrected aerial or satellite imagery. These road mouth paints (such as road centerlines, dashed lines, and stop lines) can be used to find a match in the image to be corrected/corrected. This will provide additional ground control points to correct/correct the image along the road segment. The laser data can be further used to assign elevation information to each pixel of the orthorectified image associated with a 3D model. The elevation information can be used to transform the orthorectified image within an image corresponding to when viewed from the location at which the image is to be corrected. This improvement in the corrective procedure matches the accuracy of the procedure and reduces the chance of mismatching. It should be noted that the size of a 3D ortho-corrected image (which includes elevation information) should not be limited to the area of the surface of the road. It may represent the place in the road corridor

There is an orthorectified view of the Earth's surface that can be derived from the image data and ranging data. It should be noted that instead of using one of the three sections of the road section to correct the image, an image patch can be produced. An image slice is represented by one of the features of a stationary earth surface. Examples of surface features of stationary roads: a stop line, 'Give Way' just ahead, guide arrows, sewer fences, speed limits, pedestrian crossings, tapered road edges at the exit, sharp edge edges, for people The metal cap of the hole cover and any other direction of Figure 9 indicate 9 〇. Other road surface features are hatch marks or chevron marks, reflecting arrows, and dividing arrows. Traffic Sign Handbook 2003 (Chapter 5, Road Marking, 137051. Doc -32· 201024664 ISBN 0 η 552479 7) Provide usable road markings - an overview. Figure 10 explains &quot;Warning of,Give Way,just 讣(5)士,1〇〇, hatch mark 102, parking line H) 4 and direction indication 1〇6β $ 夕, any other nozzle paint, sidewalk type light (four), stone base, unique low quasi-geological features can be used to generate an image. - Video film &amp; A stationary earth surface obtained by imaging an image. A snapshot image of the feature and metadata representing the elevation of the χγ position and height information in the coordinate reference system. At least one pixel of an image must be There is associated location information to define the location of the image within the coordinate reference system. This may be a relative position relative to the associated 3D model. Optionally, the image may have the initial positive duration of the motion mapping session A reference to correct an image or block or image to allow manual verification of the image slice. Each pixel of an image slice may contain an associated tagged message within the coordinate reference system. The image slice is also a 3D positive Correcting the image. The size of the image depends on the size of the surface features of the stationary road and the size of the pixel. A preferred area of φ pixels represents an area of 3 to 15 by 3 to 15 〇^, which has a standard reference system. An absolute horizontal resolution above 5〇cm and an absolute vertical resolution higher than .i, 5m. The resolution in the database depends on the accuracy of the image data, ranging data and position/orientation data. Sex/resolution and the application that the database product is intended for. The video images have a link to the corresponding 3〇 model and are stored in the geodetic reference library. GCP is used to discover in air or satellite imaging and to guide the discovery of a matching location for a 3D model in air or satellite imaging. 137051.doc -33· 201024664 The method according to the invention has been The bait material captured by the lower-cost vehicle produces a large geodetic reference library product that can be equipped with a relatively inexpensive digital camera, laser sensor and position determining component. This method establishes a photo-recogniable data set, which can Used as a &amp; face control object in an orthorectification procedure. The present invention allows us to obtain a collection of ground control objects and GCPs that are orders of magnitude larger than conventional ground control. The method has a versatile and verifiable accuracy round φ ^ ° in all geodetic dimensions. This method does not require the creation of special photographs on the scene to identify the Earth's surface f for use in orthophotos to correct future aerial imaging. In addition, the database produces «port 3's essentially photo-recognizable material that will exist for many years. It can also be used to correct 3D surface models because the library product contains 3 pieces of information. Another advantage of the use of MMS bedding is that during the active mapping session, the image data and the laser data record the surface of the earth surface more than once across a joint or a road segment. These areas may contain - stationary road surface features that can be used as a ground control object. • #上上' The stationary road surface features have the same location within the coordinate reference system. However, this positioning decision may have some absolute and relative inaccuracies during the action charting session. Two or more methods are employed in accordance with the present invention: the linear stationary road surface features are under-selected and each time a corresponding χ gamma position and elevation information 2 coordinates are determined. For each decision on the linear stationary road surface feature, a record can be made in a library containing a 3D model and a meta-information describing the position of the χγζ and a reference to the original ortho-corrected image. Redundant information can be removed from this resource by analyzing the records associated with the same linear earth surface features. For example, by combining (ie averaging) or anomalous 137051.doc •34· 201024664 to exclude such images and meta-data of the same linear stationary road surface features, redundant information can be removed ^ by averaging the XY position and elevation Information can be calculated for a 3D model with metadata for the average of the coordinates. The average will generally define the location of the 3D model more accurately within the coordinate reference system. In FIG. 5, a computer configuration suitable for implementing the present invention is proposed.

I. Overview. Computer configuration 500 includes a processor 511 for performing arithmetic operations. The processor 511 is connected to a plurality of memory components, including a hard disk 5 12, a read-only memory (R〇M) 5 〖3, an electrically erasable programmable read-only memory (EEPR 〇 M) 514, and Random access memory (RAM) 515. The memory components include a computer program containing data, ie, instructions configured to allow the processor 5 11 to perform a method for generating a spatial data change message or a method for processing a spatial data change message in accordance with the present invention. . It is not necessary to provide all of these memory types. Moreover, the memory, the cow does not have to be physically located close to the processor 511, but can be positioned at the processor 5&quot; far end. (4) The input data and output data associated with the method may or may not be stored as part of the computer configuration. For example, the input data can be accessed via a web service. It is even possible to perform an action by a program running on another processor. The processing benefit 511 is also connected to means for inputting instructions, data, etc., such as - keyboard 516 and - mouse 517. Other input components known to those skilled in the art, such as a touch screen, a trackball, and/or a voice converter, may also be provided. A read-read unit 519 is provided that is coupled to the processor 5ι. Reading a single 137051.doc -35- 201024664 π519 is a ge&lt;&lt;&gt;&lt;&gt;&lt;&gt;&lt;&gt;&lt;&gt;&gt; On it. Other removable data carriers may be tapes, DVDs, CD_Rs, DVD-Rs, memory sticks, solid state memories (SD cards, usb sticks), cf cards, HDDVDs, Blu-rays, etc., as is known to those skilled in the art. . The processor 511 can be connected to a printer 523 for printing output data on paper, and a display 518 such as a monitor or lcd (liquid crystal φ display) screen, a head up display (projected to the front window) or a habit Know any other type of display known to the person skilled in the art. The processor 511 can be coupled to a speaker 529 and/or an optical reader 531, such as a digital camera/network camera or a scanner, the scanner being configured to scan graphics and other files. In addition, the processor 511 can be connected to a communication network 527 by an I/O component 525, such as a public switched telephone network (PSTN), a regional network (LAN), a wide area network (WAN), and a wireless LAN. (wlan), GPRS GPRS, UMTS, Internet, etc. Processor 511 can be configured to communicate with other communication configurations over network 527. The data carrier 52, 521 may contain a computer program product in the form of data and instructions configured to provide the processor with the ability to perform a method in accordance with the present invention. However, such computer program products can alternatively be downloaded to a rough body via telecommunications network 527. The processor 5 11 can be implemented as a stand-alone system or a plurality of parallel operation processors, each of which is configured to implement a sub-task of a larger computer program, 137051.doc • 36· 201024664 or one with several sub-processors Or multiple main processors. Portions of the functionality of the present invention may even be implemented by a remote processor that communicates with processor 51 through telecommunications network 527. The components contained within the computer system of Figure 5 are typically found in a general purpose computer system and are desirably broadly classified on behalf of one of such computer components well known in the art. • Thus the computer system of Figure 5 can be a personal computer, a workstation, a mini computer, a host computer, and the like. The computer can also include different bus configurations, network platforms, multi-processor platforms, and more. A variety of operating systems are available, including UNIX, Solaris, Linux, Windows, Macintosh, and other suitable operating systems. Figure 11 shows a flow chart of one method for correcting an aerial or satellite image. In act 1100, the aerial or satellite imagery is acquired. Preferably, the ΔHai image is a perspective image instead of one of the perspective images, and the ortho-corrected image may contain distortion caused by performing an orthorectification sequence φ, which cannot be corrected. Or may cause additional distortion when performing this correction procedure. In act 1102, the geodetic reference material library is obtained, which includes the 3] model obtained by the present invention. In act 11〇4, a 3D model is retrieved from the geodetic reference library. Preferably, only the 3D model is selected, which is intended to be covered by the image to be corrected. In action ^(10) 'search for a location in the image that matches the 3D model. The buckle model illustrates the boundary of a photo-recognizable area. Therefore, a corresponding matching area can be found in the image. One of the 3D models of the orthorectified correction view can be used to find a matching area. 137051.doc -37- 201024664 Under normal circumstances, the position of a digital camera that takes aerial or satellite imagery is known within the coordinate reference system. This allows us to transform the model into a perspective image as viewed from the position of the digital camera and find the corresponding location in the image. This transformation improves the success rate of finding the right place in the image. After the location is discovered, adjacent 3D models are used to find the corresponding locations in the image. Repeat this procedure until the adjacent 3〇 model falls into the image. In this manner, the input φ is provided between the 3D model and the corresponding location within the image to correct the image in action 1. Each matched 3D model is used as a ground control object. Now all 3] 〇 models with a location within the assumed region of the Earth's surface visualized by individual images are used as DSMs on which the images should be projected. The matching of the matching points in the image corresponds to the 3D model so that we can correctly orthographically correct the portions of the image corresponding to the matching locations. The areas covered by the three 〇 models that form the network can be corrected by a generally known correction algorithm. φ Figure 12 shows a flow chart of a method for correcting an OEM. In action 1200, a digital scale model or a digital surface model to be corrected is acquired. In act 1202, the geodetic reference library is obtained. In act 1204, one or more 3〇 models are retrieved from the geodetic reference library. In act 1206, a location is searched for at which location the one or more 3D models match the digital elevation model. In act 12-8, the deviations between the coordinates in the coordinate reference system of the 3D model and the locations within the dEm are determined. These deviations are analyzed to determine the type of error. This error may be a translation error, a scaling error, or a local error. Base 137051.doc -38- 201024664 The digital elevation model is corrected in act 1210 for the results of the analysis, i.e., the type of error. The method for correcting a DEM can be further adapted to improve the triangulation of a DEM that includes a color representative of the earth's surface. The DEM can be represented as a raster (a grid) or a triangular irregular network. When using Detroit's two-corner split, maximize the minimum angle of all angles within the triangulation. It tends to avoid extremely thin triangles. However, a quadrilateral formed by 4 points has two possible triangulations. The action 丨21〇 φ is further adapted to use a 3D model that illustrates the outer edge of a road segment having a flat or soft circular surface as a break line to control the triangulation. The 3D model representing a road segment illustrates the outer edge of a soft circular surface area. In a specific embodiment, the broken lines are used to add an additional elevation to the DEM and the triangulation uses the additional point. In another specific embodiment, act 1210 is adapted to select one of two possible triangulations that best corresponds to the defined surface of the 3〇 model. The geodetic database in accordance with the present invention can be further adapted to locally improve a DEM/DSM by adding a dense point network that is representative of a 3D model of a road segment. The 3D model can also be used to replace one of the S. This provides -DEM/DSM, which can be used in navigation applications, such as ADAS applications. The foregoing detailed description of the invention has been presented by way of illustration The invention is not intended to be exhaustive or to limit the invention to the precise form disclosed. The descriptions of the embodiments are intended to be illustrative of the principles of the present invention and the practice thereof so that the skilled person can use various embodiments with the 137051.doc •39· 201024664 Various modifications are made to make the best use of the present invention. It is intended that the scope of the invention be defined by the scope of the appended claims. BRIEF DESCRIPTION OF THE DRAWINGS The invention has been discussed in more detail above with reference to the accompanying drawings, in which FIG. 1 schematically shows a source of distortion in the orthorectification procedure; FIG. 2 shows a positive Shooting correction images and a use;

3 shows a use of an orthorectified image and a DSM; FIG. 4 shows a flow chart of a method according to the invention; FIG. 5 is a block diagram of an exemplary computer system for implementing the method according to the invention; 6 shows a mms system with one camera and one laser scanner; Figure 7 shows a map of the location and orientation parameters; Figure 8 illustrates an example of the surface characteristics of a linear stationary road; Figure 9 illustrates the model used to model linear 罄天旺Figure 1 shows an example of additional ground control information; Figure 11 shows a flow chart of one method for correcting an air φ red pain f (3) dimethyl or satellite imagery; and Figure 12 shows a For correction - [Main component symbol description] A flow chart of the DEM method Aircraft 2 Earth surface horizontal line 137051.doc 3 201024664

4 5 6 7 21 22 23(j) 24 28 29(i) 80 81 82 83 84 90 91 92 100 102 104 106 Building structure capture point first terrain induced error additional building height induced error car wheel laser scanner Monitor Antenna Camera Edge Line Road Center Line First Road Section Second Road Section Third Road Section Laying Edge/Direction Indicates Legal Edge Road Centerline &quot;Warning of 'Give Way' ahead&quot; Shadow Marking Parking Line Direction Indication computer configuration 137051.doc -41 - 500 201024664

511 512 513 514 515 516 517 518 519 520 521 523 525 527 529 531 Processor hard disk read-only memory (ROM) electrically erasable programmable read-only memory (EEPROM) random access memory (RAM) keyboard Mouse display read unit floppy disk / data carrier CDROM / data carrier printer I / O components communication network / telecommunication network speaker optical reader 137051.doc • 42-

Claims (1)

201024664 X. Applying for a patent: 1. A method for generating a geodetic reference library product, the method comprising: - obtaining a digital camera, a ranging sensor, and a vehicle set by a vehicle traveling on the surface of the earth The action mapping data captured by the location determining component, the motion mapping data includes capturing image data, ranging data and associated location data simultaneously in a geographic coordinate system; - processing the image data, ranging data and associated locations by processing the image data Data from the action mapping data to determine a linear stationary earth surface feature; - generating a 3D model for the linear stationary earth surface features from the image data, ranging data, and associated location data; The 3D model is stored in a database to obtain the geodetic reference library product. / 2. The method of claim 1 wherein the method further comprises: - concatenating the 3D models to obtain a continuous linear control network; and - storing the continuous linear control network in the geodetic reference library product . 3. The method of claim 2 or 2, wherein the linear stationary earth surface feature corresponds to one of a road segment selected from a group feature, the linear feature comprising the road center line, the left road edge , right road edge, road width. For example, the method of "Study Item 1 or 2", wherein the linear stationary earth surface feature is determined to include: _ detecting a road surface in the image data; and 137051.doc 201024664 by ', and. The image data, ranging data, and associated location data extract the location of the road surface within the geographic coordinate system; - calculate _ or a plurality of fold lines from the location of the road surface that represent the linear stationary earth surface feature. 5. The method of claim 1 or 2, wherein the 3D model is based on a vector. 6. The method of claim 1 or 2, wherein the method further comprises: generating an orthorectified image for the 3D models by combining the image data and the ranging data, for each _pixel of the ortho-corrected image Determining the elevation information in the geographic coordinate system; linking the ortho-correction image and the elevation information to obtain a 3d ortho-correction image; and - storing the 3DJE-corrected images and linking the images to the ground Measure individual 3Ε&gt; models in the database product. 7. A geodetic reference library product comprising a 3D model representative of a linear static ruthenium surface feature, wherein the 3D models have been generated by any of the methods of claims 1 through 6. 8. The geodetic reference library product of claim 7, further comprising a continuous linear control network constructed by linking the 3D models. 3 9. The geodetic reference library product of claim 7 or 8, wherein the product includes: an orthorectified image: each orthorectification image representation: at least a portion of the surface of the earth represented by the 3D model, Where - each pixel of the orthorectified image contains associated elevation information. 10. A method for improving triangulation of a one-digit elevation model, wherein the method I37051.doc 201024664 comprises: obtaining the digital elevation model; _ obtaining a geodetic reference library product as claimed in claim 7; _ from the geodetic measurement Retrieving a 3D model in the reference library, wherein the 3D model illustrates an outer edge of a soft circular surface region; - finding a location in which the 3D model matches within the digital elevation model; and - using the soft circular surface region The outer edges act as break lines to control the triangulation. A method of correcting a geographic coordinate of a digital elevation model, wherein the method includes: obtaining the digital elevation model; _ acquiring a geodetic reference library product as in claim 7; _ retrieving from the geodetic reference database a plurality of 3D models; - finding a location in which the 3D models match the digital elevation model; - determining a positional deviation between a location of the discovered location within the digital elevation model and a coordinate associated with the one or more branches And - using the positional deviations to correct the geographic coordinates of the digital elevation model. 12. A method of correcting an aerial or satellite image, wherein the method comprises acquiring an aerial or satellite image; _ acquiring a geodetic reference library product as claimed in claim 8; - retrieving a reference from the geodetic reference database Or a plurality of 31) models and corresponding coordinates; 137051.doc 201024664 - found in the image where the one or more 3D models match the location of the air or satellite imagery; and - use the 3D within the coordinate reference system The location of the model and the corresponding discovery location to correct the aerial or satellite imagery. - a computer implementation system comprising: a processor (5 ιι) and a memory (512, 513, 514, 515) connected to the processor, the memory comprising a computer program, the computer program comprising a spear The stipulations and instructions are configured to allow the processor to perform any of the methods of claims 1 through 6 and 10 through 12. 14. A 禋 α Λ 兮 兮 兮 兮 兮 兮 兮 兮 兮 兮 兮 兮 兮 兮 兮 兮 兮 兮 兮 兮 兮 兮 兮 兮 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 A processor readable media brain program product includes one that allows the computer to configure any of them. 15. ‘There is a computer program product that is loaded with information and instructions, such as the methods of Requests 1 to 6 and 10 to 12, 137051.doc