CN106875467B - Three-dimensional city model fast updating method - Google Patents

Abstract

The present invention provides a method for rapidly updating a three-dimensional city model, which includes the following steps: S101 , collecting street view photos from different perspectives of a changing area in a three-dimensional city; S102 , analyzing the model of the three-dimensional city, and locating the changing area in the area in the model of the three-dimensional city; S103, establish an image feature library according to the buildings in the model of the three-dimensional city; S104, perform three-dimensional space change detection on the three-dimensional city, and determine the buildings to be updated and reconstructed; S105, Perform feature matching on the street view photos from different perspectives; S106 , accurately fuse the updated and reconstructed building and the background three-dimensional model. The three-dimensional city model rapid update method is based on the existing three-dimensional city data, and uses the invariants in the existing old data as the precision source of the update data correction to control the precision of the new data, so as to solve the problem of image correction of ordinary street view photos.

Description

A Fast Update Method for 3D City Model

technical field

The invention relates to the technical field of three-dimensional city modeling, in particular to a method for rapidly updating a three-dimensional city model.

Background technique

Large-scale, high-precision, and fast fully automatic three-dimensional urban modeling technology has always been the main research direction in the field of photogrammetry and digital cities. At present, the main technical means of urban 3D reconstruction that have been proved to be accurate and effective include: (1) Stereo measurement of airborne area array or multi-line array CCD aerial survey cameras; (2) Airborne LiDAR point cloud measurement combined with high-precision 2D images; ( 3) Airborne multi-camera oblique photogrammetry; (4) Vehicle-mounted LiDAR point cloud measurement to assist panoramic photos, combined with aerial 2D data; (5) Full manual 3Dmax modeling.

The above technologies and methods provide many options for large-scale, fast, and accurate urban 3D reconstruction, but they all face an important problem, that is, the high cost of expensive equipment and the inconvenience of subsequent updates, so the platform construction Still faced with tasks that need to be updated after completion. However, if the traditional 3Dmax or large-area urban aerial photography modeling method is used to update, the maintenance is still time-consuming and expensive.

With the rapid development of terminal electronics, the cost of obtaining images is getting lower and lower. Mobile phones with camera functions, cheap high-resolution digital cameras, etc. have been fully popularized, and urban street view data can be obtained quickly and at low cost. The three-dimensional modeling of the city provides the basis and conditions. However, due to these uncalibrated ordinary cameras and the photographers, these images have problems of distortion and projection deformation. How to geometrically locate the position of these images, correct the distortion and deformation is the primary problem. Most of the 3D platforms in cities use traditional high-precision surveying and mapping data as 2D basemaps with high-precision 3D modeling methods, with fairly accurate absolute positioning accuracy and relative positioning accuracy, and high data resolution and model finesse. , which can provide a good benchmark dataset for subsequent data update in other ways.

SUMMARY OF THE INVENTION

Based on this, the present invention provides a method for rapid updating of a three-dimensional city model to effectively solve the problems existing in the prior art.

A method for rapidly updating a three-dimensional city model, comprising the following steps:

S101. Collect street view photos from different perspectives of changing areas in a three-dimensional city;

S102, analyzing the model of the three-dimensional city, and locating the area of the change area in the model of the three-dimensional city;

S103, establishing an image feature library according to the buildings in the model of the three-dimensional city;

S104, performing three-dimensional space change detection on the three-dimensional city, and determining a building to be updated and reconstructed;

S105, performing feature matching on the street view photos of different viewing angles;

S106. Perform precise fusion of the updated and reconstructed building and the background three-dimensional model.

In a preferred embodiment of the present invention, the street view photos of different viewing angles are obtained by non-measurement professional equipment.

In a preferred embodiment of the present invention, the non-measurement type professional equipment includes a mobile phone with a camera and an ordinary camera.

In a preferred embodiment of the present invention, the street view photos from different viewing angles include photos of different facades of buildings in the three-dimensional city.

In a preferred embodiment of the present invention, in step S104, the three-dimensional city is first classified, non-building objects are filtered, and then the buildings in the city are first judged based on image features, and the objects that can be confirmed to have changed are determined. Set up a buffer zone and establish a frame of reference for nearby unchanging buildings.

In a preferred embodiment of the present invention, in step S105 , the dense matching algorithm is accelerated by using the sunshine conditions and the geometric relationship of the buildings in the 3D city to optimize the reconstruction of the dense point cloud, and further reconstruct the local 3D model.

Compared with the prior art, the method for rapidly updating a three-dimensional city model provided by the present invention is based on the existing three-dimensional city data, and uses the invariants in the existing old data as the source of accuracy for the correction of the updated data, so as to control the accuracy of the new data so as to ensure that the new data is corrected. This solves the problem of image correction in ordinary Street View photos taken by non-measurement cameras. Specifically, the three-dimensional urban model rapid update method can make use of the texture features and prior knowledge of the existing high-precision digital urban spatial basic information platform results data as much as possible, and takes the invariable building model in the city as the control condition, and adopts The DSM (Digital Surface Model) model is constructed from the street view photos taken from multiple angles to realize the three-dimensional feature update of the urban changing area, which can solve the local image distortion and deviation caused by the non-mapping camera that cannot be handled by the traditional photogrammetry method. It can meet the needs of rapid urban development and has broad application prospects.

Description of drawings

FIG. 1 is a flowchart of a method for rapidly updating a three-dimensional city model provided by an embodiment of the present invention;

FIG. 2 is a schematic diagram of the method for fast updating of the three-dimensional city model shown in FIG. 1 .

Detailed ways

In order to facilitate understanding of the present invention, the present invention will be described more fully hereinafter with reference to the related drawings. The preferred embodiments of the invention are shown in the accompanying drawings. The above are only preferred embodiments of the present invention, and are not intended to limit the scope of the present invention. Any equivalent structure or equivalent process transformation made by using the contents of the description and drawings of the present invention, or directly or indirectly applied in other related technical fields , are similarly included in the scope of patent protection of the present invention.

Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. The terms used herein in the description of the present invention are for the purpose of describing specific embodiments only, and are not intended to limit the present invention. As used herein, the term "and/or" includes any and all combinations of one or more of the associated listed items.

Please refer to FIG. 1 and FIG. 2 together. A method for quickly updating a three-dimensional city model provided by an embodiment of the present invention includes the following steps:

S101. Collect street view photos from different perspectives of a changing area in a three-dimensional city.

In this embodiment, non-measurement professional equipment such as handheld consumer-grade ordinary cameras and mobile phones are mainly used to obtain street view photos of different perspectives of changing areas in a three-dimensional city. The street view photos of different perspectives can be different exteriors of buildings. Faces provide the texture information needed for stereo mapping. It is understandable that, for the image of the top of the building, it can be collected with unmanned aerial photography aircraft, such as a small six-rotor aircraft.

S102. Analyze the three-dimensional city model, and locate the area of the change area in the three-dimensional city model.

It can be understood that the change area is the area that needs to be updated in the three-dimensional city model.

S103. Establish an image feature library according to the buildings in the three-dimensional city model.

Therefore, the matching speed and accuracy of feature points can be improved for the subsequent reconstruction of the local three-dimensional model.

S104. Perform three-dimensional space change detection on the three-dimensional city, and determine a building to be updated and reconstructed.

In this embodiment, the three-dimensional city is first classified, and non-building objects, such as trees, etc. are filtered out, and then the buildings in the city are first judged based on image features, and a buffer zone is established for the objects that can be confirmed to have changed. Identify nearby invariant buildings to establish a frame of reference.

S105. Perform feature matching on the street view photos of different viewing angles.

In this embodiment, in the process of matching the street view photos of different viewing angles, a dense matching algorithm with more prior knowledge is used. Specifically, the dense matching algorithm is accelerated by using the sunshine conditions and the geometric relationship of the buildings in the three-dimensional city, the reconstruction of the dense point cloud is optimized, and the local three-dimensional model is further reconstructed.

S106. Perform precise fusion of the updated and reconstructed building and the background three-dimensional model.

It can be understood that the background three-dimensional model is the unchanged area other than the changed area in the model of the three-dimensional city.

After the updated and reconstructed building and the background three-dimensional model are accurately fused, the rapid update of the three-dimensional city model can be completed.

Compared with the prior art, the method for rapidly updating a three-dimensional city model provided by the present invention is based on the existing three-dimensional city data, and uses the invariants in the existing old data as the source of accuracy for the correction of the updated data, so as to control the accuracy of the new data so as to ensure that the new data is corrected. This solves the problem of image correction in ordinary Street View photos taken by non-measurement cameras. Specifically, the three-dimensional urban model rapid update method can make use of the texture features and prior knowledge of the existing high-precision digital urban spatial basic information platform results data as much as possible, and takes the invariable building model in the city as the control condition, and adopts The DSM (Digital Surface Model) model is constructed from the street view photos taken from multiple angles to realize the three-dimensional feature update of the urban changing area, which can solve the local image distortion and deviation caused by the non-mapping camera that cannot be handled by the traditional photogrammetry method. It can meet the needs of rapid urban development and has broad application prospects.

The above-mentioned embodiments only represent several embodiments of the present invention, and the descriptions thereof are specific and detailed, but should not be construed as a limitation on the scope of the patent of the present invention. It should be pointed out that for those skilled in the art, without departing from the concept of the present invention, several modifications and improvements can be made, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention should be subject to the appended claims.

Claims (4)

1. a three-dimensional city model rapid update method, is characterized in that, comprises the steps: S101. Collect street view photos from different perspectives of changing areas in a three-dimensional city; S102, analyzing the model of the three-dimensional city, and locating the area of the change area in the model of the three-dimensional city; S103, establishing an image feature library according to the buildings in the model of the three-dimensional city; S104, performing three-dimensional space change detection on the three-dimensional city, and determining a building to be updated and reconstructed; S105, performing feature matching on the street view photos of different viewing angles; S106, accurately fuse the updated and reconstructed building and the background three-dimensional model; The changed area is the area that needs to be updated in the model of the three-dimensional city; the update method is based on the existing three-dimensional city data, and uses the invariants in the existing old data as the source of accuracy for the correction of the updated data. Carry out precision control; The street view photos from different perspectives are obtained by non-measurement professional equipment; The background three-dimensional model is the unchanged area except the changed area in the model of the three-dimensional city; In the step S104, first classify the three-dimensional city, filter out the non-building objects, and then judge the buildings based on the image features, establish a buffer area for the objects that can be confirmed to have changed, and determine that there are no nearby objects. Buildings that change to establish a frame of reference. 2 . The method for rapidly updating a three-dimensional city model according to claim 1 , wherein the non-measurement professional equipment comprises a mobile phone with a camera and a common camera. 3 . 3 . The method for rapidly updating a three-dimensional city model according to claim 2 , wherein the street view photos from different viewing angles include photos of different facades of buildings in the three-dimensional city. 4 . 4. The method for quick update of a three-dimensional city model according to claim 1, wherein in step S105, the dense matching algorithm is accelerated by using the sunshine conditions and the geometric relationship of the buildings in the three-dimensional city to optimize the reconstruction of the dense point cloud, and The local 3D model is further reconstructed.

Images