Building change area recognition method and system based on drone, and electronic equipment

Abstract

The invention provides a method and system for identifying building change areas based on an unmanned aerial vehicle, and an electronic device, comprising: using an unmanned aerial vehicle to collect orthophotos of a target area at different periods, respectively obtaining digital orthophotos and digital surface models of the target area at different periods; performing image segmentation on the digital orthophotos of the target area at different periods to obtain segmented digital orthophotos containing boundaries of all objects, and performing band calculation on the digital surface models of the target area at different periods to obtain digital surface models annotated with surface elevation differences at different periods; superimposing the digital orthophotos containing boundaries of all objects after image segmentation of the target area at different periods, and superimposing the superimposed digital orthophotos with the digital surface models annotated with surface elevation differences at different periods to obtain building change areas, thereby greatly improving the accuracy of identifying building change areas by the unmanned aerial vehicle.

Description

Technical Field

[0001] The present invention relates to the technical field of drone-based building change area recognition, and in particular to a drone-based building change area recognition method and system, and electronic equipment. Background Art

[0002] In recent years, with the development of the drone industry, drone imaging technology has received more and more attention and has gradually been widely used in various fields. At present, there are few methods for identifying changed areas of buildings in drone images, manual analysis is inefficient, and there are large errors.

[0003] In the existing technology, most of them use deep learning models based on semantic segmentation, such as R-CNN, FastR-CNN, FasterR-CNN, MaskR-CNN, U-net and other models for comparative analysis and testing. After selecting a suitable model for training, the AI ​​model is used to directly identify the change results. This method is not combined with the DSM results. The results of AI recognition contain erroneous results of changes in the building facades. This part of the building itself may not have changed, and the color difference caused by the weather environment is recognized. At present, the AI ​​models or algorithms related to surveying and mapping image recognition are basically recognition algorithms for satellite images, that is, the model training data sets are basically from public and accessible satellite images. Due to the large difference in accuracy between satellite images and drone orthophotos (drone orthophotos are generally at the centimeter level), the posture of ground objects is also different (for example, buildings in drone orthophotos are vertically projected, and high-rise buildings in satellite images are tilted). Therefore, most of the existing AI image recognition cannot be directly applied to drone orthophotos. In addition, drone orthophotos are difficult to obtain as data sets for model training due to their high accuracy, strong confidentiality and inability to be disclosed to the public. In addition, due to the low flight altitude of drones and the limited collection range, the cost of collecting them alone as AI model training data sets is relatively high.

[0004] In the application of DSM results in the prior art, the elevation data in the DSM results are usually aggregated and converted into vector data, and then the vector data of the two phases of data are compared in range. However, due to the error in data collection, the range error of the two phases of buildings obtained by simply relying on the aggregation of elevation values ​​is large and the range is irregular, which leads to a large error in the final comparison results.

[0005] Therefore, the prior art needs to be further developed. Summary of the invention

[0006] The purpose of the present invention is to overcome the above-mentioned technical deficiencies and provide a method and system for identifying building change areas based on drones, as well as electronic equipment, to solve the problems existing in the prior art.

[0007] To achieve the above technical objectives, according to a first aspect of the present invention, the present invention provides a method for identifying building change areas based on a drone, comprising:

[0008] S100, using a drone to collect orthophotos of a target area at different times, and obtaining digital orthophotos and digital surface models of the target area at different times according to the orthophotos of the target area at different times;

[0009] S200, performing image segmentation on the digital orthophotos of the target area at different periods to obtain the segmented digital orthophotos containing the boundaries of all objects, and performing band calculation on the digital surface model of the target area at different periods to obtain the digital surface model annotated with the surface elevation difference at different periods;

[0010] S300, superimposing the digital orthophotos containing all object boundaries after image segmentation of the target area at different periods, and superimposing the superimposed digital orthophotos with the digital surface model marked with surface elevation differences at different periods to obtain the changed area of ​​the building.

[0011] Specifically, the S100 includes:

[0012] The orthophotos of the target area taken by the UAV at different times are processed by data processing software to obtain digital orthophotos and digital surface models of the target area at different times.

[0013] Specifically, the S100 further includes:

[0014] The position correction module arranged on the UAV is used to correct the relative position of the target area photographed by the UAV at different times.

[0015] Specifically, the position correction module includes at least one of the following:

[0016] RTK differential system, PPK differential system, control point acquisition.

[0017] Specifically, the digital orthophotos of the target area at different periods are segmented to obtain the segmented digital orthophotos containing the boundaries of all objects, including:

[0018] The SAM model is used to identify the boundaries of all objects in the digital orthophotos of the target area at different periods, and the boundaries of all identified objects are marked, thereby obtaining a digital orthophoto with the boundaries of all objects marked.

[0019] Specifically, the digital surface model is a single-band raster data with geographic coordinate information, in which each pixel value corresponds to the elevation value of a certain point on the surface; the digital surface model annotated with surface elevation differences at different periods is obtained by performing band calculation on the digital surface model of the target area at different periods, including:

[0020] The pixel values ​​of the grids with the same coordinates in the target area at different periods are calculated one by one, so as to obtain a digital surface model with the surface elevation difference at different periods marked.

[0021] Specifically, the digital orthophotos containing all object boundaries after image segmentation of the target area at different times are superimposed, including:

[0022] The same areas in the digital orthophotos with object boundaries marked in the target area at different periods are screened out, and the same areas are removed, so as to obtain the digital orthophotos with object change areas marked.

[0023] Specifically, the superimposed digital orthophoto is superimposed with a digital surface model annotated with surface elevation differences at different periods to obtain the changed area of ​​the building, including:

[0024] The digital orthophoto with the object change area marked is superimposed with the digital surface model with the surface elevation difference at different periods marked, and the area with elevation difference in the object change area is screened out. This area is the change area of ​​the building.

[0025] According to a second aspect of the present invention, there is provided a system for identifying building change areas based on a drone, comprising:

[0026] An acquisition module is arranged on the UAV and is used to collect orthophotos of the target area at different periods, and obtain digital orthophotos and digital surface models of the target area at different periods according to the orthophotos of the target area at different periods;

[0027] Control module: used for performing image segmentation on the digital orthophotos of the target area at different periods to obtain the segmented digital orthophotos containing the boundaries of all objects, and performing band calculation on the digital surface model of the target area at different periods to obtain the digital surface model annotated with the surface elevation difference of different periods; or used for superimposing the digital orthophotos containing the boundaries of all objects after image segmentation of the target area at different periods, and superimposing the superimposed digital orthophotos with the digital surface model annotated with the surface elevation difference of different periods to obtain the changed area of ​​the building.

[0028] According to a third aspect of the present invention, there is provided an electronic device, comprising: a memory; and a processor, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the above-mentioned drone-based building change area identification method is implemented.

[0029] Beneficial effects:

[0030] The present invention collects orthophotos of target areas at different periods by using unmanned aerial vehicles, and respectively obtains digital orthophotos and digital surface models of target areas at different periods; performs image segmentation on the digital orthophotos of target areas at different periods to obtain segmented digital orthophotos containing boundaries of all objects; performs band calculation on the digital surface models of target areas at different periods to obtain digital surface models annotated with surface elevation differences at different periods; superimposes the digital orthophotos containing boundaries of all objects after image segmentation of target areas at different periods; and superimposes the superimposed digital orthophotos with the digital surface models annotated with surface elevation differences at different periods to obtain change areas of buildings; the method using the digital orthophotos and the digital surface models can effectively filter out misidentification caused by weather and illumination changes, improve the efficiency and accuracy of change area detection, greatly improve the accuracy of recognition of building change areas by unmanned aerial vehicles, and greatly improve the intelligence, availability and reliability of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 is a flow chart of a method for identifying building change areas based on a drone provided in a specific embodiment of the present invention;

[0032] Figure 2 1 is a schematic diagram of the system composition of a building change area recognition system based on a drone provided in a specific embodiment of the present invention;

[0033] Figure 3 is a schematic diagram of a building change area identification process based on a drone provided in a specific embodiment of the present invention;

[0034] Figure 4 It is a flow chart of SAM model image segmentation provided in a specific embodiment of the present invention;

[0035] Figure 5 is a schematic diagram of a digital orthophoto image dom-1 of phase 1 provided in a specific embodiment of the present invention;

[0036] Figure 6 is a schematic diagram of a phase 2 digital orthophoto image dom-2 provided in a specific embodiment of the present invention;

[0037] Figure 7It is a schematic diagram of dom-1 provided in a specific embodiment of the present invention using SAM to perform image segmentation and extract building range result FW-1;

[0038] Figure 8 It is a schematic diagram of dom-2 provided in a specific embodiment of the present invention using SAM to perform image segmentation and extract building range results FW-2;

[0039] Fig. 9 is a schematic diagram of a digital surface model dsm-1 of phase 1 provided in a specific embodiment of the present invention;

[0040] Fig.10 is a schematic diagram of a phase 2 digital surface model dsm-2 provided in a specific embodiment of the present invention;

[0041] Fig.11 It is a schematic diagram of the DSM-math obtained by performing difference calculation on the elevation of two-period data using dsm-1 and dsm-2 provided in a specific embodiment of the present invention;

[0042] Fig.12 is a schematic diagram of an image after superimposing FW-1 and FW-2 provided in a specific embodiment of the present invention;

[0043] Fig.13 It is a schematic diagram of the result FW-math marked with a red border after superimposing FW-1 and FW-2 provided in a specific embodiment of the present invention;

[0044] Fig.14 It is a schematic diagram of an image obtained by superimposing the FW-math provided in a specific embodiment of the present invention and the DSM-math of the difference between two periods of DSM;

[0045] Fig.15 It is a schematic diagram of the yellow marked building change area result FW-result obtained by superimposing the FW-math provided in the specific implementation manner of the present invention and the two-phase DSM difference result DSM-math. DETAILED DESCRIPTION

[0046] In order to enable those skilled in the art to better understand the technical solution of the present invention, the technical solution of the present invention is clearly and completely described below in conjunction with the accompanying drawings of the present invention. Based on the embodiments in this application, other similar embodiments obtained by ordinary technicians in this field without making creative work should all fall within the scope of protection of this application. In addition, the directional words mentioned in the following embodiments, such as "up", "down", "left", "right", etc., are only reference to the directions of the accompanying drawings. Therefore, the directional words used are used to illustrate rather than limit the invention.

[0047] The present invention will be further described below in conjunction with the accompanying drawings and preferred embodiments.

[0048] See also Figure 1 This embodiment provides a method for identifying building change areas based on a drone, including:

[0049] S100, using a drone to collect orthophotos of a target area at different times, and obtaining digital orthophotos and digital surface models of the target area at different times according to the orthophotos of the target area at different times;

[0050] Specifically, the S100 includes:

[0051] The orthophotos of the target area taken by the UAV at different times are processed by data processing software to obtain digital orthophotos and digital surface models of the target area at different times.

[0052] It can be understood that Digital Orthophoto Map (DOM) is a digital aerial photograph / remote sensing image (monochrome / color) scanned and processed using a digital elevation model, which is corrected for projection errors pixel by pixel, then mosaicked and clipped according to the map range to generate image data. Digital Surface Model (DSM) refers to a ground elevation model that includes the heights of surface buildings, bridges, and trees. Compared with DEM (digital elevation model), DEM only contains the elevation information of the terrain and does not contain other surface object information. DSM is based on DEM and further covers the elevation of other surface information except the ground.

[0053] Furthermore, data processing software such as Pix4D Mapper and Agisoft Metashape can be used to stitch the two phases of data to generate DOM and DSM respectively. Pix4D Mapper is a professional drone photogrammetry software that can convert photos taken by drones into three-dimensional maps, models, point clouds, digital elevation models (DEM), digital surface models (DSM) and orthophotos (DOM), etc. The software uses efficient parallel processing technology to automatically process image data and extract high-precision terrain and surface information; Agisoft Metashape is also a professional photogrammetry software that can process digital images and generate 3D spatial data, reconstruct 3D models from photos from multiple perspectives, and generate high-density point clouds, digital surface models (DSM), digital elevation models (DEM), orthophotos (DOM) and texture-mapped 3D models, etc.

[0054] Specifically, the S100 further includes:

[0055] The position correction module arranged on the UAV is used to correct the relative position of the target area photographed by the UAV at different times.

[0056] It can be understood that the present invention is to identify the changed areas of the building range of the target area photographed by the drone at different times. In order to ensure the accuracy of the obtained changed area and meet the comparison calculation of the later data, the image collection at different times should be carried out according to the same accuracy requirements, such as the setting of parameters such as camera resolution, flight altitude, overlap rate, etc. The same accuracy requirements can ensure the comparability of the two-phase data and minimize the errors in the horizontal and elevation positions of the two-phase data, which can significantly improve the reliability of the comparative analysis of the two-phase data.

[0057] Specifically, the position correction module includes at least one of the following:

[0058] RTK differential system, PPK differential system, control point acquisition.

[0059] It should be further explained that RTK (Real-time kinematic), also known as carrier phase differential technology, is a differential method for real-time processing of carrier phase observations of two measurement stations. It consists of a base station and a mobile station. The carrier phase collected by the base station is sent to the user receiver to calculate the coordinates and obtain the coordinate increment between the base station and the mobile station. The plane accuracy can reach 1-2 cm. Compared with the ordinary GPS receiver, the accuracy can only reach the meter level. The aircraft equipped with RTK can accurately calculate the precise position of the mobile station, that is, the drone, as long as the base station is set up at the position of the known coordinate point.

[0060] Furthermore, PPK (Post Processed Kinematic), also known as dynamic post-processing technology, is also composed of a base station and a mobile station. The base station receiver and the mobile station receiver that perform synchronous observations observe the satellite's carrier phase and record it in real time, eliminating the intermediate radio transmission link. After the measurement is completed, GPS processing software is used for linear combination, i.e., post-processing process, to form a virtual carrier phase observation value, determine the centimeter-level relative position between the receivers, and then perform coordinate conversion to obtain the coordinates of the mobile station, i.e., the drone, in the local coordinate system;

[0061] Furthermore, if there is no differential system, you can also use manually collected control points to adjust the relative position of the drone shooting. The specific steps are as follows:

[0062] (1) Select control points: Select points with obvious features in the target area as control points. These points should be evenly distributed, cover the entire target area, and be clearly visible in images at different times.

[0063] (2) On-site measurement: Using high-precision measurement equipment such as total stations and GPS receivers, the precise coordinates of control points (including horizontal position and elevation) are measured on-site. These devices can provide sub-meter or even centimeter-level positioning accuracy.

[0064] (3) Marking and recording: Mark the location of control points and record detailed information, including coordinates, environmental characteristics, and measurement time, to ensure the accuracy of the information;

[0065] (4) Image annotation: Accurately mark the positions of control points on the images collected by the drone. This step can be completed through professional image processing or GIS software to ensure that the marked positions correspond to the actual measured positions.

[0066] (5) Data correction: Compare the field-measured coordinates of the control points with the image-annotated coordinates, and use these data to perform geometric correction on the image, including horizontal and elevation correction. This step is usually completed in the later data processing stage and can be performed using methods such as polynomial transformation and collinearity condition equations.

[0067] It is understandable that although the collection of manual control points increases the workload in the early stage, it can significantly improve the positioning accuracy of the image and ensure the reliability of image comparison at different times. Especially in the absence of RTK or PPK system support, it is an effective means to improve data accuracy and comparative analysis quality. The specific method can be selected according to actual conditions. The present invention does not further limit the selection of the above three position correction methods.

[0068] S200, performing image segmentation on the digital orthophotos of the target area at different periods to obtain the segmented digital orthophotos containing the boundaries of all objects, and performing band calculation on the digital surface model of the target area at different periods to obtain the digital surface model annotated with the surface elevation difference at different periods;

[0069] Specifically, the digital orthophotos of the target area at different periods are segmented to obtain the segmented digital orthophotos containing the boundaries of all objects, including:

[0070] The SAM model is used to identify the boundaries of all objects in the digital orthophotos of the target area at different periods, and the boundaries of all identified objects are marked, thereby obtaining a digital orthophoto with the boundaries of all objects marked.

[0071] It is understandable that the SAM model (Segment Anything Model) is the first AI large-scale basic model for image segmentation, which can mark the boundaries of all objects recognized in the picture. The SAM model outputs a single-band raster data, in which the raster value of the recognized object is set to 255, and the raster value of the unrecognized object is set to 0. Then the raster data is converted into vector data through SAM, and the raster data marked with the object boundary is fused with the original digital orthophoto to generate a marked digital orthophoto. Different colors or border styles can be used to highlight the boundaries, which is convenient for intuitive identification and analysis, further improving the accuracy of image segmentation.

[0072] Further, such as Figure 4 As shown, converting raster data into vector data through SAM usually involves the following steps:

[0073] (1) Input the raster data of the DOM of the orthophoto. If the classification of the raster data is not clear enough, you can use binarization to enhance the contrast between the object and the background to ensure clear boundaries.

[0074] (2) Using the SAM model for intelligent recognition, image processing techniques such as Canny edge detection and Sobel operator are used to extract object boundaries from raster data;

[0075] (3) Output raster data of buildings;

[0076] (4) Convert raster data into vector data of buildings. Use GIS software or image processing libraries (such as GDAL / OGR, QGIS, etc.) to convert the extracted boundaries into vector format. This usually involves contour detection and vector conversion to convert pixel boundaries into points, lines or surface features in geographic coordinates.

[0077] Through the above steps, raster data can be converted into vector data format to facilitate spatial analysis in geographic information systems.

[0078] Specifically, the digital surface model is a single-band raster data with geographic coordinate information, in which each pixel value corresponds to the elevation value of a certain point on the surface; the digital surface model annotated with surface elevation differences at different periods is obtained by performing band calculation on the digital surface model of the target area at different periods, including:

[0079] The pixel values ​​of the grids with the same coordinates in the target area at different periods are calculated one by one, so as to obtain a digital surface model with the surface elevation difference at different periods marked.

[0080] It can be understood that the essence of band operation is to perform mathematical operations on the pixel values ​​corresponding to each pixel point. The digital surface model is a single-band raster image product with geographic coordinate information. Its pixel value is the elevation value at the coordinate position of this pixel. The raster pixel values ​​at the same position are subjected to one-to-one difference calculation through the coordinate information to obtain a single-band raster image composed of height differences. Due to errors in data collection, the elevation difference (pixel difference result) can be rounded during band calculation.

[0081] S300, superimposing the digital orthophotos containing all object boundaries after image segmentation of the target area at different periods, and superimposing the superimposed digital orthophotos with the digital surface model marked with surface elevation differences at different periods to obtain the changed area of ​​the building.

[0082] Specifically, the digital orthophotos containing all object boundaries after image segmentation of the target area at different times are superimposed, including:

[0083] The same areas in the digital orthophotos with object boundaries marked in the target area at different periods are screened out, and the same areas are removed, so as to obtain the digital orthophotos with object change areas marked.

[0084] It should be noted here that, from the digital orthophotos with object boundaries marked at different periods, the image parts with the same geographical range are screened out, the changed areas are marked, and digital orthophotos marked with the change range of objects are generated. Different colors can be used for marking, which may include the emergence of new buildings, reduction or increase of vegetation, expansion of roads, etc., and the change range of objects in the target area at different periods can be accurately identified.

[0085] Specifically, the superimposed digital orthophoto is superimposed with a digital surface model annotated with surface elevation differences at different periods to obtain the changed area of ​​the building, including:

[0086] The digital orthophoto with the object change area marked is superimposed with the digital surface model with the surface elevation difference at different periods marked, and the area with elevation difference in the object change area is screened out. This area is the change area of ​​the building.

[0087] It can be understood that superimposing the superimposed digital orthophotos again with the digital surface model annotated with the surface elevation differences of different periods and marking them with different colors not only takes into account the two-dimensional information of the image, but also adds the height and terrain dimensions, making the detection of changed areas more comprehensive and three-dimensional, which helps to accurately distinguish between actual changes and non-actual changes in complex environments, can improve the accuracy and reliability of changed area detection, and provide more stable and higher-quality change detection results for practical applications.

[0088] The working principle of the present invention is described below by means of specific examples:

[0089] See also Figure 3-Figure 15 , by combining with the UAV orthophoto results DOM and DSM data, the change range of the UAV orthophoto buildings in the two periods is extracted. The specific steps are as follows:

[0090] Step 1: UAV orthophoto data collection. To meet the needs of later data comparison calculations, the horizontal and elevation position differences between the two phases of data must be minimized. Therefore, the UAV data collection system must be equipped with an RTK or PPK differential system. If there is no differential system, control points must be collected manually.

[0091] Step 2: Process the drone orthophotos. Use data processing software (such as Pix4D Mapper, Agisoft Metashape, Reconstruction Master, etc.) to stitch the two phases of data together and generate DOM results. Figure 5 As shown in dom-1, Figure 6 dom-2 shown in the figure, and the DSM results generated, such as Fig. 9 As shown in dsm-1, Fig.10 DSM-2 shown;

[0092] Step 3: Process the drone orthophoto data results;

[0093] (1) dom-1 uses the SAM model to segment the image and extract Figure 7 Building scope outcome FW-1 shown;

[0094] (2) dom-2 uses the SAM model to segment the image and extract Figure 8 Building scope outcome FW-2 shown;

[0095] (3) Band calculation, that is, using dsm-1 and dsm-2 to calculate the difference between the elevations of the two phases of data, the following is obtained: Fig.11 The results shown are DSM-math;

[0096] Step 4: FW-1 and FW-2 are superimposed and the same range is eliminated to obtain the following Fig.12 The image of FW-1 and FW-2 is superimposed, and the changed area is marked with a red border, resulting in the following Fig.13 The result FW-math shown, where the red area is the changed area of ​​the object after screening;

[0097] Step 5: Superimpose the image segmentation and screening range FW-math with the two-phase DSM difference result DSM-math, screen the range with difference, and get the following Fig.14The image of FW-math and DSM-math superimposed is shown in yellow, and the change area is marked, as shown in the figure. Fig.15 The result FW-result is shown, where the yellow area is the changed area of ​​the building.

[0098] It should be noted here that the present invention collects orthophotos of target areas at different periods by using unmanned aerial vehicles, and respectively obtains digital orthophotos and digital surface models of target areas at different periods; performs image segmentation on the digital orthophotos of target areas at different periods to obtain segmented digital orthophotos containing boundaries of all objects; performs band calculation on the digital surface models of target areas at different periods to obtain digital surface models annotated with surface elevation differences at different periods; superimposes the digital orthophotos containing boundaries of all objects after image segmentation of target areas at different periods, and superimposes the superimposed digital orthophotos with the digital surface models annotated with surface elevation differences at different periods to obtain change areas of buildings; the method using digital orthophotos and digital surface models can effectively filter out misidentification caused by weather and light changes, improve the efficiency and accuracy of change area detection, greatly improve the accuracy of recognition of building change areas by unmanned aerial vehicles, and greatly improve the intelligence, usability and reliability of the present invention.

[0099] See also Figure 2 , this embodiment provides a building change area recognition system based on a drone, the system comprising:

[0100] Acquisition module 100: is set on the drone, and is used to collect orthophotos of the target area at different periods, and obtain digital orthophotos and digital surface models of the target area at different periods according to the orthophotos of the target area at different periods;

[0101] Control module 200: used to perform image segmentation on the digital orthophotos of the target area at different periods to obtain the segmented digital orthophotos containing the boundaries of all objects, and to perform band calculation on the digital surface model of the target area at different periods to obtain the digital surface model annotated with the surface elevation difference of different periods; or to superimpose the digital orthophotos containing the boundaries of all objects after image segmentation of the target area at different periods, and to superimpose the superimposed digital orthophotos with the digital surface model annotated with the surface elevation difference of different periods to obtain the changed area of ​​the building.

[0102] It should be noted here that the present invention collects orthophotos of target areas at different periods by using unmanned aerial vehicles, and respectively obtains digital orthophotos and digital surface models of target areas at different periods; performs image segmentation on the digital orthophotos of target areas at different periods to obtain segmented digital orthophotos containing boundaries of all objects; performs band calculation on the digital surface models of target areas at different periods to obtain digital surface models annotated with surface elevation differences at different periods; superimposes the digital orthophotos containing boundaries of all objects after image segmentation of target areas at different periods, and superimposes the superimposed digital orthophotos with the digital surface models annotated with surface elevation differences at different periods to obtain change areas of buildings; the method using digital orthophotos and digital surface models can effectively filter out misidentification caused by weather and light changes, improve the efficiency and accuracy of change area detection, greatly improve the accuracy of recognition of building change areas by unmanned aerial vehicles, and greatly improve the intelligence, usability and reliability of the present invention.

[0103] In a preferred embodiment, the present application further provides an electronic device, the electronic device comprising:

[0104] A memory; and a processor, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the method for identifying building change areas based on drones is implemented. The computer device can be broadly a server, a terminal, or any other electronic device with necessary computing and / or processing capabilities. In one embodiment, the computer device may include a processor, a memory, a network interface, a communication interface, etc. connected via a system bus. The processor of the computer device can be used to provide necessary computing, processing and / or control capabilities. The memory of the computer device may include a non-volatile storage medium and an internal memory. An operating system, a computer program, etc. may be stored in or on the non-volatile storage medium. The internal memory may provide an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface and the communication interface of the computer device can be used to connect and communicate with external devices through a network. When the computer program is executed by the processor, the steps of the method of the present invention are performed.

[0105] The present invention may be implemented as a computer-readable storage medium having a computer program stored thereon, which causes the steps of the method of an embodiment of the present invention to be executed when executed by a processor. In one embodiment, the computer program is distributed on a plurality of computer devices or processors coupled to a network so that the computer program is stored, accessed and executed in a distributed manner by one or more computer devices or processors. A single method step / operation, or two or more method steps / operations, may be performed by a single computer device or processor or by two or more computer devices or processors. One or more method steps / operations may be performed by one or more computer devices or processors, and one or more other method steps / operations may be performed by one or more other computer devices or processors. One or more computer devices or processors may perform a single method step / operation, or perform two or more method steps / operations.

[0106] It should be noted here that the present invention collects orthophotos of target areas at different periods by using unmanned aerial vehicles, and respectively obtains digital orthophotos and digital surface models of target areas at different periods; performs image segmentation on the digital orthophotos of target areas at different periods to obtain segmented digital orthophotos containing boundaries of all objects; performs band calculation on the digital surface models of target areas at different periods to obtain digital surface models annotated with surface elevation differences at different periods; superimposes the digital orthophotos containing boundaries of all objects after image segmentation of target areas at different periods, and superimposes the superimposed digital orthophotos with the digital surface models annotated with surface elevation differences at different periods to obtain change areas of buildings; the method using digital orthophotos and digital surface models can effectively filter out misidentification caused by weather and light changes, improve the efficiency and accuracy of change area detection, greatly improve the accuracy of recognition of building change areas by unmanned aerial vehicles, and greatly improve the intelligence, usability and reliability of the present invention.

[0107] The various technical features described above can be combined arbitrarily. Although all possible combinations of these technical features are not described, any combination of these technical features should be considered to be covered by this specification as long as there is no contradiction in such combination.

[0108] The specific implementation of the present invention described above does not constitute a limitation on the protection scope of the present invention. Any other corresponding changes and modifications made based on the technical concept of the present invention should be included in the protection scope of the claims of the present invention.

Claims

1. A method for identifying building change areas based on drones, characterized in that: include: S100, using a drone to collect orthophotos of a target area at different times, and obtaining digital orthophotos and digital surface models of the target area at different times according to the orthophotos of the target area at different times; S200, performing image segmentation on the digital orthophotos of the target area at different periods to obtain the segmented digital orthophotos containing the boundaries of all objects, and performing band calculation on the digital surface model of the target area at different periods to obtain the digital surface model annotated with the surface elevation difference at different periods; S300, superimposing the digital orthophotos containing all object boundaries after image segmentation of the target area at different periods, and superimposing the superimposed digital orthophotos with the digital surface model marked with surface elevation differences at different periods to obtain the changed area of ​​the building; The digital surface model is a single-band raster data with geographic coordinate information, in which each pixel value corresponds to the elevation value of a certain point on the surface; the digital surface model annotated with surface elevation differences at different periods is obtained by performing band calculation on the digital surface model of the target area at different periods, including: The pixel values ​​of the grids with the same coordinates in the target area at different periods are calculated one by one, so as to obtain a digital surface model with the surface elevation difference values ​​at different periods marked; The digital orthophotos containing all object boundaries after image segmentation of the target area at different times are superimposed, including: Screening out the same areas in the digital orthophotos with object boundaries marked in the target area at different periods, and removing the same areas, so as to obtain the digital orthophotos with object change areas marked; The method of superimposing the superimposed digital orthophoto image with the digital surface model annotated with the surface elevation difference of different periods to obtain the changed area of ​​the building includes: The digital orthophoto with the object change area marked is superimposed with the digital surface model with the surface elevation difference at different periods marked, and the area with elevation difference in the object change area is screened out. This area is the change area of ​​the building.

2. The method for identifying building change areas based on drones according to claim 1, characterized in that: The S100 includes: The orthophotos of the target area taken by the UAV at different times are processed by data processing software to obtain digital orthophotos and digital surface models of the target area at different times.

3. The method for identifying building change areas based on drones according to claim 1, characterized in that: The S100 further includes: The position correction module arranged on the UAV is used to correct the relative position of the target area photographed by the UAV at different times.

4. The method for identifying building change areas based on drones according to claim 3 is characterized in that: The position correction module includes at least one of the following: RTK differential system, PPK differential system, control point acquisition.

5. The method for identifying building change areas based on drones according to claim 1, characterized in that: The digital orthophotos of the target area at different times are segmented to obtain the segmented digital orthophotos containing the boundaries of all objects, including: The SAM model is used to identify the boundaries of all objects in the digital orthophotos of the target area at different periods, and the boundaries of all identified objects are marked, thereby obtaining a digital orthophoto with the boundaries of all objects marked.

6. A building change area recognition system based on drone, characterized in that: The method for identifying building change areas based on a drone as described in any one of claims 1 to 5 comprises: An acquisition module (100) is arranged on the drone and is used to collect orthophotos of the target area at different periods, and to respectively acquire digital orthophotos and digital surface models of the target area at different periods based on the orthophotos of the target area at different periods; The control module (200) is used to perform image segmentation on the digital orthophotos of the target area at different periods to obtain the segmented digital orthophotos containing the boundaries of all objects, and to perform band calculation on the digital surface models of the target area at different periods to obtain the digital surface models annotated with the surface elevation differences at different periods; and to superimpose the digital orthophotos containing the boundaries of all objects after the image segmentation of the target area at different periods, and to superimpose the superimposed digital orthophotos with the digital surface models annotated with the surface elevation differences at different periods to obtain the changed area of ​​the building.

7. An electronic device, characterized in that: include: Memory; and a processor, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the method for identifying building change areas based on a drone according to any one of claims 1 to 5 is implemented.

Images