Building change detection method and related equipment based on satellite three-dimensional images

Abstract

The present application discloses a building change detection method based on satellite three-dimensional images and related equipment, and relates to the field of computer technology. The method includes: obtaining a previous phase orthophoto fusion image, a later phase orthophoto fusion image, a previous phase DSM value, and a later phase DSM value; identifying the building outlines in the previous phase orthophoto fusion image and the later phase orthophoto fusion image through a deep learning algorithm; determining the scene of the building outline in the previous phase orthophoto fusion image according to the previous phase DSM value corresponding to the building outline in the previous phase orthophoto fusion image, and determining the scene of the building outline in the later phase orthophoto fusion image according to the later phase DSM value corresponding to the building outline in the later phase orthophoto fusion image; detecting the change type of the building in the previous phase orthophoto fusion image and the later phase orthophoto fusion image according to the scene and position relationship of the building outline in the previous phase orthophoto fusion image and the later phase orthophoto fusion image. The present application can improve the accuracy of change detection.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a building change detection method based on satellite three-dimensional images and related equipment. Background Art

[0002] The sustainable development of cities requires timely understanding of changes in buildings, ground greening, lake distribution, etc. in cities. The traditional method of detecting building changes mainly relies on manually comparing multiple images of the target area to determine whether the building type in the target area has changed. However, the current urban area is large and the scope of detection is wide. The manual comparison method is inefficient and has poor accuracy. Therefore, how to improve the accuracy of building change type detection is an urgent problem to be solved. Summary of the invention

[0003] In view of the above problems, the present application provides a building change detection method and related equipment based on satellite three-dimensional images, which solves the problem of how to improve the accuracy of building change type detection.

[0004] In order to solve the above technical problems, this application proposes the following solutions:

[0005] In the first aspect, the present application provides a building change detection method based on satellite three-dimensional images, the method comprising: obtaining a previous phase orthophoto fusion image, a subsequent phase orthophoto fusion image, a previous phase DSM value, and a subsequent phase DSM value; identifying the building contours in the previous phase orthophoto fusion image and the subsequent phase orthophoto fusion image; determining the scene of the building contour in the previous phase orthophoto fusion image according to the previous phase DSM value corresponding to the building contour in the previous phase orthophoto fusion image, and determining the scene of the building contour in the subsequent phase orthophoto fusion image according to the subsequent phase DSM value corresponding to the building contour in the subsequent phase orthophoto fusion image; detecting the change type of the building in the previous phase orthophoto fusion image and the subsequent phase orthophoto fusion image according to the scene and position relationship of the building contour in the previous phase orthophoto fusion image and the subsequent phase orthophoto fusion image.

[0006] In combination with the first aspect, in a possible implementation method, the building cluster area is determined according to the brightness value of the building contour in the previous phase orthophoto fusion image; the previous phase DSM value of the building cluster area is calculated according to the previous phase DSM value corresponding to each building contour in the building cluster area; when the previous phase DSM value of the building cluster area is greater than a first threshold, it indicates that the scene of the building cluster area is a building cluster; when the previous phase DSM value of the building cluster area is less than or equal to the first threshold, it indicates that the scene of the building cluster area is a bungalow cluster.

[0007] In combination with the first aspect, in another possible implementation method, when the scene is a gathering of buildings, the positional relationship between the outlines of the buildings in the previous phase orthophoto fusion image and the subsequent phase orthophoto fusion image is determined; when there are building outlines in the current phase orthophoto fusion image and there are no building outlines in the subsequent phase orthophoto fusion image, the change type indicated is a reduction in buildings; when there are no building outlines in the current phase orthophoto fusion image and there are building outlines in the subsequent phase orthophoto fusion image, and the outlines of adjacent buildings in the subsequent phase orthophoto fusion image do not exist in the previous phase orthophoto fusion image, the change type indicated is an increase in buildings; when there are no building outlines in the current phase orthophoto fusion image and there are building outlines in the subsequent phase orthophoto fusion image, and there are If the adjacent building outlines of the building outline in the image exist in the previous phase orthophoto fusion image, it indicates that the change type is building expansion; if the building outline exists in both the current phase orthophoto fusion image and the later phase orthophoto fusion image, and the image band standard deviation of the building outline in the previous phase orthophoto fusion image and the later phase orthophoto fusion image is greater than the second threshold, or the difference in brightness value of the building outline in the previous phase orthophoto fusion image and the later phase orthophoto fusion image is greater than the third threshold, it indicates that the change type is building reconstruction; if the building outline exists in both the current phase orthophoto fusion image and the later phase orthophoto fusion image, and the difference between the later phase DSM value corresponding to the building outline and the previous phase DSM value corresponding to the building outline is greater than the fourth threshold, it indicates that the change type is building heightening.

[0008] In combination with the first aspect, in another possible implementation method, when the scene of the building gathering area is a gathering of bungalows, multi-scale segmentation and spectral difference segmentation are performed on the previous phase orthophoto fusion image; the brightness value and adjacent object relationship value of the building outline in the previous phase orthophoto fusion image after the segmentation are determined; when the brightness value of the building outline in the previous phase orthophoto fusion image after the segmentation is greater than the fifth threshold, and the adjacent object relationship value of the building outline in the previous phase orthophoto fusion image after the segmentation is less than the sixth threshold, it indicates that the scene of the building outline in the previous phase orthophoto fusion image after the segmentation is a gathering of bungalows.

[0009] In combination with the first aspect, in another possible implementation method, when the scene is a gathering of bungalows, the positional relationship between the outlines of the buildings in the previous phase orthophoto fusion image and the subsequent phase orthophoto fusion image is determined; when there are building outlines in the current phase orthophoto fusion image and not in the subsequent phase orthophoto fusion image, the change type is indicated as a decrease in bungalows; when there are no building outlines in the current phase orthophoto fusion image and there are building outlines in the subsequent phase orthophoto fusion image, and the outlines of adjacent buildings in the subsequent phase orthophoto fusion image do not exist in the previous phase orthophoto fusion image, the change type is indicated as an increase in bungalows; If there is no building outline in the orthophoto fusion image of the current phase, but there is a building outline in the orthophoto fusion image of the later phase, and the adjacent building outlines of the building outlines in the orthophoto fusion image of the later phase exist in the orthophoto fusion image of the previous phase, it indicates that the change type is the expansion of the bungalow. If there is a building outline in both the orthophoto fusion image of the current phase and the orthophoto fusion image of the later phase, and the image band standard deviation of the building outline in the orthophoto fusion image of the previous phase and the orthophoto fusion image of the later phase is greater than the second threshold, or the brightness value difference of the building outline in the orthophoto fusion image of the previous phase and the orthophoto fusion image of the later phase is greater than the third threshold, it indicates that the change type is the reconstruction of the bungalow.

[0010] In combination with the first aspect, in another possible implementation method, when the scene is a gathering of buildings, multi-scale segmentation is performed on the previous phase orthophoto fusion image. When the absolute value of the difference between the previous phase DSM value corresponding to the first feature and the previous phase DSM value corresponding to the second feature is less than the seventh threshold, it indicates that the first feature is a building, the first feature is an unclassified feature in the previous phase orthophoto fusion image after segmentation, and the second feature is a building in the previous phase orthophoto fusion image after segmentation whose adjacent object relationship value with the first feature is greater than the eighth threshold.

[0011] In combination with the first aspect, in another possible implementation, when the scene is a gathering of buildings, projection difference correction is performed on the building outlines according to the satellite azimuth and incident angle in the orthophoto fusion image of the preceding phase.

[0012] In combination with the first aspect, in another possible implementation, when the difference between the DSM value of the previous phase and the DSM value of the next phase of the building whose type has changed is a first value, it is indicated that the change is a false change.

[0013] In a second aspect, the present application provides a building change detection device based on satellite three-dimensional images, and the building change detection device based on satellite three-dimensional images includes:

[0014] An acquisition module is used to acquire the front phase orthophoto fusion image, the back phase orthophoto fusion image, the front phase DSM value, and the back phase DSM value;

[0015] A recognition module, used for recognizing the building outlines in the front-phase orthophoto fusion image and the back-phase orthophoto fusion image;

[0016] A scene module is used to determine the scene of the building outline in the previous phase orthophoto fusion image according to the previous phase DSM value corresponding to the building outline in the previous phase orthophoto fusion image, and to determine the scene of the building outline in the later phase orthophoto fusion image according to the later phase DSM value corresponding to the building outline in the later phase orthophoto fusion image;

[0017] The detection module is used to detect the change type of the building in the previous phase orthophoto fusion image and the subsequent phase orthophoto fusion image according to the scene and position relationship of the building outline in the previous phase orthophoto fusion image and the subsequent phase orthophoto fusion image.

[0018] In combination with the second aspect, in a possible implementation method, the scene module is specifically used to determine the building cluster area according to the brightness value of the building contour in the previous phase orthophoto fusion image; calculate the previous phase DSM value of the building cluster area according to the previous phase DSM value corresponding to each building contour in the building cluster area; when the previous phase DSM value of the building cluster area is greater than a first threshold, it indicates that the scene of the building cluster area is a building cluster; when the previous phase DSM value of the building cluster area is less than or equal to the first threshold, it indicates that the scene of the building cluster area is a bungalow cluster.

[0019] In combination with the second aspect, in another possible implementation method, the detection module is specifically used to, when the scene is a gathering of buildings, determine the positional relationship between the outlines of buildings in the previous phase orthophoto fusion image and the subsequent phase orthophoto fusion image; when there are building outlines in the current phase orthophoto fusion image and no building outlines in the subsequent phase orthophoto fusion image, the change type indicated is a reduction in buildings; when there are no building outlines in the current phase orthophoto fusion image and there are building outlines in the subsequent phase orthophoto fusion image, and the outlines of adjacent buildings in the subsequent phase orthophoto fusion image do not exist in the previous phase orthophoto fusion image, the change type indicated is an increase in buildings; when there are no building outlines in the current phase orthophoto fusion image and there are building outlines in the subsequent phase orthophoto fusion image, and there are no adjacent building outlines in the subsequent phase orthophoto fusion image If the adjacent building outlines of the building outline in the orthophoto fusion image exist in the previous phase orthophoto fusion image, it indicates that the change type is building expansion; if the building outline exists in both the current phase orthophoto fusion image and the later phase orthophoto fusion image, and the image band standard deviation of the building outline in the previous phase orthophoto fusion image and the later phase orthophoto fusion image is greater than the second threshold, or the difference in brightness value of the building outline in the previous phase orthophoto fusion image and the later phase orthophoto fusion image is greater than the third threshold, it indicates that the change type is building reconstruction; if the building outline exists in both the current phase orthophoto fusion image and the later phase orthophoto fusion image, and the difference between the later phase DSM value corresponding to the building outline and the previous phase DSM value corresponding to the building outline is greater than the fourth threshold, it indicates that the change type is building heightening.

[0020] In combination with the second aspect, in another possible implementation method, the scene module is also used to, when the scene of the building cluster area is a cluster of bungalows, perform multi-scale segmentation and spectral difference segmentation on the previous phase orthophoto fusion image; determine the brightness value and adjacent object relationship value of the building outline in the previous phase orthophoto fusion image after the segmentation; when the brightness value of the building outline in the previous phase orthophoto fusion image after the segmentation is greater than the fifth threshold, and the adjacent object relationship value of the building outline in the previous phase orthophoto fusion image after the segmentation is less than the sixth threshold, it indicates that the scene of the building outline in the previous phase orthophoto fusion image after the segmentation is a cluster of bungalows.

[0021] In combination with the second aspect, in another possible implementation, the detection module is specifically used to, when the scene is a cluster of bungalows, determine the positional relationship between the building outlines in the previous phase orthophoto fusion image and the subsequent phase orthophoto fusion image; when there is a building outline in the current phase orthophoto fusion image and there is no building outline in the subsequent phase orthophoto fusion image, the change type is indicated as a decrease in bungalows; when there is no building outline in the current phase orthophoto fusion image and there is a building outline in the subsequent phase orthophoto fusion image, and the adjacent building outline of the building outline in the subsequent phase orthophoto fusion image does not exist in the previous phase orthophoto fusion image, the change type is indicated as an increase in bungalows; If there is no building outline in the current phase orthophoto fusion image, but there is a building outline in the later phase orthophoto fusion image, and the adjacent building outlines of the building outline in the later phase orthophoto fusion image exist in the previous phase orthophoto fusion image, it indicates that the change type is bungalow expansion. If there is a building outline in both the current phase orthophoto fusion image and the later phase orthophoto fusion image, and the image band standard deviation of the building outline in the previous phase orthophoto fusion image and the later phase orthophoto fusion image is greater than the second threshold, or the brightness value difference of the building outline in the previous phase orthophoto fusion image and the later phase orthophoto fusion image is greater than the third threshold, it indicates that the change type is bungalow reconstruction.

[0022] In combination with the second aspect, in another possible implementation method, the recognition module is also used to, when the scene is a gathering of buildings, perform multi-scale segmentation on the previous phase orthophoto fusion image. When the absolute value of the difference between the previous phase DSM value corresponding to the first feature and the previous phase DSM value corresponding to the second feature is less than a seventh threshold, it indicates that the first feature is a building, the first feature is an unclassified feature in the previous phase orthophoto fusion image after segmentation, and the second feature is a building in the previous phase orthophoto fusion image after segmentation whose adjacent object relationship value with the first feature is greater than an eighth threshold.

[0023] In combination with the second aspect, in another possible implementation, the recognition module is also used to, when the scene is a cluster of buildings, perform projection difference correction on the building outline according to the satellite azimuth and incident angle in the orthophoto fusion image of the previous phase of the building outline.

[0024] In combination with the second aspect, in another possible implementation, the detection module is further configured to indicate that the change is a false change when the difference between the previous phase DSM value and the later phase DSM value of the building whose type has changed is a first value.

[0025] In order to achieve the above-mentioned purpose, according to the third aspect of the present application, a storage medium is provided, which includes a stored program, wherein when the program is running, the device where the storage medium is located is controlled to execute the building change detection method based on satellite three-dimensional images of the above-mentioned first aspect.

[0026] In order to achieve the above-mentioned purpose, according to the fourth aspect of the present application, an electronic device is provided, which includes at least one processor, and at least one memory and a bus connected to the processor; wherein the processor and the memory communicate with each other through the bus; the processor is used to call the program instructions in the memory to execute the building change detection method based on satellite three-dimensional images of the above-mentioned first aspect.

[0027] By means of the above technical solution, the technical solution provided by this application has at least the following advantages:

[0028] This application provides a building change detection method based on satellite three-dimensional images and related equipment. This application combines deep learning and object-oriented methods to distinguish between building clusters and bungalow clusters, and can quickly and accurately obtain building change data in a large area, providing timely support for urban planning, resource monitoring and other work. In addition, the building change results divide the change types into five categories: new addition, reduction, reconstruction, expansion, and floor height increase, which enriches the building change types and is closer to the needs of building change monitoring.

[0029] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present application. Also, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0031] Figure 1 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application is shown;

[0032] Figure 2 A schematic diagram of a flow chart of a building change detection method based on satellite three-dimensional images provided in an embodiment of the present application is shown;

[0033] Figure 3 A schematic diagram of a specific process of a building change detection method based on satellite three-dimensional images provided in an embodiment of the present application is shown;

[0034] Figure 4 A schematic diagram of a building gathering area provided in an embodiment of the present application is shown;

[0035] Figure 5A schematic diagram of a bungalow gathering area provided in an embodiment of the present application is shown;

[0036] Figure 6 A schematic diagram of a single building before and after supplementation provided by an embodiment of the present application is shown;

[0037] Figure 7 A schematic diagram of a building cluster area before and after supplementation provided by an embodiment of the present application is shown;

[0038] Figure 8 A schematic diagram of a bungalow cluster area before segmentation provided in an embodiment of the present application is shown;

[0039] Fig. 9 A schematic diagram of a bungalow cluster area after segmentation provided by an embodiment of the present application is shown;

[0040] Fig.10 A schematic diagram of a building before regularization provided by an embodiment of the present application is shown;

[0041] Fig.11 A schematic diagram of a building after regularization provided by an embodiment of the present application is shown;

[0042] Fig.12 A schematic structural diagram of a building change detection device based on satellite three-dimensional images provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0043] The exemplary embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present application and to fully convey the scope of the present application to those skilled in the art.

[0044] In the embodiments of the present application, the terms "first" and "second" do not have a logical or temporal dependency, nor do they limit the quantity and execution order. It should also be understood that although the following description uses the terms first, second, etc. to describe various elements, these elements should not be limited by the terms. These terms are only used to distinguish one element from another element.

[0045] In the embodiments of the present application, the term "at least one" means one or more, and the term "plurality" means two or more.

[0046] It should also be understood that the term "if" may be interpreted as "when" or "upon" or "in response to determining" or "in response to detecting." Similarly, the phrase "if it is determined that ..." or "if [a stated condition or event] is detected" may be interpreted as "upon determining that ..." or "in response to determining that ..." or "upon detecting [a stated condition or event]" or "in response to detecting [a stated condition or event]," depending on the context.

[0047] The embodiment of the present application provides a building change detection device based on satellite 3D images, which can be used to execute the above-mentioned building change detection method based on satellite 3D images. Optionally, the building change detection device based on satellite 3D images can be an electronic device with data processing capabilities, or a functional module in the electronic device, which is not limited.

[0048] For example, the electronic device may be a server, which may be a single server, or a server cluster consisting of multiple servers. For another example, the electronic device may be a mobile phone, a tablet computer, a desktop, a laptop, a handheld computer, a notebook computer, an ultra-mobile personal computer (UMPC), a netbook, and a cellular phone, a personal digital assistant (PDA), an augmented reality (AR), a virtual reality (VR) device and other terminal devices. For another example, the electronic device may also be a recording device, a video surveillance device and other devices. The present application does not impose any special restrictions on the specific form of the electronic device.

[0049] The following takes the building change detection device based on satellite three-dimensional image as an electronic device as an example. Figure 1 As shown, Figure 1 The present application provides a hardware structure of an electronic device 100.

[0050] like Figure 1 As shown, the electronic device 100 includes a processor 110 , a communication circuit 120 and a communication interface 130 .

[0051] Optionally, the electronic device 100 may further include a memory 140 . The processor 110 , the memory 140 and the communication interface 130 may be connected via a communication line 120 .

[0052] The processor 110 may be a central processing unit (CPU), a general-purpose processor, a network processor (NP), a digital signal processor (DSP), a microprocessor, a microcontroller, a programmable logic device (PLD), or any combination thereof. The processor 110 may also be any other device having a processing function, such as a circuit, a device, or a software module, without limitation.

[0053] In one example, the processor 110 may include one or more CPUs, such as Figure 1 CPU0 and CPU1 in.

[0054] As an optional implementation, the electronic device 100 includes multiple processors, for example, in addition to the processor 110, it may also include a processor 170. The communication line 120 is used to transmit information between the components included in the electronic device 100.

[0055] The communication interface 130 is used to communicate with other devices or other communication networks. The other communication networks may be Ethernet, Radio Access Network (RAN), Wireless Local Area Networks (WLAN), etc. The communication interface 130 may be a module, a circuit, a transceiver or any device capable of achieving communication.

[0056] The memory 140 is used to store instructions, where the instructions may be computer programs.

[0057] Among them, the memory 140 can be a read-only memory (ROM) or other types of static storage devices that can store static information and / or instructions, or a random access memory (RAM) or other types of dynamic storage devices that can store information and / or instructions, or an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical disc, laser disc, optical disc, digital versatile disc, Blu-ray disc, etc.), magnetic disk storage medium or other magnetic storage devices, etc., without limitation.

[0058] It should be noted that the memory 140 may exist independently of the processor 110 or may be integrated with the processor 110. The memory 140 may be used to store instructions or program codes or some data, etc. The memory 140 may be located inside the electronic device 100 or outside the electronic device 100, without limitation.

[0059] The processor 110 is used to execute the instructions stored in the memory 140 to implement the communication method provided in the following embodiments of the present application. For example, when the electronic device 100 is a terminal or a chip in a terminal, the processor 110 can execute the instructions stored in the memory 140 to implement the steps performed by the transmitting end in the following embodiments of the present application.

[0060] As an optional implementation, the electronic device 100 further includes an output device 150 and an input device 160. The output device 150 may be a device such as a display screen, a speaker, etc. that can output data of the electronic device 100 to a user. The input device 160 may be a device such as a keyboard, a mouse, a microphone, or a joystick that can input data to the electronic device 100.

[0061] It should be pointed out that Figure 1 The structure shown in the figure does not constitute a limitation on the computing device, except Figure 1 In addition to the components shown, the computing device may include more or fewer components than shown, or combine certain components, or arrange the components differently.

[0062] The building change detection device based on satellite three-dimensional images and the application scenarios described in the embodiments of the present application are intended to more clearly illustrate the technical solutions of the embodiments of the present application, and do not constitute a limitation on the technical solutions provided in the embodiments of the present application. Ordinary technicians in this field can know that with the evolution of the building change detection device based on satellite three-dimensional images and the emergence of new business scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0063] Next, the building change detection method based on satellite three-dimensional images is described in detail with reference to the accompanying drawings. Figure 2 A schematic flow chart of a building change detection method based on satellite three-dimensional images provided in this application. Figure 3 The present invention provides a specific flow chart of a building change detection method based on satellite 3D images. Figure 1 The building change detection device based on satellite three-dimensional imagery of the hardware structure shown specifically comprises the following steps:

[0064] Step 210: Obtain the front-phase orthophoto fusion image, the back-phase orthophoto fusion image, the front-phase DSM value, and the back-phase DSM value.

[0065] Orthophoto fusion images refer to high-resolution image data with multiple source data information formed by fusing the information of different images after orthophoto correction, color correction, geometric correction and other processing of image data taken at different angles or at different times using a specific fusion algorithm. Orthophoto fusion images can improve the spatial and temporal resolution of images, making the image data more visual and valuable for application.

[0066] From the above, it can be seen that the front-phase orthophoto fusion image and the back-phase orthophoto fusion image of the present application cannot be directly obtained by shooting. Therefore, in the embodiment of the present application, satellite stereo image pair image data of the target area is obtained. The satellite stereo image pair image data is two remote sensing images taken by a satellite at different times of the same target area. In the embodiment of the present application, the remote sensing image taken at the first moment is called the front-phase remote sensing image, and the remote sensing image taken at the second moment is called the back-phase remote sensing image.

[0067] Furthermore, the satellite stereo image data is processed by atmospheric correction, orthorectification, data fusion, etc. to form a high-resolution orthorectified fusion image, namely, the front phase orthorectified fusion image and the back phase orthorectified fusion image.

[0068] In order to detect more accurate and comprehensive building changes, for example, building heightening, etc. This application also needs to generate digital surface model (DSM) data of the target area based on the satellite stereo image pair image data. For example, the Match-T image matching technology can be used to select applicable strategies and mathematical models to automatically extract the DSM values ​​of various objects in the target area. In the embodiments of the present application, the specific implementation method of generating DSM data of the target area based on satellite stereo image pair image data is not limited, and the above is only used as an example for explanation.

[0069] The DSM data includes the elevation values ​​of the surface buildings, bridges, trees and other objects in the target area. Since the satellite stereo image pair image data of the present application refers to the front-phase remote sensing image and the back-phase remote sensing image of the same target area. Therefore, the present application can obtain the front-phase DSM value and the back-phase DSM value of each object in the target area based on the satellite stereo image pair image data.

[0070] Since the orthophoto fusion images of the front phase, the orthophoto fusion images of the back phase, the DSM values ​​of the front phase, and the DSM values ​​of the back phase of the present application are all obtained through the same satellite stereo image data, the orthophoto fusion images and DSM values ​​of the present application are of the same source. Since the two-dimensional information (the orthophoto fusion images of the front phase, the orthophoto fusion images of the back phase) and the three-dimensional information (the DSM values ​​of the front phase, the DSM values ​​of the back phase) used in the present application are of the same source, it is possible to avoid the problem of phase errors or difficulty in obtaining phase data due to different sources of the two-dimensional information and the three-dimensional information, thereby leading to the problem of inaccurate detection of building changes.

[0071] Step 220: Identify the building outlines in the previous phase orthophoto fusion image and the next phase orthophoto fusion image.

[0072] After the front-phase orthophoto fusion image and the back-phase orthophoto fusion image are obtained in step 210 , the building outlines in the front-phase orthophoto fusion image and the back-phase orthophoto fusion image are identified by a deep learning algorithm.

[0073] For example, a convolutional neural network model is constructed to extract building contours; the constructed convolutional neural network model is used to extract the building position and reference contours in the front-phase orthophoto fusion image and the back-phase orthophoto fusion image to obtain the building mask image corresponding to the building; based on the combination of linear features and image segmentation algorithm, the building mask image is processed to obtain the corresponding line segment information, corner point information and orientation information; using the line segment information, corner point information and orientation information corresponding to the building, the building position and the reference contour are optimized and contour corrected on the ground object vector to obtain the revised building contour.

[0074] In the embodiment of the present application, the specific implementation method of identifying the building outlines in the front-phase orthophoto fusion image and the back-phase orthophoto fusion image is not limited, and the above is only used as an example for explanation.

[0075] Step 230: determine the scene of the building outline in the previous phase orthophoto fusion image according to the previous phase DSM value corresponding to the building outline in the previous phase orthophoto fusion image, and determine the scene of the building outline in the subsequent phase orthophoto fusion image according to the subsequent phase DSM value corresponding to the building outline in the subsequent phase orthophoto fusion image.

[0076] First, based on the previous phase DSM value corresponding to the previous phase orthophoto fusion image obtained in step 210, determine the previous phase DSM value corresponding to each building outline in the previous phase orthophoto fusion image identified in step 220. Specifically, in step 210, the previous phase DSM value of each pixel in the previous phase orthophoto fusion image can be obtained. Obtain the previous phase DSM values ​​of the pixels included in each building outline in the previous phase orthophoto fusion image. The average of the previous phase DSM values ​​of the pixels included in each building outline is taken as the previous phase DSM value of the building outline. Similarly, based on the previous phase DSM value corresponding to the previous phase orthophoto fusion image obtained in step 210, determine the subsequent phase DSM value corresponding to each building outline in the subsequent phase orthophoto fusion image identified in step 220.

[0077] Furthermore, since there is a great difference in brightness between buildings and other objects (such as trees, bridges, etc.) in the orthophoto fusion image, in the embodiment of the present application, the building clustering area in the previous phase of the orthophoto fusion image is determined according to the brightness value of the outline of each building in the previous phase of the orthophoto fusion image, and the building clustering area in the later phase of the orthophoto fusion image is determined according to the brightness value of the outline of each building in the later phase of the orthophoto fusion image.

[0078] Take the previous phase orthophoto fusion image as an example. First, obtain the spectral values ​​of the green band, red band and blue band of each pixel in the outline of each building from the previous phase orthophoto fusion image, and determine the brightness value Bri of the pixel according to the spectral values ​​of the green band, red band and blue band. Among them, DN green DN red DN blue are the spectral values ​​of the green band, red band and blue band of the pixel point respectively. Furthermore, the average brightness value of all pixels in the building outline is taken as the brightness value of the building outline. Secondly, the building outlines with brightness values ​​greater than the brightness threshold in the previous phase orthophoto fusion image are extracted, and these building outlines are merged, and the merged area is determined as the building cluster area.

[0079] After determining the building cluster area in the orthophoto fusion image, it is necessary to further determine whether the building cluster area is a building cluster area or a bungalow cluster area, so that when performing subsequent building change detection, more accurate detection can be performed based on the characteristics of buildings and bungalows. It should be noted that in this application, the scene of the building cluster area is not limited. The scene in this application: the building cluster area and the bungalow cluster area are only explained as an example. For example, the scene of the building cluster area can also be an exhibition hall cluster area, a school cluster area, etc.

[0080] In one embodiment, the pre-phase DSM value of the building cluster area is calculated based on the pre-phase DSM value corresponding to each building outline in the building cluster area. The specific implementation method of calculating the pre-phase DSM value of the building cluster area is the same as the specific implementation method of calculating the pre-phase DSM value corresponding to the building outline. That is, the mean of the pre-phase DSM values ​​corresponding to all building outlines in the building cluster area is used as the pre-phase DSM value of the building cluster area. In the embodiment of the present application, the specific implementation method of calculating the pre-phase DSM value of the building cluster area is not specifically limited, and the above is only explained as an example. The mean square value of the pre-phase DSM values ​​corresponding to all building outlines in the building cluster area can also be used as the pre-phase DSM value of the building cluster area.

[0081] When the previous phase DSM value of the building cluster area is greater than the first threshold, it indicates that the scene of the building cluster area is a building cluster; when the previous phase DSM value of the building cluster area is less than or equal to the first threshold, it indicates that the scene of the building cluster area is a bungalow cluster. Figure 4 A schematic diagram of a building cluster area is provided. Figure 5 A schematic diagram of a bungalow cluster area is provided.

[0082] The calculation method of the post-phase DSM value of the building cluster area in the post-phase orthophoto fusion image and the determination method of the scene are the same as those of the pre-phase orthophoto image, which will not be repeated here.

[0083] When the scene is a building cluster, due to the complex roof structure of some buildings or the presence of large buildings in the building cluster area, the building outline extracted by step 220 will be incomplete. The DSM value reflects the height of the ground surface including buildings, structures, vegetation, etc., and can fully identify the building. Therefore, after determining the building cluster area, the present application supplements the incomplete building outlines in the building cluster area according to the DSM.

[0084] In one embodiment, the previous phase orthophoto fusion image is segmented at multiple scales. In the embodiment of the present application, the previous phase orthophoto fusion image is segmented at multiple scales based on the regional merging algorithm with the minimum regional heterogeneity. That is, the single pixel points in the previous phase orthophoto fusion image are gradually merged into smaller image objects, and then the smaller image objects are gradually merged into larger image objects to complete the final segmentation.

[0085] After completing multi-scale segmentation of the previous phase orthophoto fusion image, the adjacent object relationship is used to determine whether the unclassified ground objects (first ground objects) in the segmented previous phase orthophoto fusion image are adjacent to the buildings that have been identified in the previous phase orthophoto fusion image.

[0086] In one embodiment, the adjacent object relationship between two objects in the orthophoto fusion image is represented by the boundary overlap. The boundary overlap rel is used to represent the boundary overlap ratio of the two objects. That is: Among them, x is the unclassified ground object (the first ground object) in the segmented previous phase orthophoto fusion image, y is the identified building in the previous phase orthophoto fusion image, and e is the x,y represents the overlap length of feature x and feature y, p x Indicates the perimeter of the feature x. When the adjacent object relationship value between the building identified in the segmented previous phase orthophoto fusion image and the first feature is greater than the eighth threshold, it indicates that the building is adjacent to the first feature.

[0087] After the building adjacent to the first feature is detected, the first feature is further judged based on the previous phase DSM values ​​of the first feature and the adjacent building. When the absolute value of the difference between the previous phase DSM value corresponding to the first feature and the previous phase DSM value corresponding to the second feature is less than the seventh threshold, it indicates that the first feature and the adjacent building are at the same or similar height. Since the heights of different vertices of the same building will not differ too much, in the embodiment of the present application, the first feature that is at the same or similar height as the adjacent building is determined to be a building. Since the first feature and the adjacent building are at the same or similar height, the two features are still in an adjacent relationship. Therefore, it can be considered that the first feature is part of the adjacent building, thereby completing the building outline. Figure 6 A schematic diagram of a single building before and after supplementation is provided. Figure 7 A schematic diagram of a building cluster before and after supplementation is provided.

[0088] In addition, when the scene in the building cluster area is a cluster of buildings, the different image shooting angles will cause the buildings to swing at different angles in the image, resulting in more pseudo-changes in the building change recognition results in the previous and next phases. Therefore, the present application performs projection difference correction on the building outline according to the satellite azimuth and incident angle of the building outline in the previous phase orthophoto fusion image, so that the top outline and bottom outline of the building in the previous phase orthophoto fusion image and the subsequent phase orthophoto fusion image coincide, thereby reducing the pseudo-changes of buildings in the building cluster area caused by projection differences.

[0089] In one embodiment, the satellite azimuth and the incident angle of the preceding orthophoto fusion image and the succeeding orthophoto fusion image are obtained. The moving direction of the top outline of the building in the preceding orthophoto fusion image or the succeeding orthophoto fusion image is determined according to the satellite azimuth of the preceding orthophoto fusion image or the succeeding orthophoto fusion image. The moving distance of the top outline of the building in the preceding orthophoto fusion image or the succeeding orthophoto fusion image is determined according to the incident angle of the building in the preceding orthophoto fusion image or the succeeding orthophoto fusion image.

[0090] For example, the height of a single building in the previous phase orthophoto fusion image or the next phase orthophoto fusion image is recorded as H, the incident angle is recorded as θ, and the satellite azimuth is ω. Then the moving distance of the top outline of the building in the previous phase or the next phase orthophoto fusion image is L = tan (θ) * H, and the moving direction of the top outline of the building in the previous phase or the next phase orthophoto fusion image is k = ω + 180°. When the value of k is greater than 360°, the value after subtracting 360° from k is used as the moving direction of the top outline of the building in the previous phase or the next phase orthophoto fusion image.

[0091] In the orthophoto fusion image, the building outline can be composed of one mosaic tile or two or more mosaic tiles. If a single building in the orthophoto fusion image spans two or more mosaic tiles, the satellite azimuth and incidence angle corresponding to which mosaic tile is to be moved is determined according to the principle of area dominance. That is, the top outline of the single building is moved according to the satellite azimuth and incidence angle corresponding to which mosaic tile the single building falls within has the largest area.

[0092] On the other hand, when the scene of the building cluster area is a cluster of bungalows, due to the characteristics of bungalows, for example, a bungalow is composed of several rooms, or there is a yard in the middle of a bungalow, some open spaces that are not yards are also identified as buildings in the bungalow cluster area. Therefore, it is necessary to further determine the scene of determining the outline of the building in the bungalow cluster area in the above step. Figure 8 A schematic diagram of a bungalow cluster area before segmentation is provided. Fig. 9 A schematic diagram of the segmentation of a bungalow cluster area is provided.

[0093] First, the orthophoto fusion image is subjected to multi-scale segmentation and spectral difference segmentation. Specifically, the orthophoto fusion image is subjected to multi-scale segmentation so that some unsegmented spots are segmented at a smaller scale. Furthermore, in order to avoid excessive segmentation of the orthophoto fusion image caused by multi-scale segmentation, the method of spectral difference segmentation is used to merge adjacent small spots with consistent spectral performance. Spectral difference segmentation is based on the spectral difference between different bands in the orthophoto fusion image. Cutting can obtain a more uniform and consistent image object. This method is suitable for the extraction of objects with obvious spectral characteristics and large differences. For example, the spectral characteristics between the building and the open space in the present application are obviously different, and spectral difference segmentation can be used.

[0094] Further, determine the brightness value and adjacent object relationship value of the building outline in the segmented previous phase orthophoto fusion image. The brightness value and adjacent object relationship value of the building have been described in detail in this step and will not be repeated here. When the brightness value of the building outline in the segmented previous phase orthophoto fusion image is greater than the fifth threshold value, and the adjacent object relationship value of the building outline in the segmented previous phase orthophoto fusion image is less than the sixth threshold value, it indicates that the scene of the building outline in the segmented previous phase orthophoto fusion image is a cluster of bungalows.

[0095] In another embodiment, the present application can classify the building cluster areas by introducing preset background vectors of building cluster areas and bungalow cluster areas.

[0096] Step 240: Detect the change type of the building in the previous phase orthophoto fusion image and the next phase orthophoto fusion image according to the scene and position relationship of the building outline in the previous phase orthophoto fusion image and the next phase orthophoto fusion image.

[0097] When the scene of the building gathering area is a gathering of buildings, determine the positional relationship of the same building outline in the previous phase orthophoto fusion image and the next phase orthophoto fusion image. When the building outline exists in the current phase orthophoto fusion image and does not exist in the next phase orthophoto fusion image, the change type is indicated as a reduction of buildings. If the building outline does not exist in the current phase orthophoto fusion image and exists in the next phase orthophoto fusion image, then further analyze whether the adjacent buildings of the building in the next phase orthophoto fusion image exist in the previous phase orthophoto fusion image based on the adjacent object relationship. When the adjacent building outline of the building outline in the next phase orthophoto fusion image does not exist in the previous phase orthophoto fusion image, the change type is indicated as an increase of buildings. If the building outline does not exist in the current phase orthophoto fusion image, the building outline exists in the next phase orthophoto fusion image, and the adjacent building outline of the building outline in the next phase orthophoto fusion image exists in the previous phase orthophoto fusion image, the change type is indicated as a building expansion.

[0098] When there are building outlines in both the current phase orthophoto fusion image and the next phase orthophoto fusion image, and the image band standard deviation of the building outline in the previous phase orthophoto fusion image and the next phase orthophoto fusion image is greater than the second threshold, or the brightness difference of the building outline in the previous phase orthophoto fusion image and the next phase orthophoto fusion image is greater than the third threshold, it indicates that the change type is building reconstruction. Calculate the image band standard deviation of the building outline in the orthophoto fusion image, std green 、std red 、std blue are the mean standard deviations of the green, red, and blue bands, respectively.

[0099] When there are building outlines in both the current phase orthophoto fusion image and the later phase orthophoto fusion image, and the difference between the later phase DSM value corresponding to the building outline and the previous phase DSM value corresponding to the building outline is greater than the fourth threshold, it indicates that the change type is building heightening.

[0100] When the scene of the building cluster area is bungalow clustering, the same applies to the scene of the building cluster area is building clustering. First, determine the positional relationship of the building outlines in the previous phase orthophoto fusion image and the next phase orthophoto fusion image. When there is a building outline in the current phase orthophoto fusion image and no building outline in the next phase orthophoto fusion image, the indication change type is bungalow reduction. When there is no building outline in the current phase orthophoto fusion image, there is a building outline in the next phase orthophoto fusion image, and the adjacent building outline of the building outline in the next phase orthophoto fusion image does not exist in the previous phase orthophoto fusion image, the indication change type is bungalow increase. When there is no building outline in the current phase orthophoto fusion image, there is a building outline in the next phase orthophoto fusion image, and the adjacent building outline of the building outline in the next phase orthophoto fusion image exists in the previous phase orthophoto fusion image, the indication change type is bungalow expansion. When there are building outlines in both the current phase and the later phase orthophoto fusion images, and the image band standard deviation of the building outlines in the previous phase and the later phase is greater than the second threshold, or the brightness value difference of the building outlines in the previous phase and the later phase is greater than the third threshold, it indicates that the change type is bungalow reconstruction.

[0101] It should be noted that in the implementation of the present application, the threshold of the image band standard deviation, the threshold of the brightness value, etc. are determined by the sample extreme value method. That is, Q = maxormin (k1, k2, ..., k n ). Where Q represents the determined threshold, k1, k2,…, k n Indicates the value of sample 1, sample 2, ..., sample n on a certain feature, max() represents the maximum value function, and min() represents the minimum value function. Sample extreme value method determines the threshold requirement. When selecting samples, select samples close to the critical point between the target and the non-target.

[0102] In order to avoid some pseudo-changes, for example, a certain area is an open space in the previous phase orthophoto fusion image, and a parking lot in the next phase orthophoto fusion image. At this time, due to the vehicles parked in the parking lot, it will cause recognition errors, and the vehicles will be identified as buildings, and it is considered that the change type in the area is the increase of bungalows. Therefore, the present application makes further judgments based on the previous phase DSM value and the next phase DSM value of the building whose type has changed. When the difference between the previous phase DSM value and the next phase DSM value of the building whose type has changed is the first value, it indicates that the change is a pseudo-change.

[0103] In order to make the detection of building type changes more visible, the present application further processes the changed building outlines based on the above steps.

[0104] In one implementation, firstly, the internal holes of the changed building are removed, then the building in the orthophoto fusion image is polished by an open operation to remove the burrs of the building in the orthophoto fusion image, and the building in the orthophoto fusion image is filled by a closed operation, so that the boundary of the building is smoother. The vector nodes are thinned by a vector simplification algorithm, that is, the boundary of the building is straightened, so as to achieve the effect of regularizing the boundary and reducing the amount of data. Fig.10 A schematic diagram of a building before regularization is provided. Fig.11 A schematic diagram of a building after regularization is provided.

[0105] In summary, this application combines deep learning and object-oriented methods to distinguish between building clusters and bungalow clusters, and can quickly and accurately obtain building change data in a large area, providing timely support for urban planning, resource monitoring, and other work. In addition, the building change results divide the change types into five categories: new addition, reduction, reconstruction, expansion, and floor height increase, which enriches the building change types and is closer to the needs of building change monitoring.

[0106] In addition, this application proposes corresponding solutions to interference factors in building change detection, such as incomplete building identification and poor projection. When detecting building change types based on solving the above-mentioned interference factors, the accuracy can be further improved.

[0107] It is understandable that in order to implement the functions in the above embodiments, the computer device includes hardware structures and / or software modules corresponding to the execution of each function. Those skilled in the art should easily realize that, in combination with the units and method steps of each example described in the embodiments disclosed in this application, the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in the form of hardware or computer software driving hardware depends on the specific application scenario and design constraints of the technical solution.

[0108] Furthermore, as a response to the above Figure 2 The implementation of the method embodiment shown in the figure, the embodiment of the present application provides a building change detection device based on satellite 3D images. The embodiment of the device corresponds to the aforementioned method embodiment. For ease of reading, this embodiment will not repeat the details of the aforementioned method embodiment one by one, but it should be clear that the device in this embodiment can correspond to all the contents of the aforementioned method embodiment. Specifically, Fig.12 As shown, the building change detection device 1200 based on satellite three-dimensional images includes:

[0109] An acquisition module 1210 is used to acquire a front-phase orthophoto fusion image, a rear-phase orthophoto fusion image, a front-phase DSM value, and a rear-phase DSM value;

[0110] The recognition module 1220 is used to recognize the building outlines in the front-phase orthophoto fusion image and the back-phase orthophoto fusion image;

[0111] The scene module 1230 is used to determine the scene of the building outline in the previous phase orthophoto fusion image according to the previous phase DSM value corresponding to the building outline in the previous phase orthophoto fusion image, and determine the scene of the building outline in the later phase orthophoto fusion image according to the later phase DSM value corresponding to the building outline in the later phase orthophoto fusion image;

[0112] The detection module 1240 is used to detect the change type of the building in the previous phase orthophoto fusion image and the next phase orthophoto fusion image according to the scene and position relationship of the building outline in the previous phase orthophoto fusion image and the next phase orthophoto fusion image.

[0113] Further, such as Fig.12 As shown, the scene module 1230 is specifically used to determine the building cluster area according to the brightness value of the building contour in the previous phase orthophoto fusion image; calculate the previous phase DSM value of the building cluster area according to the previous phase DSM value corresponding to each building contour in the building cluster area; when the previous phase DSM value of the building cluster area is greater than a first threshold value, it indicates that the scene of the building cluster area is a building cluster; when the previous phase DSM value of the building cluster area is less than or equal to the first threshold value, it indicates that the scene of the building cluster area is a bungalow cluster.

[0114] Further, such as Fig.12As shown, the detection module 1240 is specifically used to, when the scene is a gathering of buildings, determine the positional relationship between the outlines of buildings in the previous phase orthophoto fusion image and the subsequent phase orthophoto fusion image; when there are building outlines in the current phase orthophoto fusion image and there are no building outlines in the subsequent phase orthophoto fusion image, the change type is indicated as a reduction in buildings; when there are no building outlines in the current phase orthophoto fusion image, there are building outlines in the subsequent phase orthophoto fusion image, and the outlines of adjacent buildings in the subsequent phase orthophoto fusion image do not exist in the previous phase orthophoto fusion image, the change type is indicated as an increase in buildings; when there are no building outlines in the current phase orthophoto fusion image, there are building outlines in the subsequent phase orthophoto fusion image, and there are building outlines in the subsequent phase orthophoto fusion image If the adjacent building outlines of the building outline exist in the previous phase orthophoto fusion image, it indicates that the change type is building expansion; if the building outline exists in both the current phase orthophoto fusion image and the later phase orthophoto fusion image, and the image band standard deviation of the building outline in the previous phase orthophoto fusion image and the later phase orthophoto fusion image is greater than the second threshold, or the brightness value difference of the building outline in the previous phase orthophoto fusion image and the later phase orthophoto fusion image is greater than the third threshold, it indicates that the change type is building reconstruction; if the building outline exists in both the current phase orthophoto fusion image and the later phase orthophoto fusion image, and the difference between the later phase DSM value corresponding to the building outline and the previous phase DSM value corresponding to the building outline is greater than the fourth threshold, it indicates that the change type is building heightening.

[0115] Further, such as Fig.12 As shown, the scene module 1230 is also used to, when the scene of the building gathering area is a gathering of bungalows, perform multi-scale segmentation and spectral difference segmentation on the previous phase orthophoto fusion image; determine the brightness value and adjacent object relationship value of the building outline in the previous phase orthophoto fusion image after segmentation; when the brightness value of the building outline in the previous phase orthophoto fusion image after segmentation is greater than the fifth threshold, and the adjacent object relationship value of the building outline in the previous phase orthophoto fusion image after segmentation is less than the sixth threshold, it indicates that the scene of the building outline in the previous phase orthophoto fusion image after segmentation is a gathering of bungalows.

[0116] Further, such as Fig.12As shown, the detection module 1240 is specifically used to, when the scene is a gathering of bungalows, determine the positional relationship between the building outlines in the previous phase orthophoto fusion image and the subsequent phase orthophoto fusion image; when there are building outlines in the current phase orthophoto fusion image and there are no building outlines in the subsequent phase orthophoto fusion image, it indicates that the change type is a decrease in bungalows; when there are no building outlines in the current phase orthophoto fusion image and there are building outlines in the subsequent phase orthophoto fusion image, and the adjacent building outlines of the building outlines in the subsequent phase orthophoto fusion image do not exist in the previous phase orthophoto fusion image, it indicates that the change type is an increase in bungalows; the current phase orthophoto fusion image If there is no building outline in the fused image, but there is a building outline in the later orthogonal fusion image, and the adjacent building outlines of the building outlines in the later orthogonal fusion image exist in the previous orthogonal fusion image, it indicates that the change type is bungalow expansion. If there are building outlines in both the current orthogonal fusion image and the later orthogonal fusion image, and the image band standard deviation of the building outline in the previous orthogonal fusion image and the later orthogonal fusion image is greater than the second threshold, or the brightness value difference of the building outline in the previous orthogonal fusion image and the later orthogonal fusion image is greater than the third threshold, it indicates that the change type is bungalow reconstruction.

[0117] Further, such as Fig.12 As shown, the identification module 1220 is also used to, when the scene is a gathering of buildings, perform multi-scale segmentation on the previous phase orthophoto fusion image. When the absolute value of the difference between the previous phase DSM value corresponding to the first feature and the previous phase DSM value corresponding to the second feature is less than the seventh threshold, it indicates that the first feature is a building, the first feature is an unclassified feature in the segmented previous phase orthophoto fusion image, and the second feature is a building in the segmented previous phase orthophoto fusion image whose adjacent object relationship value with the first feature is greater than the eighth threshold.

[0118] Further, such as Fig.12 As shown, the recognition module 1220 is also used to, when the scene is a gathering of buildings, perform projection difference correction on the building outline according to the satellite azimuth and incident angle in the orthophoto fusion image of the previous phase of the building outline.

[0119] Further, such as Fig.12 As shown, the detection module 1240 is further configured to indicate that the change is a false change when the difference between the DSM value of the previous phase and the DSM value of the later phase of the building whose type has changed is a first value.

[0120] An embodiment of the present application provides a storage medium having a program stored thereon, which, when executed by a processor, implements the building change detection method based on satellite three-dimensional images.

[0121] An embodiment of the present application provides a processor, which is used to run a program, wherein the program executes the building change detection method based on satellite three-dimensional images when running.

[0122] The present application also provides a computer program product, which, when executed on a data processing device, is suitable for executing a program for initializing the following method steps:

[0123] Obtain the front-phase orthophoto fusion image, the back-phase orthophoto fusion image, the front-phase DSM value, and the back-phase DSM value; identify the building outlines in the front-phase orthophoto fusion image and the back-phase orthophoto fusion image; determine the scene of the building outline in the front-phase orthophoto fusion image according to the front-phase DSM value corresponding to the building outline in the front-phase orthophoto fusion image, and determine the scene of the building outline in the back-phase orthophoto fusion image according to the back-phase DSM value corresponding to the building outline in the back-phase orthophoto fusion image; detect the change type of the building in the front-phase orthophoto fusion image and the back-phase orthophoto fusion image according to the scene and position relationship of the building outline in the front-phase orthophoto fusion image and the back-phase orthophoto fusion image.

[0124] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0125] In a typical configuration, the device includes one or more processors (CPU), memory and bus. The device may also include input / output interface, network interface and the like.

[0126] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip. The memory is an example of a computer-readable medium.

[0127] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0128] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0129] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0130] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. A building change detection method based on satellite three-dimensional images, characterized in that: The method comprises: Obtaining the front phase orthophoto fusion image, the back phase orthophoto fusion image, the front phase DSM value, and the back phase DSM value; Identifying the outlines of buildings in the front-phase orthophoto fusion image and the back-phase orthophoto fusion image; Determine the scene of the building outline in the front-phase orthophoto fusion image according to the front-phase DSM value corresponding to the building outline in the front-phase orthophoto fusion image, and determine the scene of the building outline in the back-phase orthophoto fusion image according to the back-phase DSM value corresponding to the building outline in the back-phase orthophoto fusion image, including: determine the building gathering area according to the brightness value of the building outline in the front-phase orthophoto fusion image; calculate the front-phase DSM value of the building gathering area according to the front-phase DSM value corresponding to each building outline in the building gathering area; when the front-phase DSM value of the building gathering area is greater than a first threshold value, indicate that the scene of the building gathering area is a building gathering; when the front-phase DSM value of the building gathering area is less than or equal to the first threshold value, indicate that the scene of the building gathering area is a bungalow gathering; The change types of buildings in the previous phase orthophoto fusion image and the next phase orthophoto fusion image are detected according to the scene and position relationship of the building outlines in the previous phase orthophoto fusion image and the next phase orthophoto fusion image.

2. The method according to claim 1, characterized in that Detecting the change type of the building in the previous phase orthophoto fusion image and the next phase orthophoto fusion image according to the scene and position relationship of the building outline in the previous phase orthophoto fusion image and the next phase orthophoto fusion image, including: When the scene is a gathering of buildings, determining the positional relationship between the outlines of the buildings in the previous phase orthophoto fusion image and the next phase orthophoto fusion image; When the building outline exists in the orthophoto fusion image of the previous phase, but does not exist in the orthophoto fusion image of the subsequent phase, it is indicated that the change type is a reduction in buildings; When the building outline does not exist in the orthophoto fusion image of the previous phase, the building outline exists in the orthophoto fusion image of the next phase, and the building outline adjacent to the building outline in the orthophoto fusion image of the next phase does not exist in the orthophoto fusion image of the previous phase, it indicates that the change type is the addition of buildings; When the building outline does not exist in the orthophoto fusion image of the previous phase, the building outline exists in the orthophoto fusion image of the next phase, and the building outline adjacent to the building outline in the orthophoto fusion image of the next phase exists in the orthophoto fusion image of the previous phase, it indicates that the change type is building expansion; When the building outline exists in both the previous phase orthophoto fusion image and the next phase orthophoto fusion image, and the image band standard deviation of the building outline in the previous phase orthophoto fusion image and the next phase orthophoto fusion image is greater than the second threshold, or the brightness value difference of the building outline in the previous phase orthophoto fusion image and the next phase orthophoto fusion image is greater than the third threshold, it is indicated that the change type is building reconstruction; When the building outline exists in both the front-phase orthophoto fusion image and the rear-phase orthophoto fusion image, and the difference between the rear-phase DSM value corresponding to the building outline and the front-phase DSM value corresponding to the building outline is greater than a fourth threshold, it indicates that the change type is building heightening.

3. The method according to claim 1, characterized in that The method further comprises: When the scene of the building cluster area is a cluster of bungalows, multi-scale segmentation and spectral difference segmentation are performed on the previous phase orthophoto fusion image; Determine the brightness value and adjacent object relationship value of the building outline in the segmented previous phase orthophoto fusion image; When the brightness value of the building outline in the previous phase orthophoto fusion image after the segmentation is greater than the fifth threshold, and the adjacent object relationship value of the building outline in the previous phase orthophoto fusion image after the segmentation is less than the sixth threshold, it indicates that the scene of the building outline in the previous phase orthophoto fusion image after the segmentation is a cluster of bungalows.

4. The method according to claim 3, characterized in that Detecting the change type of the building in the previous phase orthophoto fusion image and the next phase orthophoto fusion image according to the scene and position relationship of the building outline in the previous phase orthophoto fusion image and the next phase orthophoto fusion image, further comprising: When the scene is a cluster of bungalows, determining the positional relationship between the outlines of the buildings in the previous phase orthophoto fusion image and the next phase orthophoto fusion image; When the building outline exists in the orthophoto fusion image of the previous phase and does not exist in the orthophoto fusion image of the subsequent phase, it is indicated that the change type is a reduction in bungalows; When the building outline does not exist in the orthophoto fusion image of the previous phase, the building outline exists in the orthophoto fusion image of the next phase, and the building outline adjacent to the building outline in the orthophoto fusion image of the next phase does not exist in the orthophoto fusion image of the previous phase, it indicates that the change type is the increase of bungalows; When the building outline does not exist in the orthophoto fusion image of the previous phase, the building outline exists in the orthophoto fusion image of the next phase, and the building outline adjacent to the building outline in the orthophoto fusion image of the next phase exists in the orthophoto fusion image of the previous phase, it indicates that the change type is bungalow expansion; When the building outline exists in both the previous phase orthophoto fusion image and the next phase orthophoto fusion image, and the image band standard deviation of the building outline in the previous phase orthophoto fusion image and the next phase orthophoto fusion image is greater than the second threshold, or the brightness value difference of the building outline in the previous phase orthophoto fusion image and the next phase orthophoto fusion image is greater than the third threshold, it indicates that the change type is bungalow reconstruction.

5. The method according to claim 1, characterized in that The method further comprises: When the scene is a gathering of buildings, multi-scale segmentation is performed on the previous phase orthophoto fusion image; When the absolute value of the difference between the previous phase DSM value corresponding to the first ground object and the previous phase DSM value corresponding to the second ground object is less than the seventh threshold, it indicates that the first ground object is a building, the first ground object is an unclassified ground object in the segmented previous phase orthophoto fusion image, and the second ground object is a building in the segmented previous phase orthophoto fusion image whose adjacent object relationship value with the first ground object is greater than the eighth threshold.

6. The method according to claim 1, characterized in that The method further comprises: When the difference between the DSM value of the previous phase and the DSM value of the next phase of the building whose type has changed is a first value, it is indicated that the change is a false change.

7. A building change detection device based on satellite three-dimensional images, characterized in that: The device comprises: An acquisition module is used to acquire the front phase orthophoto fusion image, the back phase orthophoto fusion image, the front phase DSM value, and the back phase DSM value; A recognition module, used for recognizing the building outlines in the front-phase orthophoto fusion image and the back-phase orthophoto fusion image; A scene module, used to determine the scene of the building outline in the previous phase orthophoto fusion image according to the previous phase DSM value corresponding to the building outline in the previous phase orthophoto fusion image, and determine the scene of the building outline in the later phase orthophoto fusion image according to the later phase DSM value corresponding to the building outline in the later phase orthophoto fusion image, including: determining the building gathering area according to the brightness value of the building outline in the previous phase orthophoto fusion image; calculating the previous phase DSM value of the building gathering area according to the previous phase DSM value corresponding to each building outline in the building gathering area; when the previous phase DSM value of the building gathering area is greater than a first threshold value, indicating that the scene of the building gathering area is a building gathering; when the previous phase DSM value of the building gathering area is less than or equal to the first threshold value, indicating that the scene of the building gathering area is a bungalow gathering; The detection module is used to detect the change type of the building in the previous phase orthophoto fusion image and the next phase orthophoto fusion image according to the scene and position relationship of the building outline in the previous phase orthophoto fusion image and the next phase orthophoto fusion image.

8. A storage medium, characterized in that: The storage medium includes a stored program, wherein when the program is running, the device where the storage medium is located is controlled to execute the building change detection method based on satellite three-dimensional images as described in any one of claims 1-6.

9. An electronic device, characterized in that: The device includes at least one processor, and at least one memory and a bus connected to the processor; wherein the processor and the memory communicate with each other through the bus; the processor is used to call the program instructions in the memory to execute the building change detection method based on satellite three-dimensional images as described in any one of claims 1-6.

Images