Method, device and server for determining construction progress of target building

Abstract

This specification provides a method, device, and server for determining the construction progress of a target building. Based on this method, a target image group is first obtained, comprising multiple satellite images collected at multiple time points within a target time period. A preset image segmentation model is then used to process the target image group to obtain a corresponding black-and-white image group. Based on preset processing rules, the KL divergence of the black-and-white images in the black-and-white image group and the area of ​​the white image regions are calculated. The construction progress of the target building within the target time period is determined based on the KL divergence of the black-and-white images and the area of ​​the white image regions in the black-and-white image group. This method effectively reduces image processing errors and allows for efficient and accurate determination of the target building's construction progress.

Description

Technical Field

[0001] This specification belongs to the field of artificial intelligence technology, and in particular relates to a method, device, and server for determining the construction progress of a target building. Background Art

[0002] In the construction management of engineering projects, it is often necessary to regularly monitor the actual construction progress of a building in order to effectively supervise the construction process of the building.

[0003] However, existing methods often suffer from technical problems such as large errors and poor accuracy when using satellite images containing buildings to determine the construction progress of buildings.

[0004] To address the above issues, no effective solutions have been proposed so far. Summary of the Invention

[0005] This specification provides a method, device, and server for determining the construction progress of a target building, so as to effectively reduce errors and efficiently and accurately determine the construction progress of the target building.

[0006] This specification provides a method for determining the construction progress of a target building, including:

[0007] Acquire a target image group; wherein the target image group includes a plurality of satellite images respectively acquired at a plurality of time points within a target time period; the satellite images include the target area to be monitored;

[0008] Calling a preset image segmentation model to process the target image group to obtain a corresponding black-and-white image group; wherein the black-and-white image group includes a plurality of black-and-white images corresponding to the plurality of satellite images; and the white image area in the black-and-white images is used to represent the target building area in the target area to be monitored;

[0009] According to a preset processing rule, the KL divergence of the black and white images in the black and white image group and the area of ​​the white image region are calculated;

[0010] The construction progress of the target building within the target time period is determined according to the KL divergence of the black and white images in the black and white image group and the area of ​​the white image region.

[0011] In some embodiments, the plurality of satellite images in the target image group are arranged in chronological order according to corresponding time points;

[0012] Correspondingly, the plurality of black-and-white images in the black-and-white image group are arranged in chronological order according to the corresponding time points.

[0013] In some embodiments, after calling a preset image segmentation model to process the target image group to obtain a corresponding black and white image group, the method further includes:

[0014] Obtain the first-ranked satellite image in the target image group as the starting reference image;

[0015] Performing image recognition on the starting reference image to determine interfering buildings that are not target buildings in the starting reference image;

[0016] According to the interfering buildings, correction processing is performed on white image areas in a plurality of black and white images in the black and white image group.

[0017] In some embodiments, calculating the KL divergence of the black-and-white images in the black-and-white image group according to a preset processing rule includes:

[0018] The KL divergence of the current black-and-white image in the black-and-white image group is calculated according to the preset processing rules in the following manner:

[0019] Calculate the position distribution mean of the white image area and the position distribution mean of the black image area in the current black and white image;

[0020] The KL divergence of the current black and white image is calculated based on the position distribution mean of the white image area and the position distribution mean of the black image area in the current black and white image.

[0021] In some embodiments, determining the construction progress of a target building within a target time period based on the KL divergence of the black and white images and the area of ​​the white image region in the black and white image group includes:

[0022] Sequentially comparing the KL divergence of the black-and-white images in the black-and-white image group with a preset KL divergence threshold, and comparing the area of ​​the white image region of the black-and-white image with a preset area threshold, to determine whether a trigger image is detected; wherein the trigger image is the first black-and-white image in the black-and-white image group whose KL divergence is greater than the preset KL divergence threshold and whose white image region has an area greater than the preset area threshold;

[0023] When it is determined that no triggering image is detected, it is determined that the target building is in an unstarted stage within the target time period.

[0024] In some embodiments, the method further comprises:

[0025] When it is determined that the trigger image is detected, it is determined that the target building is in the construction phase within the target time period.

[0026] In some embodiments, after determining that the target building is in the construction start phase within the target time period, the method further includes:

[0027] Extracting a trigger image and a black-and-white image in the black-and-white image group that is sorted after the trigger image, and constructing a corresponding progress image group;

[0028] The construction progress between adjacent time points in the commencement phase within the target time period is determined by calculating and based on the KL divergence difference between adjacent images in the progress image group and the area difference of the white image region.

[0029] In some embodiments, the target building includes at least one of the following: a highway, a power station, and a factory building.

[0030] The embodiment of this specification also provides a device for determining the construction progress of a target building, including:

[0031] An acquisition module is configured to acquire a target image group, wherein the target image group includes a plurality of satellite images acquired at a plurality of time points within a target time period, and the satellite images include a target area to be monitored;

[0032] a segmentation module, configured to call a preset image segmentation model to process the target image group to obtain a corresponding black-and-white image group; wherein the black-and-white image group includes a plurality of black-and-white images corresponding to the plurality of satellite images; and a white image area in the black-and-white images is used to represent a target building area in the target area to be monitored;

[0033] a calculation module, configured to calculate the KL divergence of the black and white images and the area of ​​the white image region in the black and white image group according to a preset processing rule;

[0034] The determination module is configured to determine the construction progress of the target building within the target time period according to the KL divergence of the black and white images in the black and white image group and the area of ​​the white image region.

[0035] An embodiment of the present specification also provides a server, comprising a processor and a memory for storing processor-executable instructions, wherein the processor implements the following steps when executing the instructions: obtaining a target image group; wherein the target image group includes multiple satellite images collected at multiple time points within a target time period; the satellite images include a target area to be monitored; calling a preset image segmentation model to process the target image group to obtain a corresponding black-and-white image group; wherein the black-and-white image group includes multiple black-and-white images corresponding to the multiple satellite images; the white image area in the black-and-white images is used to represent a target building area in the target area to be monitored; according to a preset processing rule, calculating the KL divergence of the black-and-white images in the black-and-white image group and the area area of ​​the white image area; and determining the construction progress of the target building within the target time period based on the KL divergence of the black-and-white images in the black-and-white image group and the area area of ​​the white image area.

[0036] An embodiment of the present specification also provides a computer-readable storage medium having computer instructions stored thereon, which, when executed, implement the following steps: obtaining a target image group; wherein the target image group includes multiple satellite images collected at multiple time points within a target time period; the satellite images include a target area to be monitored; calling a preset image segmentation model to process the target image group to obtain a corresponding black-and-white image group; wherein the black-and-white image group includes multiple black-and-white images corresponding to the multiple satellite images; the white image area in the black-and-white images is used to represent the target building area in the target area to be monitored; according to preset processing rules, calculating the KL divergence of the black-and-white images in the black-and-white image group and the area area of ​​the white image area; and determining the construction progress of the target building within the target time period based on the KL divergence of the black-and-white images in the black-and-white image group and the area area of ​​the white image area.

[0037] The embodiment of this specification also provides an image processing method, including:

[0038] Acquire a target image group; wherein the target image group includes a plurality of satellite images respectively acquired at a plurality of time points within a target time period; the satellite images include a target area;

[0039] Calling a preset image segmentation model to process the target image group to obtain a corresponding black-and-white image group; wherein the black-and-white image group includes a plurality of black-and-white images corresponding to the plurality of satellite images; and the white image area in the black-and-white images is used to represent the target object in the target area;

[0040] According to a preset processing rule, the KL divergence of the black and white images in the black and white image group and the area of ​​the white image region are calculated;

[0041] A change trend of the target object within a target time period is determined according to the KL divergence of the black and white images and the area of ​​the white image region in the black and white image group.

[0042] This specification provides a method, device, and server for determining the construction progress of a target building. Based on this method, a target image group comprising multiple satellite images acquired at multiple time points within a target time period is first obtained; a preset image segmentation model is invoked to process the target image group to obtain a corresponding black-and-white image group; then, according to preset processing rules, the KL divergence of the black-and-white images and the area of ​​the white image regions in the black-and-white image group are calculated; and the construction progress of the target building within the target time period is determined based on the KL divergence of the black-and-white images and the area of ​​the white image regions in the black-and-white image group. By first processing the target image group comprising multiple satellite images at multiple time points within the target time period into corresponding black-and-white image groups, and then calculating and utilizing the KL divergence of the black-and-white images and the area of ​​the white image regions in the black-and-white image groups to determine the construction progress of the target building, errors can be effectively reduced, allowing for efficient and accurate determination of the construction progress of the target building. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the embodiments of this specification, the following will briefly introduce the drawings required for use in the embodiments. The drawings described below are only some embodiments recorded in this specification. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0044] Figure 1 is a flowchart of a method for determining the construction progress of a target building provided by an embodiment of this specification;

[0045] Figure 2 is a flowchart of an image processing method provided by an embodiment of this specification;

[0046] Figure 3 This is a schematic diagram of the structure of a server provided by an embodiment of this specification;

[0047] Figure 4 This is a schematic diagram of the structure of a device for determining the construction progress of a target building provided by one embodiment of this specification;

[0048] Figure 5 This is a schematic diagram of an embodiment of a method for determining the construction progress of a target building provided by an embodiment of this specification, in a scenario example;

[0049] Figure 6This is a schematic diagram of an embodiment of a method for determining the construction progress of a target building provided by an embodiment of this specification, in a scenario example. DETAILED DESCRIPTION

[0050] To help those skilled in the art better understand the technical solutions in this specification, the following will provide a clear and complete description of the technical solutions in the embodiments of this specification, in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of this specification, not all of them. All other embodiments derived by those skilled in the art based on the embodiments in this specification without creative effort shall fall within the scope of protection of this specification.

[0051] See Figure 1 As shown, the embodiment of this specification provides a method for determining the construction progress of a target building. The method is specifically applied to the server side. When implemented, the method may include the following:

[0052] S101: Acquire a target image group; wherein the target image group includes a plurality of satellite images acquired at a plurality of time points within a target time period; the satellite images include a target area to be monitored;

[0053] S102: Calling a preset image segmentation model to process the target image group to obtain a corresponding black-and-white image group; wherein the black-and-white image group includes a plurality of black-and-white images corresponding to the plurality of satellite images; and the white image area in the black-and-white images is used to represent the target building area in the target area to be monitored;

[0054] S103: Calculating the KL divergence of the black and white images and the area of ​​the white image region in the black and white image group according to a preset processing rule;

[0055] S104: Determine the construction progress of the target building within the target time period according to the KL divergence of the black and white images and the area of ​​the white image region in the black and white image group.

[0056] Through the above embodiment, the server can first obtain and, based on a target image group of multiple satellite images collected at multiple time points within a target time period, perform corresponding segmentation processing and image processing to obtain a black-and-white image group including multiple black-and-white images corresponding to the multiple satellite images. Then, by calculating and utilizing the KL divergence of each black-and-white image in the black-and-white image group and the area of ​​the white image region representing the target building, the server can specifically analyze and accurately determine the construction progress of the target building of interest, thereby effectively reducing data errors in the satellite images, efficiently and accurately determining the construction progress of the target building, and monitoring and managing the construction progress of the target building in the target area.

[0057] In some embodiments, the target area may be specifically understood as a range area where the target building is planned to be constructed, or where the construction of the target building has already begun.

[0058] The target building may specifically be a building that the user is interested in or that requires construction monitoring.

[0059] In some embodiments, the target building may specifically include at least one of the following: a highway, a power station, a factory building, etc. Of course, it should be noted that the target buildings listed above are merely illustrative. In practice, depending on the specific application scenario and processing requirements, the target building may also include other types of building objects or project objects. For example, a dam, a windbreak, a residential building, etc.

[0060] Through the above embodiments, the method for determining the construction progress of a target building provided in this specification can be used to more effectively and accurately determine and monitor the construction progress of various types of buildings in various application scenarios.

[0061] In some embodiments, the above-mentioned satellite image can be specifically understood as a satellite image (or remote sensing image) containing the target area to be monitored, which is collected by a remote sensing satellite or other equipment at a certain point in time.

[0062] It should be noted that for some target buildings in a target area, the background color of the target area is the same or similar to that of the target building, resulting in a relatively blurred building outline in a single satellite image. For example, a wind turbine station in the snow. Therefore, processing and recognition based on only a single satellite image is prone to errors, affecting the accuracy of subsequent processing.

[0063] Based on these considerations, each processing step uses not a single satellite image corresponding to a single point in time, but rather a set of target images corresponding to a target time period, encompassing multiple satellite images collected at multiple points in time. Compared to relying on a single satellite image, relying on this set of target images effectively reduces the impact of image errors on subsequent data processing, improving processing accuracy.

[0064] In some embodiments, the target image group may specifically include multiple satellite images corresponding to a target time period. Each of the multiple satellite images corresponds to a time point within the target time period, and the satellite images were acquired at the corresponding time point. The target time period may specifically be a month, a week, a quarter, or the like.

[0065] In some embodiments, the multiple satellite images in the target image group are arranged in chronological order according to the corresponding time points; correspondingly, the multiple black-and-white images in the black-and-white image group are arranged in chronological order according to the corresponding time points.

[0066] Specifically, in the target picture group, the plurality of satellite pictures may be arranged in chronological order (e.g., in descending order) according to the corresponding time points. Specifically, for example, the time point corresponding to the first satellite picture in the target picture group may be the earliest time point in the target time period, and the time point corresponding to the last satellite picture in the target picture group may be the latest time point in the target time period. The time points corresponding to satellite pictures arranged relatively earlier in the target picture group may be earlier than the time points corresponding to satellite pictures arranged relatively later in the target picture group.

[0067] In some embodiments, the black-and-white image group may include multiple black-and-white images. Each black-and-white image corresponds to a satellite image in the target image group. Accordingly, the time point corresponding to the black-and-white image is the same as the time point of the corresponding satellite image. Accordingly, within the black-and-white image group, the multiple black-and-white images may be arranged in chronological order (e.g., in descending order) according to the corresponding time points.

[0068] Through the above embodiment, the target picture group and the black-and-white image group arranged in time sequence according to the corresponding time points can be obtained and used, so that the target picture group and the black-and-white image group can be better used for specific data processing later.

[0069] In some embodiments, the shooting angle, resolution, and image size of each of the multiple satellite images included in the target image group may be the same. Furthermore, the time intervals between the time points corresponding to any two adjacent satellite images in the target image group may also be the same.

[0070] In some embodiments, the black and white images are each a satellite image. Specifically, the black and white images only include white image areas and / or black image areas. The white image areas can be used to represent target buildings to be monitored in the target area, and the black image areas can be used to represent other objects in the target area other than the target building, for example, natural environment objects (mountains, rivers, lakes, forests, etc.) in the target area, and other building objects in the target area other than the target building (existing houses, roads, etc. in the target area).

[0071] In some embodiments, the above-mentioned preset image segmentation model can specifically be a pre-trained algorithm model that can detect and distinguish buildings and non-buildings in the picture, and segment the image area where the building is located and mark it as white, and segment the image area where the non-building is located and mark it as black.

[0072] In some embodiments, the preset image segmentation model may specifically be a model obtained by combining a convolutional neural network and a deconvolutional neural network.

[0073] In some embodiments, during implementation, a preset image segmentation model can be used to process each satellite image in the target image group of the output model to output a corresponding black and white image, thereby obtaining a corresponding black and white image group. This processing can convert satellite images, which originally have complex image information, into relatively more targeted black and white images that are easier to identify target buildings, thereby improving subsequent processing efficiency and accuracy.

[0074] In some embodiments, when a satellite image is input into a preset image segmentation model and the preset image segmentation model is run to specifically process the satellite image, the model will first perform image segmentation on the satellite image based on image features, and simultaneously identify each image unit after segmentation to obtain a set of result matrices; then, by performing corresponding processing (for example, normalization processing, coloring processing, etc.) on the result matrix, a black and white image corresponding to the satellite image is output.

[0075] In some embodiments, it is considered that in the black-and-white images directly obtained by processing satellite images using a preset image segmentation model, non-target buildings (which may be referred to as interfering buildings) in the target area may be mistakenly identified as target buildings, and the image area where the interfering buildings are located may be marked as white, resulting in errors. Therefore, after calling the preset image segmentation model to process the target image group and obtain the corresponding black-and-white image group, the black-and-white image group can be corrected accordingly to eliminate these errors and obtain a more accurate black-and-white image group.

[0076] In some embodiments, after calling a preset image segmentation model to process the target image group to obtain a corresponding black and white image group, the method may further include the following steps when implemented:

[0077] S1: Obtain the first-ranked satellite image in the target image group as the starting reference image;

[0078] S2: performing image recognition on the starting reference image to determine interfering buildings in the starting reference image that do not belong to the target building;

[0079] S3: performing correction processing on the white image areas of the plurality of black and white images in the black and white image group according to the interfering buildings.

[0080] Through the above embodiments, the interference and influence of interfering buildings on the target building in the black and white image can be eliminated in a timely manner, thereby obtaining a black and white image group with higher accuracy and smaller error.

[0081] In some embodiments, the above-mentioned image recognition is performed on the starting reference picture to determine the interfering buildings that do not belong to the target building in the starting reference picture. When implemented specifically, it can include: obtaining and based on the planning and design drawings of the target building, combining the image recognition results of the actual reference picture, finding the buildings in the starting reference picture that obviously do not belong to the target building and marking them as interfering buildings.

[0082] In some embodiments, based on the interfering buildings, the white image areas in the multiple black and white images in the black and white image group are respectively corrected, which may specifically include: modifying the image area belonging to the interfering buildings in the white image area of ​​the black and white images into a black image area.

[0083] In some embodiments, the above-mentioned calculation of the KL divergence of the black-and-white images in the black-and-white image group according to the preset processing rules may include the following steps: calculating the KL divergence of the current black-and-white image in the black-and-white image group according to the preset processing rules in the following manner:

[0084] S1: Calculate the position distribution mean of the white image area and the position distribution mean of the black image area in the current black and white image;

[0085] S2: Calculate the KL divergence of the current black and white image based on the position distribution mean of the white image area and the position distribution mean of the black image area in the current black and white image.

[0086] The KL divergence of the current black-and-white image can specifically reflect the difference in position distribution probability of the target building area in the current black-and-white image relative to other areas in the target area except the target building based on the statistical dimension.

[0087] According to the above-mentioned method of processing the current black-and-white image, other black-and-white images in the black-and-white image group may be processed in sequence to obtain the KL divergence of each black-and-white image in the black-and-white image group.

[0088] Through the above embodiment, the KL divergence based on the statistical dimension can be calculated to better reflect the relative difference in the distribution probability of the target building at the time point corresponding to each black and white image relative to other areas in the target area except the target building; and then the above KL divergence can be used to analyze and determine the construction changes of the target building at each time point within the target time period more finely based on the statistical dimension.

[0089] In some embodiments, the KL divergence, which may also be referred to as relative entropy, Kullback-Leibler divergence, or information divergence, is a parameter data that can characterize the asymmetric measure of the difference between two probability distributions.

[0090] In some embodiments, the above-mentioned determination of the construction progress of the target building within the target time period based on the KL divergence of the black and white images in the black and white image group and the area of ​​the white image area may include the following content during specific implementation: sequentially comparing the KL divergence of the black and white images in the black and white image group with a preset KL divergence threshold, and comparing the area of ​​the white image area of ​​the black and white image with a preset area threshold to determine whether a trigger image is detected; wherein, the trigger image is the first black and white image in the black and white image group whose KL divergence is greater than the preset KL divergence threshold and whose area of ​​the white image area is greater than the preset area threshold; if it is determined that no trigger image is detected, it is determined that the target building is in the unstarted stage within the target time period.

[0091] The above-mentioned preset KL divergence threshold and the preset region area threshold may specifically be threshold data obtained by taking the average value after pre-training a large amount of historical sample data.

[0092] Through the above embodiment, the KL divergence of each black and white image in the black and white image group and the area of ​​the white image region can be combined to more accurately identify and determine whether the target building is in the construction stage within the target time period.

[0093] In some embodiments, specifically, for example, the server may first detect, in chronological order, whether the KL divergence and the area of ​​the white image region of the first-ranked black-and-white image in the black-and-white image group are greater than a preset KL divergence threshold and a preset area threshold, respectively. If it is determined that the KL divergence and the area of ​​the white image region of the first-ranked black-and-white image are greater than the preset KL divergence threshold and the preset area threshold, respectively, the black-and-white image may be determined as a triggering image, thereby determining that the triggering image has been detected, and further determining that the target building is in the construction phase within the target time period.

[0094] Conversely, if it is determined that the KL divergence of the first-ranked black-and-white image is less than or equal to a preset KL divergence threshold, and / or the area of ​​the white image region is less than or equal to a preset area threshold, the black-and-white image can be determined not to be a triggering image. Furthermore, the above detection process can be repeated, sequentially detecting the second-ranked black-and-white image following the first-ranked black-and-white image in the black-and-white image group to determine whether the second-ranked black-and-white image is a triggering image. This process can be repeated until a triggering image is found in the black-and-white image group, or all black-and-white images in the black-and-white image group have been detected.

[0095] In some embodiments, if a triggering image is detected, it can be determined that the target building is in the construction phase during the target time period. Conversely, if no triggering image is detected after all black-and-white images in the black-and-white image group have been detected, it can be determined that the target building is not in the construction phase during the target time period.

[0096] In some embodiments, when the method is specifically implemented, it may further include the following: when it is determined that the trigger image is detected, determining that the target building is in the construction phase within the target time period.

[0097] Through the above embodiment, when it is determined that a trigger image is detected in the black and white image group, it can be determined that the target building is in the construction stage within the target time period. Furthermore, the time point corresponding to the trigger image can be determined as the construction start time point within the target time period.

[0098] In some embodiments, after determining that the target building is in the construction start phase within the target time period, the method may further include the following steps during implementation:

[0099] S1: extracting a trigger image and black-and-white images that are sorted after the trigger image in the black-and-white image group, and constructing a corresponding progress image group;

[0100] S2: Calculate and determine the construction progress between adjacent time points in the commencement phase within the target time period based on the KL divergence difference between adjacent images in the progress image group and the area difference of the white image region.

[0101] Through the above embodiment, after the trigger image is detected, the black and white image sorted after the trigger image in the black and white image group can be extracted and obtained, together with the trigger image, as a progress image to construct a progress image group; then, only the images in the above progress image group can be targetedly calculated and processed to determine the specific construction progress between different time points, so that the specific construction progress of the target building between different time points after the start of construction within the target time period can be determined more efficiently and accurately.

[0102] In some embodiments, the above calculation and determination of the construction progress between adjacent time points in the commencement phase within the target time period based on the KL divergence difference between adjacent images in the progress image group and the area difference of the white image area may include the following during specific implementation: calculating the construction progress between adjacent first and second time points in the commencement phase in the following manner; wherein the first image in the progress image group corresponds to the first time point, the second image in the progress image group corresponds to the second time point, and the first time point is the previous time point of the second time point: subtracting the KL divergence of the first image from the KL divergence of the second image to obtain a KL divergence difference; subtracting the area of ​​the white image area of ​​the first image from the area of ​​the white image area of ​​the second image to obtain an area area difference; detecting whether the KL divergence difference is greater than 0 and whether the area area difference is greater than 0; and when it is determined that the KL divergence difference is greater than 0 and the area area difference is greater than 0, determining that the construction status of the target building in the time period between the first and second time points is normal construction.

[0103] Otherwise, it is temporarily impossible to determine whether the target building is under normal construction. In this case, image detail recognition can be performed on the satellite images corresponding to the first time point and the second time point, or an on-site inspection of the target building in the target area can be initiated to obtain image detail recognition results or on-site inspection results. Then, based on the image detail recognition results or on-site inspection results, it can be accurately determined whether the construction status of the target building is normal.

[0104] The above method can accurately identify and determine the specific construction status of the target building between two adjacent time points within the target time period, and then combine the above multiple construction statuses to obtain a more detailed and specific construction progress record of the target building within the target time period.

[0105] In some embodiments, when the method is implemented, it also includes the following contents: generating a construction progress record for the target building based on the construction progress within the identified target time period and the construction status between adjacent time points; and providing corresponding prompts based on the construction progress record of the target building to more effectively monitor and manage the construction progress of the target building.

[0106] As can be seen from the above, the method for determining the construction progress of a target building provided in the embodiments of this specification can first obtain a target image group comprising multiple satellite images collected at multiple time points within a target time period; then invoke a preset image segmentation model to process the target image group to obtain a corresponding black-and-white image group; calculate the KL divergence of the black-and-white images and the area of ​​the white image regions in the black-and-white image group according to preset processing rules; and determine the construction progress of the target building within the target time period based on the KL divergence of the black-and-white images and the area of ​​the white image regions in the black-and-white image group. By first processing the target image group comprising multiple satellite images into a corresponding black-and-white image group, and then calculating and utilizing the KL divergence of the black-and-white images and the area of ​​the white image regions to determine the construction progress of the target building, errors can be effectively reduced, and the construction progress of the target building can be determined efficiently and accurately.

[0107] See Figure 2 As shown, the embodiment of this specification also provides an image processing method. Wherein, when the method is specifically implemented, it may include the following contents:

[0108] S201: Acquire a target image group; wherein the target image group includes a plurality of satellite images acquired at a plurality of time points within a target time period; the satellite images include a target area;

[0109] S202: Invoking a preset image segmentation model to process the target image group to obtain a corresponding black-and-white image group; wherein the black-and-white image group includes a plurality of black-and-white images corresponding to the plurality of satellite images; and a white image area in the black-and-white images is used to represent a target object in the target area;

[0110] S203: Calculating the KL divergence of the black and white images and the area of ​​the white image region in the black and white image group according to a preset processing rule;

[0111] S204: Determine a change trend of the target object within a target time period according to the KL divergence of the black-and-white images and the area of ​​the white image region in the black-and-white image group.

[0112] In some embodiments, the target object may be an object that has appeared in the target area or may appear in the target area and that the user is interested in or needs to monitor. Specifically, the target object may be a building or a windbreak.

[0113] The image processing method provided in the embodiments of this specification can effectively reduce errors and efficiently and accurately determine the change trend of the target object in the target area within the target time period.

[0114] An embodiment of the present specification further provides a server, comprising a processor and a memory for storing processor-executable instructions. When the processor is specifically implemented, it can perform the following steps according to the instructions: obtaining a target image group; wherein the target image group includes multiple satellite images respectively collected at multiple time points within a target time period; the satellite images include a target area to be monitored; calling a preset image segmentation model to process the target image group to obtain a corresponding black-and-white image group; wherein the black-and-white image group includes multiple black-and-white images corresponding to the multiple satellite images; the white image area in the black-and-white images is used to represent a target building area in the target area to be monitored; according to a preset processing rule, calculating the KL divergence of the black-and-white images in the black-and-white image group and the area area of ​​the white image area; and determining the construction progress of the target building in the target time period based on the KL divergence of the black-and-white images in the black-and-white image group and the area area of ​​the white image area.

[0115] In order to complete the above instructions more accurately, refer to Figure 3 As shown, the embodiment of this specification also provides another specific server, wherein the server includes a network communication port 301, a processor 302 and a memory 303, and the above structures are connected through internal cables so that each structure can perform specific data interaction.

[0116] The network communication port 301 can be used to obtain a target image group, wherein the target image group includes multiple satellite images collected at multiple time points within a target time period; and the satellite images include a target area to be monitored.

[0117] The processor 302 can be specifically used to call a preset image segmentation model to process the target image group to obtain a corresponding black-and-white image group; wherein the black-and-white image group includes multiple black-and-white images corresponding to the multiple satellite images respectively; the white image area in the black-and-white image is used to represent the target building area in the target area to be monitored; according to a preset processing rule, the KL divergence of the black-and-white images in the black-and-white image group and the area area of ​​the white image area are calculated; based on the KL divergence of the black-and-white images in the black-and-white image group and the area area of ​​the white image area, the construction progress of the target building within the target time period is determined.

[0118] The memory 303 may be specifically used to store corresponding instruction programs.

[0119] In this embodiment, the network communication port 301 can be a virtual port that is bound to different communication protocols, thereby being capable of sending or receiving different data. For example, the network communication port can be a port responsible for web data communication, a port responsible for FTP data communication, or a port responsible for email data communication. Furthermore, the network communication port can also be a physical communication interface or communication chip. For example, it can be a wireless mobile network communication chip, such as GSM or CDMA; it can also be a Wi-Fi chip; or it can be a Bluetooth chip.

[0120] In this embodiment, the processor 302 can be implemented in any suitable manner. For example, the processor can take the form of a microprocessor or a processor and a computer-readable medium storing computer-readable program code (such as software or firmware) executable by the (micro)processor, a logic gate, a switch, an application-specific integrated circuit (ASIC), a programmable logic controller, an embedded microcontroller, etc. This specification is not limited to this.

[0121] In this embodiment, the memory 303 may include multiple levels. In a digital system, anything that can store binary data can be a memory. In an integrated circuit, a circuit with a storage function that does not have a physical form is also called a memory, such as RAM, FIFO, etc. In a system, a storage device with a physical form is also called a memory, such as a memory stick, TF card, etc.

[0122] An embodiment of the present specification also provides another server, including a processor and a memory for storing processor-executable instructions. When specifically implemented, the processor can perform the following steps according to the instructions: obtaining a target image group; wherein the target image group includes multiple satellite images collected at multiple time points within a target time period; the satellite images include a target area; calling a preset image segmentation model to process the target image group to obtain a corresponding black-and-white image group; wherein the black-and-white image group includes multiple black-and-white images corresponding to the multiple satellite images; the white image area in the black-and-white images is used to represent the target object in the target area; according to a preset processing rule, calculating the KL divergence of the black-and-white images in the black-and-white image group and the area of ​​the white image area; and determining the change trend of the target object within the target time period based on the KL divergence of the black-and-white images and the area of ​​the white image area in the black-and-white image group.

[0123] An embodiment of the present specification also provides a computer-readable storage medium for a method for determining the construction progress of a target building, wherein the computer-readable storage medium stores computer program instructions, which, when executed, implement the following steps: obtaining a target image group; wherein the target image group includes multiple satellite images collected at multiple time points within a target time period; the satellite images include a target area to be monitored; calling a preset image segmentation model to process the target image group to obtain a corresponding black-and-white image group; wherein the black-and-white image group includes multiple black-and-white images corresponding to the multiple satellite images; the white image area in the black-and-white images is used to represent the target building area in the target area to be monitored; according to preset processing rules, calculating the KL divergence of the black-and-white images in the black-and-white image group and the area area of ​​the white image area; and determining the construction progress of the target building within the target time period based on the KL divergence of the black-and-white images in the black-and-white image group and the area area of ​​the white image area.

[0124] In this embodiment, the storage medium includes, but is not limited to, random access memory (RAM), read-only memory (ROM), cache, hard disk drive (HDD), or memory card. The memory can be used to store computer program instructions. The network communication unit can be an interface configured in accordance with the standards specified by the communication protocol for network connection communication.

[0125] In this embodiment, the functions and effects specifically implemented by the program instructions stored in the computer-readable storage medium can be explained in comparison with other implementations and will not be repeated here.

[0126] An embodiment of the present specification further provides a computer-readable storage medium based on the above-mentioned image processing method, wherein the computer-readable storage medium stores computer program instructions, which, when executed, implement the following steps: obtaining a target image group; wherein the target image group includes multiple satellite images collected at multiple time points within a target time period; wherein the satellite images include a target area; calling a preset image segmentation model to process the target image group to obtain a corresponding black-and-white image group; wherein the black-and-white image group includes multiple black-and-white images corresponding to the multiple satellite images; wherein the white image areas in the black-and-white images are used to represent target objects in the target area; calculating the KL divergence of the black-and-white images and the area area of ​​the white image areas in the black-and-white image group according to preset processing rules; and determining a change trend of the target objects within the target time period based on the KL divergence of the black-and-white images and the area area of ​​the white image areas in the black-and-white image group.

[0127] See Figure 4 As shown, at the software level, the embodiment of this specification further provides a device for determining the construction progress of a target building, which may specifically include the following structural modules:

[0128] The acquisition module 401 may be specifically configured to acquire a target image group, wherein the target image group includes a plurality of satellite images acquired at a plurality of time points within a target time period, and the satellite images include a target area to be monitored;

[0129] The segmentation module 402 may be specifically configured to call a preset image segmentation model to process the target image group to obtain a corresponding black-and-white image group; wherein the black-and-white image group includes a plurality of black-and-white images corresponding to the plurality of satellite images; and the white image area in the black-and-white images is used to represent the target building area in the target area to be monitored;

[0130] The calculation module 403 may be specifically configured to calculate the KL divergence of the black and white images and the area of ​​the white image region in the black and white image group according to a preset processing rule;

[0131] The determination module 404 may be specifically configured to determine the construction progress of the target building within the target time period according to the KL divergence of the black and white images and the area of ​​the white image regions in the black and white image group.

[0132] It should be noted that the units, devices or modules described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. For the convenience of description, the above devices are described in terms of functions and are divided into various modules and described separately. Of course, when implementing this specification, the functions of each module can be implemented in the same or multiple software and / or hardware, or the module that implements the same function can be implemented by a combination of multiple sub-modules or sub-units. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0133] As can be seen from the above, the device for determining the construction progress of a target building provided in the embodiments of this specification can effectively reduce errors and determine the construction progress of the target building efficiently and accurately.

[0134] In a specific scenario example, the method for determining the construction progress of a target building provided in the embodiments of this specification can be applied to continuously monitor the progress of a project (eg, a target building). The specific implementation process can include the following steps.

[0135] S0: Input a set of satellite images within a continuous period of time (e.g., a target set of images).

[0136] See Figure 5 The pictures included in the picture group can be satellite images (for example, satellite pictures) of the same location, same time interval, same resolution, and continuous time. Each picture group includes at least one picture.

[0137] S2: Use an image segmentation model (eg, a preset image segmentation model) to perform image segmentation on the image group to generate a black and white image.

[0138] In specific implementations, the image segmentation model can be used to segment the satellite images in the input image set, thereby obtaining a set of result matrices. The result matrices are then processed accordingly to generate a black and white image set containing multiple black and white images. White (e.g., white image areas) in the black and white images represent project buildings, and black represents non-project buildings. Figure 6 As shown, the white portion in the black and white image may specifically be a photovoltaic panel.

[0139] S4: Use the generated black and white image group to compare the area and distribution, and output the evolution process.

[0140] Specifically, the black and white image group generated by the above-mentioned image segmentation algorithm can be used to perform position distribution calculations (for example, calculating the distribution mean of the white image area and the black image area) to obtain the corresponding KL divergence and the white area (for example, the area of ​​the white image area) for detection. If the KL divergence of an image in the group is greater than a certain threshold t1 (for example, a preset KL divergence threshold) and the white area is greater than a certain threshold t2 (for example, a preset area threshold), it can be considered that the project has started at the current time point.

[0141] It should be noted that by calculating and using the KL divergence, we can use the peak difference between the two distributions in the image (white and black) as the basis for judgment. The so-called KL divergence can specifically be the difference between the two distributions, that is, the difference between white and black. Generally, the larger the difference value, the lower the probability of misidentification.

[0142] Under normal construction conditions, as subsequent projects proceed, the distribution mean and area of ​​subsequent images will continue to increase, which can also be used to determine whether the project has made progress within a period of time.

[0143] In this example scenario, using an image segmentation model to segment the image group and generate a black-and-white image group can include: generating a result matrix using the image segmentation model; normalizing the result matrix so that the matrix values ​​are only 0 or 1, where 1 represents project buildings (such as photovoltaic panels, factory buildings, or roads) and 0 represents non-project buildings (such as surrounding grass and other buildings). This 0,1 matrix can be mapped into a black-and-white image, where white represents project buildings and black represents non-project buildings.

[0144] In this scenario example, the image segmentation model used can specifically be a model composed of a convolutional neural network and a deconvolutional neural network.

[0145] In this scenario example, during specific implementation, the generated black-and-white image group can be used to calculate the area occupied by the project building in the image.

[0146] Specifically, for the generated black and white image, OpenCV can be used to calculate the area size of the white part and the area size of the black part in the image, and at the same time, the ratio of the white area to the black area can be obtained.

[0147] In this scenario example, during specific implementation, the generated black-and-white image group can be used to calculate the distribution of project buildings and non-project buildings in the image, and perform a distribution comparison.

[0148] In this scenario example, during implementation, the distribution of the black and white areas in the black and white image M can be calculated, that is, their probability density functions can be obtained. This can yield the probability density function F0 (M=0) for the non-project building area (black area) and the probability density function F1 (M=1) for the project building area (white area).

[0149] Specifically, for the two obtained probability density functions, KL divergence can be applied (KL divergence is used to measure the difference between two probability distributions) to output the KL divergence value.

[0150] In this scenario example, during specific implementation, the time node whose distribution comparison result is greater than a certain threshold t1 and whose area is greater than a certain threshold t2 can be selected as the start time, and the changes in the area and distribution in the image at subsequent times can be output as the project progress.

[0151] Specifically, the KL divergence threshold t1 is the average KL divergence at the time of construction commencement in the historical data. The area ratio threshold t2 is also the average area at the time of construction commencement in the historical data. Only when the KL divergence is greater than threshold t1 and the area ratio is greater than threshold t2 can it be determined that the project in the current input satellite image has commenced.

[0152] After determining the start time of construction, subsequent images can only determine whether their area ratio and KL divergence continue to increase to determine whether the subsequent construction is progressing steadily.

[0153] Through the above scenario examples, it is verified that the method for determining the construction progress of the target building provided in the embodiment of this specification can effectively reduce the errors caused by factors such as shooting angle, weather, season and time when relying solely on a single satellite image by performing a series of operations such as image segmentation, area comparison and distribution calculation; at the same time, by analyzing a set of continuous satellite images, the project progress can be monitored more accurately.

[0154] Although this specification provides the method operation steps as described in the embodiments or flow charts, more or fewer operation steps may be included based on conventional or non-creative means. The order of steps listed in the embodiments is only one way of executing the order of many steps and does not represent the only execution order. When the device or client product in practice is executed, it can be executed in sequence or in parallel according to the method shown in the embodiments or the drawings (for example, a parallel processor or a multi-threaded processing environment, or even a distributed data processing environment). The term "comprise", "include" or any other variant thereof is intended to cover non-exclusive inclusion, so that the process, method, product or device including a series of elements includes not only those elements, but also includes other elements that are not explicitly listed, or also includes elements inherent to such process, method, product or device. In the absence of more restrictions, it is not excluded that there are other identical or equivalent elements in the process, method, product or device including the elements. Words such as first and second are used to represent names and do not represent any particular order.

[0155] Those skilled in the art will also appreciate that, in addition to implementing the controller in pure computer-readable program code, it is entirely possible to implement the same functionality by logically programming the method steps in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, embedded microcontrollers, and the like. Therefore, such a controller can be considered a hardware component, and the devices included therein for implementing various functions can also be considered structures within the hardware component. Alternatively, the devices for implementing various functions can be considered both software modules implementing the method and structures within the hardware component.

[0156] This specification may be described in the general context of computer-executable instructions, such as program modules, executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, classes, and the like that perform specific tasks or implement specific abstract data types. This specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected through a communications network. In a distributed computing environment, program modules may be located in local and remote computer-readable storage media, including storage devices.

[0157] Through the description of the above embodiments, it can be seen that those skilled in the art can clearly understand that this specification can be implemented by means of software plus the necessary general hardware platform. Based on this understanding, the technical solution of this specification can essentially be embodied in the form of a software product. This computer software product can be stored in a storage medium such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, a mobile terminal, a server, or a network device, etc.) to execute the methods described in various embodiments or certain parts of the embodiments of this specification.

[0158] The various embodiments in this specification are described in a progressive manner. References to the common or similar parts of the various embodiments are sufficient. Each embodiment focuses on the differences from the other embodiments. This specification can be used in a variety of general-purpose or specialized computer system environments or configurations. For example, personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable electronic devices, network PCs, minicomputers, mainframe computers, and distributed computing environments that include any of the above systems or devices.

[0159] Although the present specification is described through embodiments, those skilled in the art will appreciate that there are many modifications and variations to the present specification without departing from the spirit of the present specification. It is intended that the appended claims include these modifications and variations without departing from the spirit of the present specification.

Claims

1. A method for determining the construction progress of a target building, characterized in that: include: Acquire a target image group; wherein the target image group includes a plurality of satellite images respectively acquired at a plurality of time points within a target time period; the satellite images include the target area to be monitored; Calling a preset image segmentation model to process the target image group to obtain a corresponding black-and-white image group; wherein the black-and-white image group includes a plurality of black-and-white images corresponding to the plurality of satellite images; the white image area in the black-and-white images is used to represent the target building area in the target area to be monitored; the preset image segmentation model is a model obtained by combining a convolutional neural network and a deconvolutional neural network; According to a preset processing rule, the KL divergence of the black and white images in the black and white image group and the area of ​​the white image region are calculated; Determining the construction progress of a target building within a target time period based on the KL divergence of the black and white images in the black and white image group and the area of ​​the white image area; the method comprising: sequentially comparing the KL divergence of the black and white images in the black and white image group with a preset KL divergence threshold, and comparing the area of ​​the white image area of ​​the black and white image with a preset area threshold to determine whether a trigger image is detected; wherein the trigger image is the first black and white image in the black and white image group whose KL divergence is greater than the preset KL divergence threshold and the area of ​​the white image area is greater than the preset area threshold; if it is determined that no trigger image is detected, determining that the target building is in a non-construction stage within the target time period; the preset KL divergence threshold and the preset area threshold are obtained by taking the average value after pre-training a large amount of historical sample data; Wherein, calculating the KL divergence of the black-and-white images in the black-and-white image group according to the preset processing rules includes: calculating the KL divergence of the current black-and-white image in the black-and-white image group according to the preset processing rules in the following manner: calculating the position distribution mean of the white image area and the position distribution mean of the black image area in the current black-and-white image; calculating the KL divergence of the current black-and-white image according to the position distribution mean of the white image area and the position distribution mean of the black image area in the current black-and-white image.

2. The method according to claim 1, characterized in that Arrange the plurality of satellite images in the target image group in chronological order according to the corresponding time points; Correspondingly, the plurality of black-and-white images in the black-and-white image group are arranged in chronological order according to the corresponding time points.

3. The method according to claim 2, characterized in that After calling a preset image segmentation model to process the target image group to obtain a corresponding black and white image group, the method further includes: Obtain the first-ranked satellite image in the target image group as the starting reference image; Performing image recognition on the starting reference image to determine interfering buildings that are not target buildings in the starting reference image; According to the interfering buildings, correction processing is performed on white image areas in a plurality of black and white images in the black and white image group.

4. The method according to claim 1, wherein The method further comprises: When it is determined that the trigger image is detected, it is determined that the target building is in the construction phase within the target time period.

5. The method according to claim 4, characterized in that After determining that the target building is in the construction start phase within the target time period, the method further includes: Extracting a trigger image and a black-and-white image in the black-and-white image group that is sorted after the trigger image, and constructing a corresponding progress image group; The construction progress between adjacent time points in the commencement phase within the target time period is determined by calculating and based on the KL divergence difference between adjacent images in the progress image group and the area difference of the white image region.

6. The method according to claim 1, characterized in that The target building includes at least one of the following: a highway, a power station, and a factory building.

7. A device for determining the construction progress of a target building, characterized in that: include: An acquisition module is configured to acquire a target image group, wherein the target image group includes a plurality of satellite images acquired at a plurality of time points within a target time period, and the satellite images include a target area to be monitored; a segmentation module, configured to invoke a preset image segmentation model to process the target image group to obtain a corresponding black-and-white image group; wherein the black-and-white image group includes a plurality of black-and-white images corresponding to the plurality of satellite images; the white image region in the black-and-white images is used to represent the target building region in the target area to be monitored; and the preset image segmentation model is a model obtained by combining a convolutional neural network and a deconvolutional neural network; a calculation module, configured to calculate the KL divergence of the black and white images and the area of ​​the white image region in the black and white image group according to a preset processing rule; A determination module is configured to determine the construction progress of a target building within a target time period based on the KL divergence of the black-and-white images in the black-and-white image group and the area of ​​the white image area; the determination module is specifically configured to: sequentially compare the KL divergence of the black-and-white images in the black-and-white image group with a preset KL divergence threshold, and compare the area of ​​the white image area of ​​the black-and-white image with a preset area threshold, to determine whether a trigger image is detected; wherein the trigger image is the first black-and-white image in the black-and-white image group whose KL divergence is greater than the preset KL divergence threshold and whose white image area is greater than the preset area threshold; if it is determined that no trigger image is detected, it is determined that the target building is in a non-construction stage within the target time period; the preset KL divergence threshold and the preset area threshold are obtained by taking the average value after pre-training a large amount of historical sample data; Among them, the calculation module specifically calculates the KL divergence of the current black and white image in the black and white image group in the following manner according to the preset processing rules: calculates the position distribution mean of the white image area and the position distribution mean of the black image area in the current black and white image; calculates the KL divergence of the current black and white image based on the position distribution mean of the white image area and the position distribution mean of the black image area in the current black and white image.

8. A server, characterized in that: The method comprises a processor and a memory for storing processor-executable instructions, wherein the processor implements the steps of the method according to any one of claims 1 to 6 when executing the instructions.

9. A computer-readable storage medium, characterized in that Computer instructions are stored thereon, and when the instructions are executed, the steps of the method according to any one of claims 1 to 6 are implemented.

10. An image processing method, characterized in that: include: Acquire a target image group; wherein the target image group includes a plurality of satellite images respectively acquired at a plurality of time points within a target time period; the satellite images include a target area; Calling a preset image segmentation model to process the target image group to obtain a corresponding black-and-white image group; wherein the black-and-white image group includes a plurality of black-and-white images corresponding to the plurality of satellite images; the white image area in the black-and-white images is used to represent the target object in the target area; the preset image segmentation model is a model obtained by combining a convolutional neural network and a deconvolutional neural network; According to a preset processing rule, the KL divergence of the black and white images in the black and white image group and the area of ​​the white image region are calculated; Determining a change trend of a target object within a target time period based on the KL divergence of the black-and-white images and the area of ​​the white image area in the black-and-white image group; the method comprising: sequentially comparing the KL divergence of the black-and-white images in the black-and-white image group with a preset KL divergence threshold, and comparing the area of ​​the white image area of ​​the black-and-white images with a preset area threshold, to determine whether a trigger image is detected; wherein the trigger image is the first black-and-white image in the black-and-white image group whose KL divergence is greater than the preset KL divergence threshold and whose white image area is greater than the preset area threshold; if it is determined that no trigger image is detected, determining that the change trend of the target object within the target time period is no change; the preset KL divergence threshold and the preset area threshold are obtained by taking the average value after pre-training a large amount of historical sample data; Wherein, calculating the KL divergence of the black-and-white images in the black-and-white image group according to the preset processing rules includes: calculating the KL divergence of the current black-and-white image in the black-and-white image group according to the preset processing rules in the following manner: calculating the position distribution mean of the white image area and the position distribution mean of the black image area in the current black-and-white image; calculating the KL divergence of the current black-and-white image according to the position distribution mean of the white image area and the position distribution mean of the black image area in the current black-and-white image.

Images