KR102484401B1 - Method for classifying building change in a user-participating recognition platform using aritificial intelligence and apparatus therefore - Google Patents

Abstract

Disclosed is a method and apparatus for classifying building changes using aerial photographic analysis based on artificial intelligence object recognition. A building change classification method according to an embodiment of the present invention is an artificial intelligence object recognition based building change classification method, with respect to a target area, a first aerial photograph at a first point in time, which is the current point of view, and a first point in time prior to the first point in time. analyzing second aerial photographs of two viewpoints and detecting buildings included in the first and second aerial photographs; Comparing shapes of the detected buildings of the first aerial photograph and the buildings of the second aerial photograph with respect to each of the buildings in the same location; and obtaining, for each of the buildings at the same location, a changed part at the first point of view based on a comparison result of a building shape.

Description

A method and device for classifying building changes in a user-participating recognition platform using artificial intelligence

The present invention relates to a technology for classifying changes in buildings using aerial photographic analysis, and more specifically, a method for distinguishing building changes that can identify changes in buildings and distances by analyzing aerial photographs based on artificial intelligence object recognition, and It's about the device.

In general, a Geographic Information System (GIS) refers to a computer-based system that uses and manages geographic data to solve problems related to space. When GIS was first developed in the early 1960s, GIS was nothing more than a computer-based program for processing map data. However, today, GIS is not only positioned as one of the important academic research fields in itself, but also is actively applied to each sector of society while being combined with the latest information technology.

A fundamental characteristic of GIS is that it can handle attribute information related to the location of objects in spatial, that is, geographic space. A map is the most efficient way to represent the location and properties of spatially distributed objects. Therefore, the most basic data in GIS construction is a digital map. Unlike the classic paper maps, digital maps can be said to be indexed by interpreting, numerically editing, and indexing various topographical data obtained from survey maps, aerial photographs, and satellite images.

The process of constructing a digital map is completed through several processes as follows. First, the paper map is converted into a digital map through digitizing or scanning, and then various input errors are corrected. Then, after transforming into a real coordinate system to suit the user's purpose through coordinate transformation, a topological structure is established to determine mutual location and correlation between spatial objects. Afterwards, attribute data related to each figure data is entered into the numerical map that has undergone the establishment of the topological structure.

Compared to paper maps, digital maps produced in this way enable faster and more accurate map retrieval, and are superior in information management and usability, enabling various planning and decision-making to be supported more effectively.

Embodiments of the present invention provide a building change classification method and apparatus capable of identifying a change in a building and a change in a street by analyzing an aerial photograph based on artificial intelligence object recognition.

A building change classification method according to an embodiment of the present invention is an artificial intelligence object recognition based building change classification method, with respect to a target area, a first aerial photograph at a first point in time, which is the current point of view, and a first point in time prior to the first point in time. analyzing second aerial photographs of two viewpoints and detecting buildings included in the first and second aerial photographs; Comparing shapes of the detected buildings of the first aerial photograph and the buildings of the second aerial photograph with respect to each of the buildings in the same location; and obtaining, for each of the buildings at the same location, a changed part at the first point of view based on a comparison result of a building shape.

The detecting step may include analyzing the first aerial photo and the second aerial photo of the target area to detect a road network included in the first aerial photo and the second aerial photo, and the comparing step may include: In the step of comparing the road network detected in the first aerial photograph and the road network detected in the second aerial photograph, and acquiring, a changed part of the road network at the first time point may be obtained based on the comparison result of the road network.

The detecting may include analyzing the first and second aerial photos of the target area using a preset segmentation technique, thereby analyzing the first and second aerial photos included in the first and second aerial photos. Buildings, roads, trees and forests can be distinguished and detected.

The detecting may include detecting a preset target building by analyzing the first aerial photograph and the second aerial photograph of the target area, and removing remaining parts except for the detected target building, thereby capturing the first aerial photograph and the second aerial photograph. Only the target building may be detected in each of the second aerial photographs.

Furthermore, the building change classification method according to an embodiment of the present invention may further include providing information on a building in which the changed portion is obtained when the changed portion is acquired at the first point of view.

A building change classification device according to an embodiment of the present invention is an artificial intelligence object recognition based building change classification device, with respect to a target area, a first aerial photograph at a first point in time, which is the current point of view, and a first point in time before the first point in time. a detector configured to analyze second aerial photographs of two viewpoints and detect buildings included in the first and second aerial photographs; a determination unit comparing the shapes of the detected buildings in the first aerial photograph with those of the second aerial photograph for each of the buildings in the same location; and an acquiring unit that obtains a changed part from the first viewpoint based on a comparison result of a building shape for each of the buildings at the same location.

The detection unit analyzes the first aerial photograph and the second aerial photograph of the target area to detect a road network included in the first aerial photograph and the second aerial photograph, and the determination unit analyzes the first aerial photograph. The road network detected in the second aerial photograph may be compared with the road network detected in the second aerial photograph, and the acquisition unit may obtain a changed part of the road network at the first point in time based on the comparison result of the road network.

The detection unit analyzes the first and second aerial photographs of the target area using a preset segmentation technique, so that the buildings included in the first and second aerial photographs; Roads, trees and forests can be distinguished and detected.

The detection unit detects a preset target building by analyzing the first and second aerial photographs of the target area, and removes remaining parts except for the detected target building, thereby capturing the first aerial photograph and the first aerial photograph. In each of the 2 aerial photographs, only the target building can be detected.

Furthermore, the apparatus for classifying building change according to an embodiment of the present invention may further include a provision unit that provides information on a building in which the changed portion is obtained when the changed portion is acquired at the first point of view.

According to embodiments of the present invention, changes in a specific building, for example, cracks or damage to a roof of a building, can be detected by analyzing aerial photographs based on artificial intelligence object recognition to determine changes in buildings and distances. In addition, it is possible to identify the damage situation of buildings before and after disasters. Accordingly, the present invention can provide such information to insurance companies and repair companies.

According to the embodiments of the present invention, since it is possible to determine the change in the street, it is possible to reduce the cost of examining street scenery, etc., to determine whether a road is illegally occupied, and to determine whether a public space is illegally occupied. there is.

The present invention can be used for various services such as urban planning, analysis of real estate commercial districts, and identification of illegal buildings violating building laws, and the information obtained through the present invention can be provided to public offices and specific private companies.

1 is an operation flowchart of a method for distinguishing building changes based on artificial intelligence object recognition according to an embodiment of the present invention.
Figure 2 shows an exemplary diagram for a first aerial photograph and a second aerial photograph.
3 shows an example of an object detection result in an aerial photograph.
Figure 4 shows the configuration of the building change classification device based on artificial intelligence object recognition according to an embodiment of the present invention.

Advantages and features of the present invention, and methods of achieving them, will become clear with reference to the detailed description of the following embodiments taken in conjunction with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below and will be implemented in various forms different from each other, only these embodiments make the disclosure of the present invention complete, and common knowledge in the art to which the present invention pertains. It is provided to completely inform the person who has the scope of the invention, and the present invention is only defined by the scope of the claims.

Terms used in this specification are for describing the embodiments and are not intended to limit the present invention. In this specification, singular forms also include plural forms unless specifically stated otherwise in a phrase. As used herein, "comprises" and/or "comprising" means that a stated component, step, operation, and/or element is present in the presence of one or more other components, steps, operations, and/or elements. or do not rule out additions.

Unless otherwise defined, all terms (including technical and scientific terms) used in this specification may be used in a meaning commonly understood by those of ordinary skill in the art to which the present invention belongs. In addition, terms defined in commonly used dictionaries are not interpreted ideally or excessively unless explicitly specifically defined.

Hereinafter, with reference to the accompanying drawings, preferred embodiments of the present invention will be described in more detail. The same reference numerals are used for the same components in the drawings, and redundant descriptions of the same components are omitted.

Embodiments of the present invention can detect changes in a specific building, for example, roof cracks or damage, etc., by analyzing aerial photographs based on AI object recognition to determine changes in buildings and distances The main point is to identify the damage situation of buildings before and after disasters.

At this time, the present invention receives an aerial photograph of a current point of view (hereinafter referred to as "first point of view") and a preset past point of view (hereinafter referred to as "second point of view") with respect to the target survey area, or It is possible to provide information on a building whose shape has changed by downloading, detecting a target building set by a user or manager through aerial photograph analysis, and comparing the shape of a building in the same location. Here, the provided information on the building may include latitude, longitude, location information, building information, and the like.

Furthermore, the present invention may detect changes in the road network by detecting the road network of the area to be surveyed through aerial photograph analysis as well as the target building, and comparing the road networks of the first and second points of view.

That is, the present invention can detect objects such as target buildings, road networks, trees, forests, etc. through analysis of aerial photographs of the first and second viewpoints, and based on the objects detected in the aerial photographs of the first viewpoint By comparing objects detected in aerial photographs of two points of view, a changed part of each object may be obtained.

In the present invention, processes such as aerial photograph analysis, object recognition, and shape change acquisition can be performed based on artificial intelligence object recognition. That is, the present invention can create each learning model or an integrated learning model that detects objects or obtains transformed parts for each object by training artificial intelligence using a preset training dataset. The learned model may be updated using a result obtained through aerial photographic analysis or may be updated using an additionally collected training dataset. Artificial intelligence may be implemented by a neural network such as a convolutional neural network (CNN), a deep learning neural network (DNN), etc., and these details are obvious to those skilled in the art, so detailed descriptions thereof is omitted.

1 is an operation flowchart for a method for distinguishing building changes based on artificial intelligence object recognition according to an embodiment of the present invention, which can be performed by a device in which an algorithm for the method of the present invention is installed.

Referring to FIG. 1, a method for classifying building changes according to an embodiment of the present invention is an aerial photograph (first aerial photograph) of a first viewpoint and an aerial photograph of a second viewpoint, which is a preset past viewpoint (with respect to an area to be surveyed). second aerial photo) is received or downloaded, and buildings included in the first and second aerial photos are detected by analyzing the first and second aerial photos based on artificial intelligence object recognition (S110).

Here, in step S110, the latitude and longitude of the coordinates of the target region are inputted by the user or administrator, the target region is set, and the first and second aerial photographs of the target region are received from the database. It is possible to detect specific buildings included in the first and second aerial photos, for example, preset target buildings, by analyzing the first and second aerial photos based on artificial intelligence object recognition.

Here, in step S110, the survey target region may be set by dragging the survey target region in the map data displayed on the screen by input of a user or manager.

For example, in step S110, as shown in FIG. 2, a first aerial photograph (before) and a second aerial photograph (after) may be received or downloaded, and target buildings may be detected from each of the first and second aerial photographs. there is.

In the present invention, the aerial photograph of the target area to be investigated may be an aerial photograph obtained by downloading only the target region from a database of a specific server, or an aerial photograph corresponding to the target region among aerial photographs downloaded and stored in a device providing the method of the present invention. It could be an aerial photo from a photo search. Of course, according to the present invention, both the aerial photograph of the first viewpoint and the aerial photograph of the second viewpoint may be stored in a corresponding device or may be downloaded from a specific server.

Target buildings in step S110 may include residential buildings, commercial buildings, buildings of a certain size or more, and these target buildings may be determined by an operator providing the technology or through selection of items capable of setting target buildings. can At this time, in step S110, information on the detected building is retrieved from a database in which information such as the purpose and use of the building according to the latitude and longitude of the aerial photograph and whether it is for home or commercial use is stored, so that the building is a target in the present invention. It is possible to determine whether a building is a building, and through this, a desired target building can be detected. Of course, according to circumstances, the present invention may detect a building of a certain size or larger without considering the use or purpose of such a building.

Furthermore, in step S110, preset target buildings are detected by analyzing the first and second aerial photographs of the area to be surveyed, and the remaining parts other than the detected target buildings, for example, including roads, parking lots, rivers, etc. Parts other than buildings are removed using a preset segmentation technique, and target buildings may be detected from an aerial photograph from which these parts are removed.

Depending on circumstances, in step S110, the first and second aerial photographs are analyzed using a segmentation technique, and through this, road networks, trees, forests, rivers, etc. as well as target buildings may be classified and detected. For example, as shown in FIG. 3 , the present invention can detect objects such as target buildings (red), roads (yellow), and forests (blue) from aerial photographs based on artificial intelligence object recognition.

In this case, the present invention may analyze and provide not only changes in buildings but also changes in road networks, changes in rivers, and changes in forests. Hereinafter, for convenience of description of the method of the present invention, only the target building is detected and the change of the target building is identified.

As described above, in step S110, a target building may be detected by recognizing each of the buildings in the first and second aerial photographs of the survey target area using artificial intelligence of a pre-learned learning model.

When the first-time buildings and the second-time buildings are detected in the area to be surveyed in step S110, the first-time buildings and the second-time buildings for each of the buildings at the same location are detected. The shape is compared, and based on the result of comparing the shape of the building, a portion where the shape of the building has changed at the first point in time is obtained (S120 and S130). That is, in step S120, roof shape difference information may be obtained by comparing the roof shapes of the buildings of the first viewpoint with the roof shapes of the buildings of the second viewpoint, and in step S130, information on the difference in roof shapes From the information, it is possible to know if the shape has changed for each of the target buildings.

Here, in step S120, the shapes of each of the buildings of the first point in time and the buildings of the second point in time detected using artificial intelligence of the pre-trained learning model may be compared. Specifically, in the shape of the buildings of the second point of view It is possible to compare how the shapes of the buildings at the first point in time have changed and whether or not they have changed.

Here, in step S130, it is possible to obtain information on a changed part for each of the buildings whose shape has changed through the comparison result in step S120. For example, in step S130, if the shape of the building at the location has changed within a certain range, it can be determined that the shape of the corresponding building has not changed. When it is determined that the building has changed, information on the change of the corresponding building or information determining that the corresponding building has changed may be obtained. For example, in operation S130 , information indicating that the shape or shape of a building has been changed and the size of the building has been reduced or expanded may be obtained with respect to a building at a specific location.

Step S130 may also be performed using artificial intelligence of a pre-learned learning model.

When information about a building whose shape has changed in step S130, for example, location information, latitude and longitude information, etc., is obtained, information about the changed part at the first point in time, that is, the building whose shape has changed compared to the past, is provided. Do (S140).

Here, in step S140, information on a building whose shape has been changed among target buildings in the survey target area may be provided in the form of a specific file, and the provided information includes address information, building information, use information, It may include at least one of change information and latitude and longitude information.

Furthermore, the method according to the embodiment of the present invention recognizes road networks, farm waterways, farm roads, rivers, forests, etc. included in the survey target area as objects, and can know how the objects have changed at the current time point. Of course, whether or not each object is changed can also be performed through the above-described process, and when an object is changed at the same location through object recognition, information on how the object has been changed may be provided.

Furthermore, the method according to an embodiment of the present invention may additionally analyze the surrounding street scenery of the target building with respect to the object where the change has occurred, for example, the target building, and additionally determine the use of the target building through the surrounding street scenery analysis. may be Of course, the surrounding street scenery in the method of the present invention may include surrounding building information, road information, etc., and the information on the surrounding street scenery is not limited to the above information, and all types of information that can be used in the method of the present invention information may be included.

As such, the method according to an embodiment of the present invention analyzes aerial photographs based on artificial intelligence object recognition to identify changes in buildings and distances, such as changes in a specific building, such as cracks or damage to the roof of the building. can be detected, and the damage situation of buildings before and after disasters can be identified. Accordingly, the present invention can provide such information to insurance companies and repair companies.

In addition, since the method according to an embodiment of the present invention can determine the change in the street, it is possible to reduce the cost of investigating the street scenery, etc., it is possible to determine whether or not the road is illegally occupied, and whether or not the public space is illegally occupied. can also figure out.

The present invention can be used for various services such as urban planning, analysis of real estate commercial districts, and identification of illegal buildings violating building laws, and the information obtained through the present invention can be provided to public offices and specific private companies.

Figure 4 shows the configuration of the building change classification device based on artificial intelligence object recognition according to an embodiment of the present invention, and shows the conceptual configuration of the device performing the method of Figs. 1 to 3.

Referring to FIG. 4 , the apparatus 400 according to an embodiment of the present invention includes a receiving unit 410, a detecting unit 420, a determining unit 430, an acquiring unit 440, a providing unit 450 and a database (DB ) (460).

DB 460 is a means for storing data for carrying out the present invention, and can store various data such as aerial photographs, map data, artificial intelligence object recognition-based algorithms, learning models, and data providing algorithms according to object changes, In addition to these data, all kinds of data required for carrying out the present invention can be stored.

Here, data stored in the DB 460 may be updated through a wired/wireless network, and the data update cycle may be determined by a business operator or individual providing the technology of the present invention.

The receiving unit 410 sets the survey target area by receiving the latitude and longitude of the coordinates of the survey target area from the user or administrator, and receives an aerial photograph of the survey target area at a first point of view and a second point in time. do.

Here, the receiver 410 may set the survey target region by dragging the survey target region in the map data according to a user's or manager's input.

The detection unit 420 analyzes the first and second aerial photos of the area to be investigated based on artificial intelligence object recognition to determine buildings included in the first and second aerial photos, for example, preset target buildings. detect

At this time, the detection unit 420 retrieves information about the detected building from a database in which information such as the purpose and use of the building according to the latitude and longitude of the aerial photograph, and whether it is for home or commercial use is stored, so that the building is found in the present invention. It can be determined whether it is a target building, and through this, a desired target building can be detected.

Depending on circumstances, the detection unit 420 removes parts other than the detected target building, for example, non-building parts including roads, parking lots, rivers, etc., using a preset segmentation technique, and these parts Target buildings may be detected in the removed aerial photograph.

Depending on circumstances, the detection unit 420 may analyze the first and second aerial photographs by using a segmentation technique, and may classify and detect target buildings as well as road networks, trees, forests, and rivers through the analysis.

When the target buildings of the first viewpoint and the target buildings of the second viewpoint are detected by the detection unit 420, the determination unit 430 determines the buildings of the first viewpoint and the buildings of the second viewpoint with respect to the buildings at the same location, respectively. Compare the shape of each of them.

Here, the determination unit 430 compares the shapes of the buildings of the first point of view with the shapes of the buildings of the second point of time detected using artificial intelligence of the pre-learned learning model, thereby determining the shape of the buildings of the second point of view. It is possible to compare how the shapes of the buildings at the first point in time have changed and whether or not they have changed.

The acquisition unit 440 acquires a part where the shape of the building has changed at the first point of view based on the comparison result of the building shape.

When information about a building whose shape has changed, for example, location information, latitude and longitude information, etc. is acquired, the provision unit 450 provides information about a building whose shape has changed at the first point in time, that is, a building whose shape has changed compared to the past. provides

Here, the provision unit 450 may provide information on a building whose shape has been changed among target buildings in the survey target area in the form of a specific file, and the provided information includes address information, building information, It may include at least one of usage information, change information, and latitude and longitude information.

Although the description is omitted in the device of FIG. 4, each constituent means constituting FIG. 4 may include all of the contents described in the method of FIGS. 1 to 3, which is obvious to those skilled in the art. Do.

The devices described above may be implemented as hardware components, software components, and/or a combination of hardware components and software components. For example, devices and components described in the embodiments may include, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable array (FPA), It may be implemented using one or more general purpose or special purpose computers, such as a programmable logic unit (PLU), microprocessor, or any other device capable of executing and responding to instructions. A processing device may run an operating system (OS) and one or more software applications running on the operating system. A processing device may also access, store, manipulate, process, and generate data in response to execution of software. For convenience of understanding, there are cases in which one processing device is used, but those skilled in the art will understand that the processing device includes a plurality of processing elements and/or a plurality of types of processing elements. It can be seen that it can include. For example, a processing device may include a plurality of processors or a processor and a controller. Other processing configurations are also possible, such as parallel processors.

Software may include a computer program, code, instructions, or a combination of one or more of the foregoing, which configures a processing device to operate as desired or processes independently or collectively. The device can be commanded. Software and/or data may be any tangible machine, component, physical device, virtual equipment, computer storage medium or device, intended to be interpreted by or provide instructions or data to a processing device. can be embodied in The software may be distributed on networked computer systems and stored or executed in a distributed manner. Software and data may be stored on one or more computer readable media.

The method according to the embodiment may be implemented in the form of program instructions that can be executed through various computer means and recorded on a computer readable medium. The computer readable medium may include program instructions, data files, data structures, etc. alone or in combination. Program commands recorded on the medium may be specially designed and configured for the embodiment or may be known and usable to those skilled in computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks and magnetic tapes, optical media such as CD-ROMs and DVDs, and magnetic media such as floptical disks. - includes hardware devices specially configured to store and execute program instructions, such as magneto-optical media, and ROM, RAM, flash memory, and the like. Examples of program instructions include high-level language codes that can be executed by a computer using an interpreter, as well as machine language codes such as those produced by a compiler.

As described above, although the embodiments have been described with limited examples and drawings, those skilled in the art can make various modifications and variations from the above description. For example, the described techniques may be performed in an order different from the method described, and/or components of the described system, structure, device, circuit, etc. may be combined or combined in a different form than the method described, or other components may be used. Or even if it is replaced or substituted by equivalents, appropriate results can be achieved.

Therefore, other implementations, other embodiments, and equivalents of the claims are within the scope of the following claims.

Claims (10)

In the building change classification method based on artificial intelligence object recognition,
For the target area, the first aerial photo of the first time point, which is the current point of time, and the second aerial photo of the second time point before the first point of time are analyzed using artificial intelligence of a pre-learned learning model to analyze the first aerial photo. and detecting a building included in the second aerial photograph;
Comparing shapes of the detected buildings of the first aerial photograph and the buildings of the second aerial photograph using artificial intelligence of the pre-learned learning model for each of the buildings in the same location;
Obtaining a changed part at the first point in time based on a comparison result of a building shape using AI of the pre-learned learning model for each of the buildings at the same location; and
Including providing information on a building in which the changed part is obtained when the changed part is obtained from the first point of view,
The step of detecting
Setting the target area dragged from map data by input of a user or administrator;
The step of detecting
With respect to the target area, buildings, roads, trees, and forests included in the first and second aerial photos are analyzed by analyzing the first and second aerial photos using a preset segmentation technique. In the case of detecting a preset target building, the first aerial photograph and the second aerial photograph are analyzed to detect a preset target building, and the remaining parts except for the detected target building are removed. Detecting only the target building in each of the first aerial photograph and the second aerial photograph;
The step of providing the information is
Provides building address information, building information, use information, change information, and latitude and longitude information of a building whose shape is deformed in the target area in the form of a specific file,
The step of detecting
When detecting a road network, with respect to the target area, analyzing the first aerial photo and the second aerial photo to detect a road network included in the first aerial photo and the second aerial photo;
The step of comparing
Comparing the road network detected in the first aerial photograph with the road network detected in the second aerial photograph;
The step of obtaining
Obtaining a changed part of the road network at the first point in time based on the comparison result of the road network;
As artificial intelligence is learned using a preset learning dataset, an object is detected or an artificial intelligence of each learning model or an integrated learning model that acquires a transformed part for each object is created, and the generated pre-learned A method for distinguishing building changes in which the artificial intelligence of the learning model is updated using the output derived from aerial photo analysis and is updated using the additionally collected learning dataset.
delete delete delete delete In the building change classification device based on artificial intelligence object recognition,
For the target area, the first aerial photo of the first time point, which is the current point of time, and the second aerial photo of the second time point before the first point of time are analyzed using artificial intelligence of a pre-learned learning model to analyze the first aerial photo. and a detector detecting a building included in the second aerial photograph;
a determination unit comparing the shape of the detected building of the first aerial photograph and the building of the second aerial photograph using artificial intelligence of the pre-learned learning model for each of the buildings in the same location;
For each of the buildings at the same location, an acquisition unit for acquiring a changed part at the first point in time based on a comparison result of a building shape using artificial intelligence of the pre-learned learning model; and
Including a provision unit for providing location information on a building from which the changed portion is obtained when the changed portion is acquired at the first point of view,
the detection unit
Setting the target area dragged from map data by input of a user or administrator;
the detection unit
With respect to the target area, buildings, roads, trees, and forests included in the first and second aerial photographs are analyzed by analyzing the first and second aerial photographs using a preset segmentation technique. Classification and detection, in case of detecting a preset target building, the first aerial photograph and the second aerial photograph are analyzed to detect a preset target building, and by removing the remaining parts except for the detected target building, Detecting only the target building in each of the first aerial photograph and the second aerial photograph;
The provisioning department
Provides building address information, building information, use information, change information, and latitude and longitude information of a building whose shape is deformed in the target area in the form of a specific file, and the detection unit
When detecting a road network, with respect to the target area, analyzing the first aerial photograph and the second aerial photograph to detect a road network included in the first aerial photograph and the second aerial photograph;
The judge
Comparing the road network detected in the first aerial photograph with the road network detected in the second aerial photograph;
the acquisition unit
Obtaining a changed part of the road network at the first point in time based on the comparison result of the road network;
As artificial intelligence is learned using a preset learning dataset, an object is detected or an artificial intelligence of each learning model or an integrated learning model that obtains a transformed part for each object is created, and the generated pre-learned The artificial intelligence of the learning model is updated using the output derived from the aerial photograph analysis, and the building change classification device is updated using the additionally collected learning dataset.
delete delete delete delete

Images