AI-based historical building digital data processing method and system

Abstract

The present invention discloses a method and system for processing digital data of historical buildings based on AI, the method comprising: obtaining real-time sensing information of a target historical building through a sensor; generating a digital building model corresponding to the target historical building based on a modeling algorithm according to the real-time sensing information; completing missing parts of the digital building model based on a neural network algorithm to obtain a completed digital model; predicting the risk location and risk type of the target historical building based on the completed digital model and the real-time sensing information based on a neural network algorithm. It can be seen that the present invention can improve the modeling efficiency and modeling accuracy of the digital model of historical buildings, and realize more intelligent risk monitoring of historical buildings, reduce monitoring costs, and improve monitoring effects.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to an AI-based historical building digital data processing method and system. Background Art

[0002] With the development of the protection demand for cultural heritage such as ancient buildings or ancient cultural relics and murals, and the improvement of digital technology, more and more historical building protection agencies have begun to use digital technology to realize data collection and risk analysis of historical buildings. Among them, how to improve the modeling accuracy and risk monitoring timeliness of historical buildings has become an important technical issue. However, when realizing the digitization of historical buildings, the existing technology does not consider combining AI algorithms to improve modeling accuracy and risk monitoring capabilities. Therefore, the accuracy of its digital model is insufficient, and the aesthetics or viewing experience of the display to users is also lacking, and risk monitoring cannot be effectively realized. It can be seen that the existing technology has defects that need to be solved urgently. Summary of the invention

[0003] The technical problem to be solved by the present invention is to provide an AI-based historical building digital data processing method and system, which can improve the modeling efficiency and modeling accuracy of the digital model of the historical building, and realize more intelligent risk monitoring of the historical building, reduce the monitoring cost and improve the monitoring effect.

[0004] In order to solve the above technical problems, the first aspect of the present invention discloses a method for processing digital data of historical buildings based on AI, the method comprising:

[0005] Acquire real-time sensing information of target historical buildings through sensors;

[0006] Generate a digital building model corresponding to the target historical building based on the real-time sensing information and a modeling algorithm;

[0007] Based on a neural network algorithm, the missing parts of the building digital model are completed to obtain a completed digital model;

[0008] Based on a neural network algorithm, the risk location and risk type of the target historical building are predicted according to the completed digital model and the real-time sensor information.

[0009] As an optional embodiment, in the first aspect of the present invention, the real-time sensing information includes image data, sound wave information, ultrasonic wave reflection information and light reflection information.

[0010] As an optional implementation, in the first aspect of the present invention, generating a digital building model corresponding to the target historical building based on the real-time sensing information and a modeling algorithm comprises:

[0011] Acquire a preset historical digital model corresponding to the target historical building;

[0012] According to the real-time sensing information, based on a three-dimensional modeling algorithm, a real-time three-dimensional model corresponding to the target historical building is generated;

[0013] The historical digital model is updated according to the real-time three-dimensional model to obtain a building digital model corresponding to the target historical building.

[0014] As an optional implementation, in the first aspect of the present invention, updating the historical digital model according to the real-time three-dimensional model to obtain the building digital model corresponding to the target historical building includes:

[0015] aligning the real-time three-dimensional model and the historical digital model into the same coordinate system;

[0016] For any spatial position in the coordinate system including both the real-time three-dimensional model and the historical digital model, determining a first model portion of the real-time three-dimensional model and a second model portion of the historical digital model in the spatial position;

[0017] Calculating an image difference and a three-dimensional difference between the first model part and the second model part;

[0018] Calculating a weighted average of the image difference and the three-dimensional difference to obtain a position update parameter corresponding to the spatial position;

[0019] Determine the spatial positions whose position update parameters are all greater than a first parameter threshold as update positions;

[0020] Based on the historical digital model, the model part corresponding to each updated position of the historical digital model is replaced with the model part of the corresponding position in the real-time three-dimensional model to obtain the building digital model corresponding to the target historical building.

[0021] As an optional implementation, in the first aspect of the present invention, the method of completing the missing parts of the building digital model based on a neural network algorithm to obtain a completed digital model includes:

[0022] Framing the building digital model based on multiple viewpoints and multiple angles to obtain multiple framing images;

[0023] Inputting each of the framing images into a preset defect recognition neural network model to obtain the image defect portion corresponding to each of the framing images;

[0024] According to the missing image parts corresponding to all the framing images, identifying the missing model parts in the digital building model;

[0025] According to the historical digital model records corresponding to the target historical building and a preset structural prediction algorithm, the model shape of each missing model part is predicted to obtain a predicted shape result;

[0026] Each missing model part of the building digital model is completed according to the predicted shape result to obtain a completed digital model.

[0027] As an optional implementation, in the first aspect of the present invention, identifying the missing model part in the digital building model according to the missing image parts corresponding to all the framing images includes:

[0028] Based on the correspondence between the preset model parts and the possibility of missing, a plurality of possible missing parts in the building digital model are determined; the possible missing parts are building connection parts, load-bearing parts in the building, non-visible surface parts of the building, or low-visibility surface parts of the building;

[0029] For each of the possible missing parts, calculating the number of occurrences of the possible missing part in the image missing parts corresponding to all the framed images;

[0030] If the number of occurrences is greater than a preset number threshold, the possible missing part is determined as a missing model part.

[0031] As an optional implementation, in the first aspect of the present invention, predicting the model shape of each missing model part according to the historical digital model record corresponding to the target historical building and a preset structure prediction algorithm to obtain a predicted shape result includes:

[0032] For each missing model part, determining a plurality of nearby model parts corresponding to the missing model part; the nearby model parts are model parts whose distances to the missing model part are less than a preset distance threshold;

[0033] Inputting all the nearby model parts and the corresponding model positions into a trained shape prediction neural network to obtain a predicted model shape corresponding to the missing model part; the shape prediction neural network is trained by a training data set including a plurality of training nearby model parts and corresponding specific area shape annotations; the training data set is obtained by annotating the historical digital model records corresponding to the target historical building;

[0034] The predicted model shapes corresponding to all the missing model parts are determined as predicted shape results.

[0035] As an optional implementation, in the first aspect of the present invention, the neural network algorithm is used to predict the risk location and risk type of the target historical building according to the completed digital model and the real-time sensor information, including:

[0036] Input the real-time sensing information corresponding to each missing model part in the completed digital model into the risk prediction neural network model corresponding to the corresponding data type to obtain the risk prediction probability and risk prediction type corresponding to each missing model part;

[0037] Input the real-time sensing information corresponding to each update position in the completed digital model into the risk prediction neural network model corresponding to the corresponding data type to obtain the risk prediction probability and risk prediction type corresponding to each update position;

[0038] All the risk prediction types whose risk prediction probabilities are greater than a preset probability threshold are determined as risk types of the target historical building, and the corresponding missing model parts or the updated positions are determined as corresponding risk positions.

[0039] A second aspect of an embodiment of the present invention discloses an AI-based historical building digital data processing system, the system comprising:

[0040] An acquisition module, used for acquiring real-time sensing information of a target historical building through sensors;

[0041] A generation module, used to generate a digital building model corresponding to the target historical building according to the real-time sensing information and based on a modeling algorithm;

[0042] A completion module, used to complete the missing parts of the building digital model based on a neural network algorithm to obtain a completed digital model;

[0043] A prediction module is used to predict the risk location and risk type of the target historical building based on a neural network algorithm according to the completed digital model and the real-time sensor information.

[0044] As an optional embodiment, in the second aspect of the present invention, the real-time sensing information includes image data, sound wave information, ultrasonic wave reflection information and light reflection information.

[0045] As an optional implementation, in the second aspect of the present invention, the specific manner in which the generation module generates the digital building model corresponding to the target historical building according to the real-time sensing information and based on the modeling algorithm includes:

[0046] Acquire a preset historical digital model corresponding to the target historical building;

[0047] According to the real-time sensing information, based on a three-dimensional modeling algorithm, a real-time three-dimensional model corresponding to the target historical building is generated;

[0048] The historical digital model is updated according to the real-time three-dimensional model to obtain a building digital model corresponding to the target historical building.

[0049] As an optional implementation, in the second aspect of the present invention, the specific manner in which the generation module updates the historical digital model according to the real-time three-dimensional model to obtain the building digital model corresponding to the target historical building includes:

[0050] aligning the real-time three-dimensional model and the historical digital model into the same coordinate system;

[0051] For any spatial position in the coordinate system including both the real-time three-dimensional model and the historical digital model, determining a first model portion of the real-time three-dimensional model and a second model portion of the historical digital model in the spatial position;

[0052] Calculating an image difference and a three-dimensional difference between the first model part and the second model part;

[0053] Calculating a weighted average of the image difference and the three-dimensional difference to obtain a position update parameter corresponding to the spatial position;

[0054] Determine the spatial positions whose position update parameters are all greater than a first parameter threshold as update positions;

[0055] Based on the historical digital model, the model part corresponding to each updated position of the historical digital model is replaced with the model part of the corresponding position in the real-time three-dimensional model to obtain the building digital model corresponding to the target historical building.

[0056] As an optional implementation, in the second aspect of the present invention, the completion module completes the missing parts of the building digital model based on a neural network algorithm, and a specific method for obtaining the completed digital model includes:

[0057] Framing the building digital model based on multiple viewpoints and multiple angles to obtain multiple framing images;

[0058] Inputting each of the framing images into a preset defect recognition neural network model to obtain the image defect portion corresponding to each of the framing images;

[0059] According to the missing image parts corresponding to all the framing images, identifying the missing model parts in the digital building model;

[0060] According to the historical digital model records corresponding to the target historical building and a preset structural prediction algorithm, the model shape of each missing model part is predicted to obtain a predicted shape result;

[0061] Each missing model part of the building digital model is completed according to the predicted shape result to obtain a completed digital model.

[0062] As an optional implementation, in the second aspect of the present invention, the specific manner in which the completion module identifies the missing model portion in the building digital model according to the missing image portions corresponding to all the viewfinder images includes:

[0063] Based on the correspondence between the preset model parts and the possibility of missing, a plurality of possible missing parts in the building digital model are determined; the possible missing parts are building connection parts, load-bearing parts in the building, non-visible surface parts of the building, or low-visibility surface parts of the building;

[0064] For each of the possible missing parts, calculating the number of occurrences of the possible missing part in the image missing parts corresponding to all the framed images;

[0065] If the number of occurrences is greater than a preset number threshold, the possible missing part is determined as a missing model part.

[0066] As an optional implementation, in the second aspect of the present invention, the completion module predicts the model shape of each missing model part according to the historical digital model record corresponding to the target historical building and a preset structural prediction algorithm, and obtains the specific method of the predicted shape result, including:

[0067] For each missing model part, determining a plurality of nearby model parts corresponding to the missing model part; the nearby model parts are model parts whose distances to the missing model part are less than a preset distance threshold;

[0068] Inputting all the nearby model parts and the corresponding model positions into a trained shape prediction neural network to obtain a predicted model shape corresponding to the missing model part; the shape prediction neural network is trained by a training data set including a plurality of training nearby model parts and corresponding specific area shape annotations; the training data set is obtained by annotating the historical digital model records corresponding to the target historical building;

[0069] The predicted model shapes corresponding to all the missing model parts are determined as predicted shape results.

[0070] As an optional implementation, in the second aspect of the present invention, the prediction module predicts the risk location and risk type of the target historical building based on the neural network algorithm according to the completed digital model and the real-time sensor information, including:

[0071] Input the real-time sensing information corresponding to each missing model part in the completed digital model into the risk prediction neural network model corresponding to the corresponding data type to obtain the risk prediction probability and risk prediction type corresponding to each missing model part;

[0072] Input the real-time sensing information corresponding to each update position in the completed digital model into the risk prediction neural network model corresponding to the corresponding data type to obtain the risk prediction probability and risk prediction type corresponding to each update position;

[0073] All the risk prediction types whose risk prediction probabilities are greater than a preset probability threshold are determined as risk types of the target historical building, and the corresponding missing model parts or the updated positions are determined as corresponding risk positions.

[0074] The third aspect of the present invention discloses another AI-based historical building digital data processing system, the system comprising:

[0075] A memory storing executable program code;

[0076] a processor coupled to the memory;

[0077] The processor calls the executable program code stored in the memory to execute part or all of the steps in the AI-based historical building digital data processing method disclosed in the first aspect of the present invention.

[0078] The fourth aspect of the present invention discloses a computer storage medium, which stores computer instructions. When the computer instructions are called, they are used to execute some or all of the steps in the AI-based historical building digital data processing method disclosed in the first aspect of the present invention.

[0079] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0080] The present invention can generate a digital building model corresponding to a target historical building based on a modeling algorithm according to real-time sensor information, and then complete the missing parts of the digital building model based on a neural network algorithm, thereby accurately predicting the risk location and risk type of the target historical building, thereby improving the modeling efficiency and accuracy of the digital model of the historical building, and realizing more intelligent risk monitoring of historical buildings, reducing monitoring costs and improving monitoring effects. BRIEF DESCRIPTION OF THE DRAWINGS

[0081] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0082] Figure 1 It is a flowchart of a method for processing digital data of historical buildings based on AI disclosed in an embodiment of the present invention.

[0083] Figure 2 It is a structural schematic diagram of an AI-based historical building digital data processing system disclosed in an embodiment of the present invention.

[0084] Figure 3 It is a structural schematic diagram of another AI-based historical building digital data processing system disclosed in an embodiment of the present invention. DETAILED DESCRIPTION

[0085] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0086] The terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, device, product or equipment that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units that are not listed, or may optionally include other steps or units that are inherent to these processes, methods, products or equipment.

[0087] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present invention. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0088] The present invention discloses an AI-based historical building digital data processing method and system, which can generate a building digital model corresponding to a target historical building based on a modeling algorithm according to real-time sensor information, and then complete the missing parts of the building digital model based on a neural network algorithm, and accurately predict the risk location and risk type of the target historical building, thereby improving the modeling efficiency and modeling accuracy of the digital model of the historical building, and realizing more intelligent risk monitoring of historical buildings, reducing monitoring costs, and improving monitoring effects. The following are detailed descriptions.

[0089] Embodiment 1

[0090] See also Figure 1 , Figure 1 : is a flowchart of a method for processing digital data of historical buildings based on AI disclosed in an embodiment of the present invention. Figure 1 The AI-based historical building digital data processing method described can be applied to a data processing device / data processing server (wherein the server includes a local processing server or a cloud processing server). Figure 1 As shown, the AI-based historical building digital data processing method may include the following operations:

[0091] 101. Obtain real-time sensing information of the target historical building through sensors.

[0092] Specifically, in a specific implementation plan, sensors such as cameras, sound sensors, ultrasonic reflection sensors and light reflection sensors are placed around and inside immovable cultural relics, historical buildings, traditional buildings, and ancient trees. By deploying a sensor network on the building, the structural health, environmental changes and other information of the building can be monitored in real time. The data can be analyzed and predicted using AI algorithms, so that potential safety hazards of the building can be discovered in a timely manner and early warnings can be issued to the information management platform. For example, the camera can be controlled to take pictures of historical buildings once an hour and update the pictures to the information management platform. The multimodal capabilities of the AI ​​model can be used to identify the current status of the historical building. If there is a risk of damage such as breakage or cracks, an early warning will be issued to the information management platform.

[0093] 102. According to the real-time sensor information and based on the modeling algorithm, a digital model of the building corresponding to the target historical building is generated.

[0094] 103. Based on the neural network algorithm, the missing parts of the building digital model are completed to obtain the completed digital model.

[0095] 104. Based on the neural network algorithm, the risk location and risk type of the target historical building are predicted according to the completed digital model and real-time sensor information.

[0096] Specifically, in order to realize the above scheme, multiple databases can be further established, and the completed digital models can be integrated into these databases to display more intuitive digital models of historical buildings to users. Among them, the basic spatial information database includes digital line drawings, government map databases, Internet map databases, image databases, digital elevation models, administrative divisions, place names and addresses, and other information. The historical building thematic database realizes the verification, scanning, extraction, storage, etc. of historical building thematic materials and data such as immovable cultural relics, historical buildings and traditional style buildings, ancient trees and famous trees, and establishes a historical building database. Specifically, the historical building thematic data needs include thematic data involved in the data census, spatial positioning of immovable cultural relics information at all levels and types, and marking them on electronic maps, so that this information can be queried intuitively and vividly on the electronic map. The historical and cultural city protection database includes this database when the area is a historical and cultural city. It is included in the historical and cultural city planning and related achievements, mainly including: famous city protection framework planning, urban historical and cultural protection planning, historical and cultural towns, villages, historical and cultural protection areas, historical and cultural blocks, historical style areas, historical and cultural city protection lines, etc. Optionally, other databases may include real-life images of immovable cultural relics, historical buildings and traditional style buildings, ancient trees and famous trees, VR panoramas, laser point clouds, real-life 3D models, etc.

[0097] Based on the above specific implementation plan, various databases can be used as the basis to comprehensively express the spatial distribution, photos, characteristics and attributes of immovable cultural relics, historical buildings and traditional style buildings, and ancient and famous trees.

[0098] In order to implement the above technical solutions, a field collection applet was developed for field collectors of historical buildings. It can collect cultural relics information using different methods such as photos, text, voice, and notebooks, and record the current geographical location, which is displayed synchronously on the map of the collection phone to provide support for collectors to view the collection results. The field collection applet mainly realizes the collection of thematic data of historical buildings. To facilitate the use of operators, the field collection applet includes modules such as creating new tasks, loading tasks, map view, track recording, exiting, data uploading, and census form loading. The map view includes functions such as adding collection points, adding photos, querying collection points, adding new bookmarks, querying bookmarks, measuring distances, zooming in and out, and positioning.

[0099] At the same time, a data management subsystem has also been developed. Its main task is to manage the comprehensive database, which mainly includes high-precision data coordinate conversion, census data management, census and historical data storage management, basic spatial information data storage management, other new types of surveying and mapping data storage management, real-time storage of camera collected data, data editing and maintenance, data resource directory management, symbol library management, metadata management, data update, data backup management and other functions.

[0100] At the same time, an information management subsystem has also been developed to display, manage, analyze, and count the special data of historical buildings formed after the data management subsystem has been checked. It mainly provides services for the daily management departments of historical buildings, including the comprehensive management and diversified display of immovable cultural relics, historical buildings, and traditional style buildings; the main functions are: common GIS functions, heritage information editing, heritage information display, AI real-time warning, archive management, print output and other functions.

[0101] It can be seen that the above-mentioned embodiments of the invention can generate a digital building model corresponding to the target historical building based on the modeling algorithm according to the real-time sensor information, and then complete the missing parts of the digital building model based on the neural network algorithm, and thereby accurately predict the risk location and risk type of the target historical building, thereby improving the modeling efficiency and modeling accuracy of the digital model of the historical building, and realizing more intelligent risk monitoring of historical buildings, reducing monitoring costs and improving monitoring effects.

[0102] As an optional embodiment, in the above steps, the real-time sensing information includes image data, sound wave information, ultrasonic wave reflection information and light reflection information.

[0103] It can be seen that through the above optional embodiments, the content of real-time sensing information is limited, which can fully characterize the structural characteristics of historical buildings and be used for more accurate modeling later, thereby assisting in improving the modeling efficiency and modeling accuracy of digital models of historical buildings, and realizing more intelligent risk monitoring of historical buildings, reducing monitoring costs and improving monitoring effects.

[0104] As an optional embodiment, in the above steps, generating a digital building model corresponding to the target historical building based on the real-time sensor information and the modeling algorithm includes:

[0105] Obtaining a preset historical digital model corresponding to the target historical building;

[0106] According to the real-time sensor information and based on the 3D modeling algorithm, a real-time 3D model corresponding to the target historical building is generated;

[0107] The historical digital model is updated according to the real-time three-dimensional model to obtain the building digital model corresponding to the target historical building.

[0108] It can be seen that through the above-mentioned optional embodiments, a real-time three-dimensional model corresponding to the target historical building can be generated based on the modeling algorithm according to real-time sensing information to update the historical digital model to obtain a building digital model corresponding to the target historical building, which can be used for missing completion and risk analysis subsequently, to assist in improving the modeling efficiency and modeling accuracy of the digital model of the historical building, and to achieve more intelligent risk monitoring of historical buildings, reduce monitoring costs, and improve monitoring effects.

[0109] As an optional embodiment, in the above step, updating the historical digital model according to the real-time three-dimensional model to obtain the building digital model corresponding to the target historical building includes:

[0110] Align the real-time 3D model and the historical digital model into the same coordinate system;

[0111] For any spatial position in the coordinate system including both the real-time three-dimensional model and the historical digital model, determining a first model portion of the real-time three-dimensional model and a second model portion of the historical digital model in the spatial position;

[0112] Calculating an image difference and a three-dimensional difference between the first model part and the second model part;

[0113] Calculate the weighted average of the image difference and the three-dimensional difference to obtain the position update parameter corresponding to the spatial position;

[0114] Determine the spatial positions where all position update parameters are greater than a first parameter threshold as update positions;

[0115] Based on the historical digital model, the model part corresponding to the historical digital model corresponding to each updated position is replaced with the model part of the corresponding position in the real-time three-dimensional model to obtain the building digital model corresponding to the target historical building.

[0116] It can be seen that through the above-mentioned optional embodiments, the necessity of updating the position can be determined by calculating the image difference and three-dimensional difference of the model part in the spatial position that includes both the real-time three-dimensional model and the historical digital model, and the part with high necessity is updated and replaced to obtain the digital model of the building, which is subsequently used for missing completion and risk analysis, to assist in improving the modeling efficiency and modeling accuracy of the digital model of the historical building, and to achieve more intelligent risk monitoring of historical buildings, reduce monitoring costs, and improve monitoring effects.

[0117] As an optional embodiment, in the above steps, based on a neural network algorithm, the missing parts of the building digital model are completed to obtain a completed digital model, including:

[0118] Framing the building digital model based on multiple viewpoints and multiple angles to obtain multiple framing images;

[0119] Input each framed image into a preset missing part recognition neural network model to obtain the missing part of the image corresponding to each framed image;

[0120] According to the missing parts of the images corresponding to all the framed images, the missing model parts in the building digital model are identified;

[0121] According to the historical digital model records corresponding to the target historical building and the preset structural prediction algorithm, the model shape of each missing model part is predicted to obtain the predicted shape result;

[0122] Each missing model part of the building digital model is completed according to the predicted shape result to obtain a completed digital model.

[0123] It can be seen that through the above-mentioned optional embodiments, it is possible to identify the parts with structural missing in the image based on the framed images obtained based on multi-angle and multi-viewpoint framing through a neural network model, and predict the shape of the missing parts based on historical model records to complete them, thereby assisting in improving the modeling efficiency and modeling accuracy of the digital model of historical buildings, and realizing more intelligent risk monitoring of historical buildings, reducing monitoring costs and improving monitoring effects.

[0124] As an optional embodiment, in the above step, identifying the missing model part in the building digital model according to the missing image parts corresponding to all the framing images includes:

[0125] Based on the correspondence between the preset model parts and the possibility of missing, multiple possible missing parts in the building digital model are determined; optionally, the possible missing parts are building connection parts, load-bearing parts in the building, non-visible surface parts of the building, or low-visibility surface parts of the building;

[0126] For each possible missing part, the number of occurrences of the possible missing part in the image missing parts corresponding to all framed images is calculated;

[0127] If the number of occurrences is greater than a preset threshold, the possible missing part is determined as a missing model part.

[0128] It can be seen that through the above-mentioned optional embodiments, after determining the possible missing parts, the missing model parts can be determined based on the counting method, so as to effectively improve the efficiency of determining the missing model parts, assist in improving the modeling efficiency and modeling accuracy of the digital model of the historical building, and realize more intelligent risk monitoring of historical buildings, reduce monitoring costs, and improve monitoring effects.

[0129] As an optional embodiment, in the above steps, according to the historical digital model record corresponding to the target historical building and the preset structure prediction algorithm, the model shape of each missing model part is predicted to obtain the predicted shape result, including:

[0130] For each missing model part, determine a plurality of nearby model parts corresponding to the missing model part; optionally, the nearby model part is a model part whose distance to the missing model part is less than a preset distance threshold;

[0131] Inputting all nearby model parts and corresponding model positions into a trained shape prediction neural network to obtain a predicted model shape corresponding to the missing model part; optionally, the shape prediction neural network is trained by a training data set including a plurality of training nearby model parts and corresponding specific area shape annotations; the training data set is obtained by annotating historical digital model records corresponding to the target historical building;

[0132] The predicted model shapes corresponding to all missing model parts are determined as the predicted shape results.

[0133] It can be seen that through the above-mentioned optional embodiments, the shape prediction neural network trained based on the historical digital model records corresponding to the target historical building can be used to accurately predict the predicted model shape corresponding to the missing model part according to the nearby model parts and the corresponding model positions, thereby assisting in improving the modeling efficiency and modeling accuracy of the digital model of the historical building, and realizing more intelligent risk monitoring of historical buildings, reducing monitoring costs and improving monitoring effects.

[0134] As an optional embodiment, in the above steps, based on the neural network algorithm, according to the completed digital model and the real-time sensor information, predicting the risk location and risk type of the target historical building includes:

[0135] The real-time sensor information corresponding to each missing model part in the completed digital model is input into the risk prediction neural network model corresponding to the corresponding data type to obtain the risk prediction probability and risk prediction type corresponding to each missing model part;

[0136] The real-time sensor information corresponding to each updated position in the completed digital model is input into the risk prediction neural network model corresponding to the corresponding data type to obtain the risk prediction probability and risk prediction type corresponding to each updated position;

[0137] All risk prediction types with risk prediction probabilities greater than a preset probability threshold are determined as risk types of the target historical building, and the corresponding missing model parts or updated locations are determined as corresponding risk locations.

[0138] It can be seen that through the above-mentioned optional embodiments, risk identification and monitoring can be prioritized based on the missing parts and updated parts of the model determined in the previous steps. These parts are often missing or have large update changes due to new building damage accidents. Therefore, this priority identification and monitoring can achieve more intelligent risk monitoring of historical buildings, reduce monitoring costs, and improve monitoring effects.

[0139] Embodiment 2

[0140] See also Figure 2 , Figure 2 : is a schematic diagram of a historical building digital data processing system based on AI disclosed in an embodiment of the present invention. Figure 2 The AI-based historical building digital data processing system described can be applied to a data processing device / data processing server (wherein the server includes a local processing server or a cloud processing server). Figure 2 As shown, the AI-based historical building digital data processing system may include:

[0141] The acquisition module 201 is used to acquire real-time sensor information of the target historical building through sensors.

[0142] The generation module 202 is used to generate a digital building model corresponding to the target historical building according to the real-time sensor information and based on a modeling algorithm.

[0143] The completion module 203 is used to complete the missing parts of the building digital model based on a neural network algorithm to obtain a completed digital model.

[0144] The prediction module 204 is used to predict the risk location and risk type of the target historical building based on the neural network algorithm, the completed digital model and the real-time sensor information.

[0145] It can be seen that the above-mentioned embodiments of the invention can generate a digital building model corresponding to the target historical building based on the modeling algorithm according to the real-time sensor information, and then complete the missing parts of the digital building model based on the neural network algorithm, and thereby accurately predict the risk location and risk type of the target historical building, thereby improving the modeling efficiency and modeling accuracy of the digital model of the historical building, and realizing more intelligent risk monitoring of historical buildings, reducing monitoring costs and improving monitoring effects.

[0146] As an optional embodiment, the real-time sensing information includes image data, sound wave information, ultrasonic wave reflection information and light reflection information.

[0147] It can be seen that through the above optional embodiments, the content of real-time sensing information is limited, which can fully characterize the structural characteristics of historical buildings and be used for more accurate modeling later, thereby assisting in improving the modeling efficiency and modeling accuracy of digital models of historical buildings, and realizing more intelligent risk monitoring of historical buildings, reducing monitoring costs and improving monitoring effects.

[0148] As an optional embodiment, the specific manner in which the generation module generates the digital building model corresponding to the target historical building according to the real-time sensor information and based on the modeling algorithm includes:

[0149] Obtaining a preset historical digital model corresponding to the target historical building;

[0150] According to the real-time sensor information and based on the 3D modeling algorithm, a real-time 3D model corresponding to the target historical building is generated;

[0151] The historical digital model is updated according to the real-time three-dimensional model to obtain the building digital model corresponding to the target historical building.

[0152] It can be seen that through the above-mentioned optional embodiments, a real-time three-dimensional model corresponding to the target historical building can be generated based on the modeling algorithm according to real-time sensing information to update the historical digital model to obtain a building digital model corresponding to the target historical building, which can be used for missing completion and risk analysis subsequently, to assist in improving the modeling efficiency and modeling accuracy of the digital model of the historical building, and to achieve more intelligent risk monitoring of historical buildings, reduce monitoring costs, and improve monitoring effects.

[0153] As an optional embodiment, the generation module updates the historical digital model according to the real-time three-dimensional model to obtain the specific manner of the building digital model corresponding to the target historical building, including:

[0154] Align the real-time 3D model and the historical digital model into the same coordinate system;

[0155] For any spatial position in the coordinate system including both the real-time three-dimensional model and the historical digital model, determining a first model portion of the real-time three-dimensional model and a second model portion of the historical digital model in the spatial position;

[0156] Calculating an image difference and a three-dimensional difference between the first model part and the second model part;

[0157] Calculate the weighted average of the image difference and the three-dimensional difference to obtain the position update parameter corresponding to the spatial position;

[0158] Determine the spatial positions where all position update parameters are greater than a first parameter threshold as update positions;

[0159] Based on the historical digital model, the model part corresponding to the historical digital model corresponding to each updated position is replaced with the model part of the corresponding position in the real-time three-dimensional model to obtain the building digital model corresponding to the target historical building.

[0160] It can be seen that through the above-mentioned optional embodiments, the necessity of updating the position can be determined by calculating the image difference and three-dimensional difference of the model part in the spatial position that includes both the real-time three-dimensional model and the historical digital model, and the part with high necessity is updated and replaced to obtain the digital model of the building, which is subsequently used for missing completion and risk analysis, to assist in improving the modeling efficiency and modeling accuracy of the digital model of the historical building, and to achieve more intelligent risk monitoring of historical buildings, reduce monitoring costs, and improve monitoring effects.

[0161] As an optional embodiment, the completion module completes the missing parts of the building digital model based on the neural network algorithm, and the specific method of obtaining the completed digital model includes:

[0162] Framing the building digital model based on multiple viewpoints and multiple angles to obtain multiple framing images;

[0163] Input each framed image into a preset missing part recognition neural network model to obtain the missing part of the image corresponding to each framed image;

[0164] According to the missing parts of the images corresponding to all the framed images, the missing model parts in the building digital model are identified;

[0165] According to the historical digital model records corresponding to the target historical building and the preset structural prediction algorithm, the model shape of each missing model part is predicted to obtain the predicted shape result;

[0166] Each missing model part of the building digital model is completed according to the predicted shape result to obtain a completed digital model.

[0167] It can be seen that through the above-mentioned optional embodiments, it is possible to identify the parts with structural missing in the image based on the framed images obtained based on multi-angle and multi-viewpoint framing through a neural network model, and predict the shape of the missing parts based on historical model records to complete them, thereby assisting in improving the modeling efficiency and modeling accuracy of the digital model of historical buildings, and realizing more intelligent risk monitoring of historical buildings, reducing monitoring costs and improving monitoring effects.

[0168] As an optional embodiment, the specific manner in which the completion module identifies the missing model portion in the building digital model according to the missing image portions corresponding to all the framing images includes:

[0169] Based on the correspondence between the preset model parts and the possibility of missing, multiple possible missing parts in the building digital model are determined; optionally, the possible missing parts are building connection parts, load-bearing parts in the building, non-visible surface parts of the building, or low-visibility surface parts of the building;

[0170] For each possible missing part, the number of occurrences of the possible missing part in the image missing parts corresponding to all framed images is calculated;

[0171] If the number of occurrences is greater than a preset threshold, the possible missing part is determined as a missing model part.

[0172] It can be seen that through the above-mentioned optional embodiments, after determining the possible missing parts, the missing model parts can be determined based on the counting method, so as to effectively improve the efficiency of determining the missing model parts, assist in improving the modeling efficiency and modeling accuracy of the digital model of the historical building, and realize more intelligent risk monitoring of historical buildings, reduce monitoring costs, and improve monitoring effects.

[0173] As an optional embodiment, the completion module predicts the model shape of each missing model part according to the historical digital model record corresponding to the target historical building and a preset structural prediction algorithm, and obtains the specific method of the predicted shape result, including:

[0174] For each missing model part, determine a plurality of nearby model parts corresponding to the missing model part; optionally, the nearby model part is a model part whose distance to the missing model part is less than a preset distance threshold;

[0175] Inputting all nearby model parts and corresponding model positions into a trained shape prediction neural network to obtain a predicted model shape corresponding to the missing model part; optionally, the shape prediction neural network is trained by a training data set including a plurality of training nearby model parts and corresponding specific area shape annotations; the training data set is obtained by annotating historical digital model records corresponding to the target historical building;

[0176] The predicted model shapes corresponding to all missing model parts are determined as the predicted shape results.

[0177] It can be seen that through the above-mentioned optional embodiments, the shape prediction neural network trained based on the historical digital model records corresponding to the target historical building can be used to accurately predict the predicted model shape corresponding to the missing model part according to the nearby model parts and the corresponding model positions, thereby assisting in improving the modeling efficiency and modeling accuracy of the digital model of the historical building, and realizing more intelligent risk monitoring of historical buildings, reducing monitoring costs and improving monitoring effects.

[0178] As an optional embodiment, the prediction module predicts the risk location and risk type of the target historical building based on the neural network algorithm according to the completed digital model and real-time sensor information, including:

[0179] The real-time sensor information corresponding to each missing model part in the completed digital model is input into the risk prediction neural network model corresponding to the corresponding data type to obtain the risk prediction probability and risk prediction type corresponding to each missing model part;

[0180] The real-time sensor information corresponding to each updated position in the completed digital model is input into the risk prediction neural network model corresponding to the corresponding data type to obtain the risk prediction probability and risk prediction type corresponding to each updated position;

[0181] All risk prediction types with risk prediction probabilities greater than a preset probability threshold are determined as risk types of the target historical building, and the corresponding missing model parts or updated locations are determined as corresponding risk locations.

[0182] It can be seen that through the above-mentioned optional embodiments, risk identification and monitoring can be prioritized based on the missing parts and updated parts of the model determined in the previous steps. These parts are often missing or have large update changes due to new building damage accidents. Therefore, this priority identification and monitoring can achieve more intelligent risk monitoring of historical buildings, reduce monitoring costs, and improve monitoring effects.

[0183] Embodiment 3

[0184] See also Figure 3 , Figure 3 This is another AI-based historical building digital data processing system disclosed in an embodiment of the present invention. Figure 3 The AI-based historical building digital data processing system described is applied to a data processing device / data processing server (wherein the server includes a local processing server or a cloud processing server). Figure 3 As shown, the AI-based historical building digital data processing system may include:

[0185] A memory 301 storing executable program codes;

[0186] a processor 302 coupled to the memory 301;

[0187] The processor 302 calls the executable program code stored in the memory 301 to execute the steps of the AI-based historical building digital data processing method described in the first embodiment.

[0188] Embodiment 4

[0189] An embodiment of the present invention discloses a computer-readable storage medium storing a computer program for electronic data exchange, wherein the computer program enables a computer to execute the steps of the AI-based historical building digital data processing method described in the first embodiment.

[0190] Embodiment 5

[0191] An embodiment of the present invention discloses a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to enable a computer to execute the steps of the AI-based historical building digital data processing method described in Example 1.

[0192] The above describes specific embodiments of the present specification, and other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily have to be performed in the specific order or sequential order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0193] The systems, devices, modules or units described in the above embodiments may be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, the computer may be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0194] For the convenience of description, the above device is described in various units according to their functions. Of course, when implementing this specification, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0195] Those skilled in the art will appreciate that the embodiments of this specification may be provided as methods, systems, or computer program products. Therefore, the embodiments of this specification may be in the form of complete hardware embodiments, complete software embodiments, or embodiments in combination with software and hardware. Moreover, the embodiments of this specification may be in the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0196] This specification is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of this specification. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0197] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0198] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0199] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0200] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0201] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0202] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0203] This specification may be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. 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 communication network. In a distributed computing environment, program modules may be located in local and remote computer storage media, including storage devices.

[0204] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0205] Finally, it should be noted that the AI-based historical building digital data processing method and system disclosed in the embodiment of the present invention only discloses the preferred embodiment of the present invention, which is only used to illustrate the technical solution of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments can still be modified, or some of the technical features therein can be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for processing digital data of historical buildings based on AI, characterized in that: The method comprises: Acquire real-time sensing information of target historical buildings through sensors; According to the real-time sensing information, based on the modeling algorithm, a digital building model corresponding to the target historical building is generated, including: Acquire a preset historical digital model corresponding to the target historical building; According to the real-time sensing information, based on a three-dimensional modeling algorithm, a real-time three-dimensional model corresponding to the target historical building is generated; aligning the real-time three-dimensional model and the historical digital model into the same coordinate system; For any spatial position in the coordinate system including both the real-time three-dimensional model and the historical digital model, determining a first model portion of the real-time three-dimensional model and a second model portion of the historical digital model in the spatial position; Calculating an image difference and a three-dimensional difference between the first model part and the second model part; Calculating a weighted average of the image difference and the three-dimensional difference to obtain a position update parameter corresponding to the spatial position; Determine the spatial positions whose position update parameters are all greater than a first parameter threshold as update positions; Based on the historical digital model, the model portion corresponding to each of the updated positions of the historical digital model is replaced with the model portion of the corresponding position in the real-time three-dimensional model to obtain the building digital model corresponding to the target historical building; Based on a neural network algorithm, the missing parts of the building digital model are completed to obtain a completed digital model; Based on a neural network algorithm, the risk location and risk type of the target historical building are predicted according to the completed digital model and the real-time sensor information.

2. The AI-based historical building digital data processing method according to claim 1 is characterized in that: The real-time sensing information includes image data, sound wave information, ultrasonic wave reflection information and light reflection information.

3. The AI-based historical building digital data processing method according to claim 1 is characterized in that: The method of completing missing parts of the building digital model based on a neural network algorithm to obtain a completed digital model includes: Framing the building digital model based on multiple viewpoints and multiple angles to obtain multiple framing images; Inputting each of the framing images into a preset defect recognition neural network model to obtain the image defect portion corresponding to each of the framing images; According to the missing image parts corresponding to all the framing images, identifying the missing model parts in the digital building model; According to the historical digital model records corresponding to the target historical building and a preset structural prediction algorithm, the model shape of each missing model part is predicted to obtain a predicted shape result; Each missing model part of the building digital model is completed according to the predicted shape result to obtain a completed digital model.

4. The AI-based historical building digital data processing method according to claim 3 is characterized in that: The step of identifying the missing model part in the digital building model according to the missing image parts corresponding to all the framing images includes: Based on the correspondence between the preset model parts and the possibility of missing, a plurality of possible missing parts in the building digital model are determined; the possible missing parts are building connection parts, load-bearing parts in the building, non-visible surface parts of the building, or low-visibility surface parts of the building; For each of the possible missing parts, calculating the number of occurrences of the possible missing part in the image missing parts corresponding to all the framing images; If the number of occurrences is greater than a preset number threshold, the possible missing part is determined as a missing model part.

5. The AI-based historical building digital data processing method according to claim 3 is characterized in that: The step of predicting the model shape of each missing model part according to the historical digital model record corresponding to the target historical building and a preset structural prediction algorithm to obtain a predicted shape result includes: For each missing model part, determining a plurality of nearby model parts corresponding to the missing model part; the nearby model parts are model parts whose distances to the missing model part are less than a preset distance threshold; Inputting all the nearby model parts and the corresponding model positions into a trained shape prediction neural network to obtain a predicted model shape corresponding to the missing model part; the shape prediction neural network is trained by a training data set including a plurality of training nearby model parts and corresponding specific area shape annotations; the training data set is obtained by annotating the historical digital model records corresponding to the target historical building; The predicted model shapes corresponding to all the missing model parts are determined as predicted shape results.

6. The AI-based historical building digital data processing method according to claim 5 is characterized in that: The method of predicting the risk location and risk type of the target historical building based on the neural network algorithm according to the completed digital model and the real-time sensor information includes: Input the real-time sensing information corresponding to each missing model part in the completed digital model into the risk prediction neural network model corresponding to the corresponding data type to obtain the risk prediction probability and risk prediction type corresponding to each missing model part; Input the real-time sensing information corresponding to each update position in the completed digital model into the risk prediction neural network model corresponding to the corresponding data type to obtain the risk prediction probability and risk prediction type corresponding to each update position; All the risk prediction types whose risk prediction probabilities are greater than a preset probability threshold are determined as risk types of the target historical building, and the corresponding missing model parts or the updated positions are determined as corresponding risk positions.

7. An AI-based historical building digital data processing system, characterized in that: The system comprises: An acquisition module, used for acquiring real-time sensing information of a target historical building through sensors; A generation module is used to generate a digital building model corresponding to the target historical building according to the real-time sensing information and based on a modeling algorithm, including: Acquire a preset historical digital model corresponding to the target historical building; According to the real-time sensing information, based on a three-dimensional modeling algorithm, a real-time three-dimensional model corresponding to the target historical building is generated; aligning the real-time three-dimensional model and the historical digital model into the same coordinate system; For any spatial position in the coordinate system including both the real-time three-dimensional model and the historical digital model, determining a first model portion of the real-time three-dimensional model and a second model portion of the historical digital model in the spatial position; Calculating an image difference and a three-dimensional difference between the first model part and the second model part; Calculating a weighted average of the image difference and the three-dimensional difference to obtain a position update parameter corresponding to the spatial position; Determine the spatial positions whose position update parameters are all greater than a first parameter threshold as update positions; Based on the historical digital model, the model portion corresponding to each of the updated positions of the historical digital model is replaced with the model portion of the corresponding position in the real-time three-dimensional model to obtain the building digital model corresponding to the target historical building; A completion module, used to complete the missing parts of the building digital model based on a neural network algorithm to obtain a completed digital model; A prediction module is used to predict the risk location and risk type of the target historical building based on a neural network algorithm according to the completed digital model and the real-time sensor information.

8. An AI-based historical building digital data processing system, characterized in that: The system comprises: A memory storing executable program code; a processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the AI-based historical building digital data processing method as described in any one of claims 1-6.

Images