Building health assessment method and system based on two / three-dimensional fusion technology

Abstract

The application discloses a building health assessment method and system based on two-dimensional / three-dimensional fusion technology, and the method comprises the following steps: performing omnibearing scanning on the indoor space of a building and collecting image data; extracting key building structures in the image data to obtain a plurality of building structure images; identifying damages in the building structure images to obtain damage identification results; taking the maximum value of the area of all damage regions of the same damage as the final damage region area of the damage, and calculating two-dimensional damage information evaluation results of the building; obtaining three-dimensional damage information evaluation results of the building based on three-dimensional reconstruction; and obtaining the health condition of the whole building based on the two-dimensional damage information evaluation results and the three-dimensional damage information evaluation results. The application is applied to the field of building detection, and through the fusion of two-dimensional damage evaluation and three-dimensional reconstruction results, comprehensive damage evaluation of the building on the three-dimensional level can be effectively realized.

Description

Technical Field

[0001] The present invention relates to the field of building detection technology, and in particular to a building health assessment method and system based on two-dimensional / three-dimensional fusion technology. Background Art

[0002] With the accelerating pace of urbanization, the monitoring and management of building structural health has become a critical issue for urban safety and sustainable development. Due to natural disasters, engineering accidents, or the long-term service life of buildings, buildings may suffer varying degrees of structural damage, which not only directly threatens their structural integrity but also poses a significant safety hazard to occupants. Therefore, the development of an efficient and accurate building health assessment system is of vital practical significance and has broad application value.

[0003] Currently, traditional methods for building health assessment rely primarily on on-site inspections and remote sensing image analysis. On-site inspections typically require professionals carrying handheld equipment to inspect and assess structural damage at the construction site, with results relying on accumulated engineering experience. This approach is not only labor-intensive and time-consuming, but also potentially dangerous and highly uncertain. Remote sensing image analysis, on the other hand, extracts image features such as color, spectrum, shape, texture, and shadows to develop algorithms for distinguishing built-up areas from non-built-up areas. However, these methods, based on prior knowledge, are susceptible to the limitations of building area characteristics and, due to resolution limitations, struggle to effectively identify even minor damage.

[0004] Furthermore, existing building health assessment systems typically focus on detecting damage to individual building components (such as walls, beams, and columns), neglecting to comprehensively assess the building's overall three-dimensional structure. This limitation makes it difficult to accurately assess the true health status of a building as a whole, especially when assessing the health of complex spatial structures such as floors and rooms. Summary of the Invention

[0005] In response to the above-mentioned deficiencies in the existing technology, the present invention provides a building health assessment method and system based on two-dimensional / three-dimensional fusion technology. By fusing two-dimensional damage assessment with three-dimensional reconstruction results, it can effectively achieve a comprehensive damage assessment of the building at the three-dimensional level.

[0006] To achieve the above objectives, the present invention provides a building health assessment method based on two-dimensional / three-dimensional fusion technology, comprising the following steps:

[0007] Step 1: Perform a full-scale scan of the indoor space of the building and collect image data during the scanning process;

[0008] Step 2: extracting key building structures from each of the image data to obtain a plurality of building structure images, wherein the key building structures include beams, columns, floors, ceilings, and walls;

[0009] Step 3: Identify damage in the building structure image based on a neural network to obtain damage identification results in all the building structure images, wherein the damage identification results include the area of ​​the damaged area and the confidence level;

[0010] Step 4: After eliminating the damage identification results with a confidence level lower than the set threshold, the maximum damage area among all damage identification results for the same damage is taken as the final damage area of ​​the damage;

[0011] Step 5: Calculate the corresponding damage assessment results based on the final damage area of ​​each damage, and combine the damage assessment results corresponding to all damages to obtain a two-dimensional damage information assessment result of the building;

[0012] Step 6: Performing three-dimensional reconstruction of the interior space of the building based on the image data, and obtaining a three-dimensional damage information assessment result of the building based on the three-dimensional reconstruction;

[0013] Step 7: Based on the two-dimensional damage information assessment result and the three-dimensional damage information assessment result, the health status of the entire building is obtained, and normalization processing is performed to obtain the final assessment result of the building.

[0014] In one embodiment, in step 5, the calculation process of the damage assessment result is:

[0015]

[0016] in, The damage assessment results are: is the final damaged area, that is, the number of damaged pixels, is the total number of pixels in the image of the building structure where the damage is located.

[0017] In one embodiment, in step 5, the damage assessment results corresponding to all damages are integrated to obtain a two-dimensional damage information assessment result of the building, specifically:

[0018] All damages are classified into one of the following categories: crack damage, hole damage, and spalling damage according to the actual situation;

[0019] Calculate the average value of the damage assessment results of all crack damages as the crack damage assessment result;

[0020] Calculate the average value of the damage assessment results of all hole damages as the hole damage assessment result;

[0021] Calculate the average value of the damage assessment results of all spalling damages as the spalling damage assessment result;

[0022] The crack damage assessment result, the hole damage assessment result, and the spalling damage assessment result are weighted to obtain the two-dimensional damage information assessment result.

[0023] In one embodiment, step 6 is specifically as follows:

[0024] Performing three-dimensional reconstruction of the interior space of the building based on the image data to obtain three-dimensional point cloud data of each key building structure in each building after damage;

[0025] Compare and calculate the 3D point cloud data of each key building structure after damage with the corresponding 3D point cloud data before damage to obtain the volume change and deflection change of each key building structure;

[0026] Classify all key building structures as beams, columns, floors, ceilings and walls;

[0027] Calculate the average volume change and average deflection change of all beams, calculate the average volume change and average deflection change of all columns, calculate the average volume change and average deflection change of all floors, and calculate the average volume change and average deflection change of all walls;

[0028] The volume damage assessment result is obtained by weighting the average value of the volume changes of beams, columns, floors, ceilings and walls;

[0029] The deflection damage assessment result is obtained by weighting the average value of the deflection changes of beams, columns, floors, ceilings and walls;

[0030] The volume damage assessment result and the deflection damage assessment result are the three-dimensional damage information assessment result.

[0031] In one embodiment, in step 7, based on the two-dimensional damage information assessment result and the three-dimensional damage information assessment result, the health status of the entire building is obtained as follows:

[0032]

[0033] in, For the health of the building, is the two-dimensional damage information assessment result, is the volume damage assessment result, is the deflection damage assessment result, 、 、 is the weight coefficient, Assess the initial condition of the building.

[0034] In one embodiment, in step 7, the normalization process is performed to obtain the final evaluation result of the building as follows:

[0035]

[0036] in, For the final evaluation results, is the normalized maximum value of the evaluation structure;

[0037] when When the building is assessed as damaged at level 1,

[0038] when When the building is assessed as damaged at level 2,

[0039] when At that time, the building's health assessment was level 3 damage;

[0040] when At that time, the building's health assessment was level 4 damage.

[0041] To achieve the above objectives, the present invention further provides a building health assessment system based on 2D / 3D fusion technology, which uses the above method to assess the health status of a building. The building health system includes:

[0042] A data acquisition module is used to collect image data of the indoor space of a building;

[0043] A two-dimensional image processing module is used to evaluate and obtain two-dimensional damage information assessment results of the building;

[0044] A three-dimensional reconstruction processing module is used to evaluate the three-dimensional damage information of the building;

[0045] The data fusion processing module is used to output the final assessment results and damage level of the building.

[0046] Compared with the prior art, the present invention has the following beneficial technical effects:

[0047] 1. This invention detects and analyzes damage to building components on a two-dimensional scale. By comprehensively assessing the damage level of each building component, it generates a weighted file for its impact on the overall building health. Simultaneously, it uses three-dimensional reconstruction technology to analyze information such as geometric volume changes, displacement, and deformation of building components. Finally, by integrating the two-dimensional damage assessment with the three-dimensional reconstruction results, it comprehensively derives health assessment weights for different building components, achieving a comprehensive three-dimensional damage assessment of the building, providing a new solution for building health assessment.

[0048] 2. The present invention has the characteristics of high accuracy, strong adaptability and wide application range, and has broad market application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. 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 the structures shown in these drawings without paying any creative work.

[0050] Figure 1 Flowchart of a building health assessment method based on two-dimensional / three-dimensional fusion technology in an embodiment of the present invention;

[0051] Figure 2 3D / 2D fusion technology is used to evaluate the health of a building.

[0052] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0053] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0054] It should be noted that all directional indications (such as up, down, left, right, front, back, etc.) in the embodiments of the present invention are only used to explain the relative position relationship, movement status, etc. between the various components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly.

[0055] In addition, the technical solutions between the various embodiments of the present invention can be combined with each other, but it must be based on the fact that ordinary technicians in this field can implement it. When the combination of technical solutions is mutually contradictory or cannot be implemented, it should be deemed that such a combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0056] Example 1

[0057] like Figure 1 The present embodiment discloses a building health assessment method based on 2D / 3D fusion technology (hereinafter referred to as the "building health assessment method"), which mainly includes the following steps:

[0058] Step 1: Perform a full-scale scan of the indoor space of the building and collect image data during the scanning process;

[0059] Step 2: extract key building structures from each image data to obtain a number of building structure images, wherein the key building structures include beams, columns, floors, ceilings, and walls;

[0060] Step 3: Identify damage in the building structure image based on the neural network to obtain damage identification results in all building structure images. The damage identification results include the damage area and confidence level.

[0061] Step 4: After eliminating the damage identification results with a confidence level lower than the set threshold, the maximum damage area among all damage identification results for the same damage is taken as the final damage area of ​​the damage;

[0062] Step 5: Calculate the corresponding damage assessment results based on the final damage area of ​​each damage, and combine the damage assessment results corresponding to all damages to obtain a two-dimensional damage information assessment result of the building;

[0063] Step 6: Perform three-dimensional reconstruction of the interior space of the building based on the image data, and obtain a three-dimensional damage information assessment result of the building based on the three-dimensional reconstruction;

[0064] Step 7: Based on the two-dimensional damage information assessment results and the three-dimensional damage information assessment results, the health status of the entire building is obtained, and normalization processing is performed to obtain the final assessment result of the building.

[0065] The building health assessment method in this embodiment primarily includes four steps: image acquisition (i.e., step 1 above), 2D image processing (i.e., steps 2-5 above), 3D point cloud processing (i.e., step 6 above), and final assessment (i.e., step 7 above). Damage detection and analysis of building components are performed on a 2D level. The degree of damage to each building component is comprehensively assessed to generate a weighted file for its impact on the overall building health. 3D reconstruction technology is then used to analyze information such as geometric volume changes, displacements, and deformations of the building components. Finally, the 2D damage assessment is integrated with the 3D reconstruction results to comprehensively derive health assessment weights for different building components, achieving a comprehensive 3D damage assessment of the building, providing a new solution for building health assessment.

[0066] In the image acquisition part of step 1, a multi-rotor drone is used as the data collection platform. The multi-rotor drone is equipped with a high-precision camera to realize remote control of the drone for obstacle avoidance and data collection, thereby realizing all-round scanning and image collection of the indoor space of the building, solving the problem of the danger of manual data collection. At the same time, the flexibility of the drone greatly improves the efficiency of data collection.

[0067] The two-dimensional image processing part in steps 2 to 5 is mainly divided into two parts: building component detection and segmentation and damage category detection.

[0068] The building component detection component primarily identifies key building structures and damaged areas from image data. For damaged buildings, the internal structure and external macroscopic damage characteristics exhibit significant differences, which can lead to inaccurate damage assessments based on remote sensing images. For example, a building's exterior wall may show only minor damage, while its internal structure may have sustained significant damage. This discrepancy prevents traditional macroscopic assessment methods from accurately reflecting the building's actual damage status.

[0069] In order to solve this problem, this embodiment proposes a method that uses building structure components as detection elements, focusing on identifying the internal structure of the building and the type of damage. First, the key building structures inside the building (including beams, columns, floors, ceilings and walls) are selected as research objects, and the three common damage categories of cracks, holes and peeling are selected as detection targets. That is, the image data is first segmented to obtain a number of building structure images, and then the building structure images are feature extracted based on the neural network to realize the identification of damage in the building structure images. Among them, in order to ensure the accuracy of feature extraction, the confidence of the feature extraction result is set, and the detection results with a confidence less than the set threshold are screened out, thereby reducing false detection. As for the specific structure of the neural network and its training process, they are relatively conventional technical means, and this embodiment will not elaborate on them.

[0070] Damage classification detection primarily aims to prevent misidentification of damage categories due to camera errors and viewing angle errors. This is because the same damage type may appear differently from different viewing angles, posing a challenge to damage identification. For example, a completely collapsed wall may appear as just a small hole from a side view. This inconsistent appearance caused by these perspective differences can lead to errors in damage area assessment, compromising the accuracy of the overall damage assessment.

[0071] To address this issue, this embodiment compares the detection areas of the same damage in different building structure images and extracts the maximum damage area as the final damage area, thereby effectively avoiding the situation where damage classification errors caused by camera errors and shooting angle errors occur. That is, in the specific implementation process of step 5, the calculation process of the damage assessment result for a certain damage is as follows:

[0072]

[0073] in, The damage assessment results are: is the final damaged area, that is, the number of damaged pixels, is the total number of pixels in the image of the building structure where the damage is located.

[0074] The above formula can be used to calculate the damage assessment results for all damage within the building's interior space. However, in actual building assessments, different damage types have different impacts on the assessment results. Therefore, the damage assessment results for all damage in this embodiment cannot be generalized. Based on this, this embodiment proposes a method for comprehensively analyzing different types of damage to obtain the overall two-dimensional damage information assessment results for the building. Specifically, the method is as follows:

[0075] First, the damages identified from all building structures are classified into one of the following categories: crack damage, hole damage, and spalling damage according to their shapes and other actual conditions.

[0076] Then, the average damage assessment results of all crack damages are calculated as the crack damage assessment result; the average damage assessment results of all hole damages are calculated as the hole damage assessment result; the average damage assessment results of all spalling damages are calculated as the spalling damage assessment result;

[0077] Finally, the crack damage assessment results, hole damage assessment results, and spalling damage assessment results are weighted to obtain the two-dimensional damage information assessment results.

[0078] In the actual building assessment process, it is difficult to observe three-dimensional physical quantity changes, such as changes in volume and deflection, from two-dimensional images. However, the detection of physical quantities is a very convincing factor in building damage assessment. Therefore, in the three-dimensional point cloud processing portion of step 6 of this embodiment, by comparing the volume and deflection changes of key building structures before and after damage, the three-dimensional damage information assessment results of the building are obtained. The specific implementation process is as follows:

[0079] Perform 3D reconstruction of the building's interior space based on image data, obtaining 3D point cloud data of each key building structure after damage;

[0080] Compare and calculate the 3D point cloud data of each key building structure after damage with the corresponding 3D point cloud data before damage to obtain the volume change and deflection change of each key building structure;

[0081] Classify all key building structures as beams, columns, floors, ceilings and walls;

[0082] Calculate the average volume change and average deflection change of all beams, calculate the average volume change and average deflection change of all columns, calculate the average volume change and average deflection change of all floors, and calculate the average volume change and average deflection change of all walls;

[0083] The volume damage assessment result is obtained by weighting the average value of the volume changes of beams, columns, floors, ceilings and walls;

[0084] The deflection damage assessment result is obtained by weighting the average value of the deflection changes of beams, columns, floors, ceilings and walls;

[0085] The volume damage assessment results and deflection damage assessment results are the three-dimensional damage information assessment results.

[0086] After obtaining the 3D damage information assessment results, the health status of the entire building can be obtained by combining the 2D and 3D damage information assessment results. Specifically:

[0087]

[0088] in, For the health of the building, is the two-dimensional damage information assessment result, is the volume damage assessment result, Deflection damage assessment results , 、 、 is the weight coefficient, To conduct an initial condition assessment of the building;

[0089] For the calculated After normalization, the final evaluation result of the building can be obtained:

[0090]

[0091] in, For the final evaluation results, is the normalized maximum value of the evaluation structure;

[0092] when When the building is assessed as damaged at level 1,

[0093] when When the building is assessed as damaged at level 2,

[0094] when At that time, the building's health assessment was level 3 damage;

[0095] when At that time, the building's health assessment was level 4 damage;

[0096] Level 1 and 2 damage are considered mild damage, meaning the building is only slightly damaged and does not affect its basic functions and structural integrity. Level 3 damage is considered moderate damage, meaning the building is deemed to have reached a point where repairs are necessary, otherwise it may be dangerous to live in and requires timely repairs. Level 4 damage is considered severe damage, meaning the building has completely lost its original architectural function.

[0097] It is worth noting that although this embodiment Figure 1 The steps in the diagram are shown in the order indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. In addition, Figure 1 At least part of the steps may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least part of the sub-steps or stages of other steps.

[0098] Example 2

[0099] Based on the building health assessment method based on two-dimensional / three-dimensional fusion technology in Example 1, this embodiment discloses a building health assessment system based on two-dimensional / three-dimensional fusion technology. Figure 2 The building health assessment system includes a data acquisition module, a two-dimensional image processing module, a three-dimensional reconstruction processing module and a data fusion processing module. Specifically:

[0100] The data acquisition module is used to collect image data of the indoor space of the building;

[0101] The two-dimensional image processing module is used to evaluate and obtain the two-dimensional damage information assessment results of the building;

[0102] The three-dimensional reconstruction processing module is used to evaluate and obtain the three-dimensional damage information assessment results of the building;

[0103] The data fusion processing module is used to output the final assessment results and damage level of the building.

[0104] In this embodiment, the specific working processes and working principles of the data acquisition module, the two-dimensional image processing module, the three-dimensional reconstruction processing module, and the data fusion processing module are the same as those in the method of Example 1, and therefore are not further described in this embodiment. Each unit module can be implemented in whole or in part through software, hardware, or a combination thereof. Each unit module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in the form of software in the computer device so that the processor can call and execute the corresponding operations of each of the above unit modules.

[0105] The above description is only a preferred embodiment of the present invention and does not limit the patent scope of the present invention. All equivalent structural transformations made by using the contents of the present invention description and drawings under the inventive concept of the present invention, or direct / indirect application in other related technical fields are included in the patent protection scope of the present invention.

Claims

1. A building health assessment method based on 2D / 3D fusion technology, characterized in that: The steps include: Step 1: Perform a full-scale scan of the indoor space of the building and collect image data during the scanning process; Step 2: extracting key building structures from each of the image data to obtain a plurality of building structure images, wherein the key building structures include beams, columns, floors, ceilings, and walls; Step 3: Identify damage in the building structure image based on a neural network to obtain damage identification results in all the building structure images, wherein the damage identification results include the area of ​​the damaged area and the confidence level; Step 4: After eliminating the damage identification results with a confidence level lower than the set threshold, the maximum damage area among all damage identification results for the same damage is taken as the final damage area of ​​the damage; Step 5: Calculate the corresponding damage assessment results based on the final damage area of ​​each damage, and combine the damage assessment results corresponding to all damages to obtain a two-dimensional damage information assessment result of the building; Step 6: Performing three-dimensional reconstruction of the interior space of the building based on the image data, and obtaining a three-dimensional damage information assessment result of the building based on the three-dimensional reconstruction; Step 7: Based on the two-dimensional damage information assessment result and the three-dimensional damage information assessment result, the health status of the entire building is obtained, and normalization processing is performed to obtain the final assessment result of the building.

2. The building health assessment method based on 2D / 3D fusion technology according to claim 1 is characterized in that: In step 5, the calculation process of the damage assessment result is: in, The damage assessment results are: is the final damaged area, that is, the number of damaged pixels, is the total number of pixels in the image of the building structure where the damage is located.

3. The building health assessment method based on 2D / 3D fusion technology according to claim 1 is characterized in that: In step 5, the damage assessment results corresponding to all damages are integrated to obtain the two-dimensional damage information assessment result of the building, specifically: All damages are classified into one of the following categories: crack damage, hole damage, and spalling damage according to the actual situation; Calculate the average value of the damage assessment results of all crack damages as the crack damage assessment result; Calculate the average value of the damage assessment results of all hole damages as the hole damage assessment result; Calculate the average value of the damage assessment results of all spalling damages as the spalling damage assessment result; The crack damage assessment result, the hole damage assessment result, and the spalling damage assessment result are weighted to obtain the two-dimensional damage information assessment result.

4. The building health assessment method based on 2D / 3D fusion technology according to claim 1, 2 or 3, characterized in that: Step 6 is as follows: Performing three-dimensional reconstruction of the interior space of the building based on the image data to obtain three-dimensional point cloud data of each key building structure in each building after damage; Compare and calculate the 3D point cloud data of each key building structure after damage with the corresponding 3D point cloud data before damage to obtain the volume change and deflection change of each key building structure; Classify all key building structures as beams, columns, floors, ceilings and walls; Calculate the average volume change and average deflection change of all beams, calculate the average volume change and average deflection change of all columns, calculate the average volume change and average deflection change of all floors, and calculate the average volume change and average deflection change of all walls; The volume damage assessment result is obtained by weighting the average value of the volume changes of beams, columns, floors, ceilings and walls; The deflection damage assessment result is obtained by weighting the average value of the deflection changes of beams, columns, floors, ceilings and walls; The volume damage assessment result and the deflection damage assessment result are the three-dimensional damage information assessment result.

5. The building health assessment method based on 2D / 3D fusion technology according to claim 4 is characterized in that: In step 7, based on the two-dimensional damage information assessment results and the three-dimensional damage information assessment results, the health status of the entire building is obtained as follows: in, For the health of the building, is the two-dimensional damage information assessment result, is the volume damage assessment result, is the deflection damage assessment result, 、 、 is the weight coefficient, Assess the initial condition of the building.

6. The building health assessment method based on 2D / 3D fusion technology according to claim 5 is characterized in that: In step 7, the normalization process is performed to obtain the final evaluation result of the building, specifically: in, For the final evaluation results, is the normalized maximum value of the evaluation structure; when When the building is assessed as damaged at level 1, when When the building is assessed as damaged at level 2, when At that time, the building's health assessment was level 3 damage; when At that time, the building's health assessment was level 4 damage.

7. A building health assessment system based on 2D / 3D fusion technology, characterized in that: Assessing the health status of a building using the method according to any one of claims 1 to 6; The building health assessment system includes: A data acquisition module is used to collect image data of the indoor space of a building; A two-dimensional image processing module is used to evaluate and obtain two-dimensional damage information assessment results of the building; A three-dimensional reconstruction processing module is used to evaluate the three-dimensional damage information of the building; The data fusion processing module is used to output the final assessment results and damage level of the building.

Images