Evaluation method of response and dynamic properties for multi-story structure using multiple unmanned aerial vehicle and computer vision

Abstract

The present invention can provide a method for evaluating the layer-by-layer response and dynamic characteristics of a multi-layer structure using multiple unmanned aerial vehicles and image recognition technology, comprising the steps of: identifying a shooting area of ​​a target structure due to vibration or external force; mounting imaging devices on multiple unmanned aerial vehicles and shooting the shooting area of ​​the target structure using multiple unmanned aerial vehicles; analyzing shooting data obtained by the imaging devices to extract layer-by-layer displacement data of the structure over time; analyzing the dynamic characteristics of the target structure based on the layer-by-layer data of the structure over time; and conducting an evaluation of the analyzed dynamic characteristics of the target structure. According to the above, by using a UAV to photograph and record the constant vibration of a structure, it becomes possible to measure all floors of the structure with minimal manpower, thereby eliminating the need for additional equipment or sensors and allowing for easier and more thorough evaluation of the dynamic characteristics of the structures. According to the above, a method for evaluating the layer-by-layer response and dynamic characteristics of a multi-layer structure can be provided by using multiple unmanned aerial vehicles to perform layer-by-layer superposition imaging of the target structure.

Description

Technology Field

[0001] The present invention relates to a method for evaluating the floor-by-floor response and dynamic characteristics of a multi-layer structure using multiple unmanned aerial vehicles and image recognition technology. More specifically, the invention relates to a method for evaluating the floor-by-floor response and dynamic characteristics of a multi-layer structure using multiple unmanned aerial vehicles and image recognition technology, wherein the dynamic characteristics of the structure can be evaluated by capturing and recording the continuous vibration of a completed structure through field measurements using cameras attached to multiple unmanned aerial vehicles, and analyzing the corresponding data for the superimposed captured areas using image recognition technology. Background Technology

[0003] In general, structural dynamics is intended to understand the patterns of movement of structures or such structural systems subjected to forces that change over time, and it is often based on methods for evaluating the dynamic characteristics of structures using normal vibration measurements, which is one of the methods for assessing the structural health.

[0005] In addition, since structures respond to externally applied forces according to their inherent characteristics, it is essential to evaluate the dynamic characteristics of structures based on appropriate design for dynamic loads. Research on techniques to evaluate the dynamic characteristics of structures by measuring externally applied wind and vibrations, comparing the forces applied to the structure with the corresponding response, which is time-history displacement, is also essential.

[0007] Meanwhile, regarding such dynamic evaluation methods, there is a conventional technique for evaluating the dynamic characteristics of structures through field measurements utilizing sensors such as accelerometers, wind direction and speed meters, and GPS antennas. However, since it is difficult to apply artificial excitation forces to large structures once construction is complete, this technique targets the constant vibration of structures swaying due to general phenomena such as wind. It evaluates the dynamic characteristics of a structure by measuring wind and vibration, comparing the forces applied to the building with the resulting time-historical displacement.

[0009] However, the aforementioned conventional dynamic evaluation method for structures had the problem of requiring the installation and management of various expensive vibration measurement equipment at each reference floor for structural integrity assessment and continuous monitoring. Furthermore, it had limitations in that it required a large amount of manpower and time, and was difficult to apply quickly to new structures.

[0011] Conventional Korean Patent Publication No. 10-2020-0030021 describes a system capable of safety inspection and maintenance through images of a structure captured using a measuring device mounted on an unmanned aerial vehicle. In this system, the shooting area is determined based on the position information of the unmanned aerial vehicle, and the measuring device is composed of a thermal imaging camera and a visible light camera. Based on the captured image data, the system detects the presence and information of defects within the shooting area and calculates defect data. Based on the calculated defect data, the system incorporates information on the displacement of the structure into a structural analysis model to produce analysis data. Based on the calculated analysis data, the system assigns a safety grade to the structure under analysis according to pre-stored safety grade values. Furthermore, it establishes a precise safety diagnosis or repair and reinforcement plan required for the structure based on the assigned safety grade. However, a limitation remains in that it is difficult to evaluate the dynamic characteristics of the structure caused by wind and vibration.

[0013] Furthermore, conventional technology utilizes images collected by a single unmanned aerial vehicle, which limits displacement measurement to specific areas such as the top or bottom floors of a structure, resulting in the problem that it is difficult to measure displacement for the entire structure.

[0015] Accordingly, there is a need for a method to evaluate the floor-by-floor response and dynamic characteristics of multi-story structures using multiple unmanned aerial vehicles and image recognition technology, which can assess the dynamic characteristics of a structure by capturing and recording the continuous vibration of a completed structure through on-site measurements using a camera attached to the unmanned aerial vehicle, and analyzing the data using image recognition technology. Prior art literature

[0016] Korean Published Patent Application No. 10-2020-0030021 (March 19, 2020) Korean Registered Patent Application No. 10-2444976 (September 21, 2022) Korean Registered Patent Application No. 10-2271251 (June 30, 2021) The problem to be solved

[0017] The present invention aims to provide a method for analyzing layer-by-layer data and evaluating dynamic characteristics of a multi-layer structure, which can evaluate the dynamic characteristics of a target structure using only an unmanned aerial vehicle and an embedded camera, requires less manpower, time, and effort compared to conventional evaluation methods, enables effective measurement of vibrations in high-rise, large structures or locations that are difficult for people to reach, enables overlapping imaging of the structure using multiple unmanned aerial vehicles and image recognition technology, and allows for the synchronization of images of the overlapping areas to enable analysis of the entire target structure based on the synchronized images. means of solving the problem

[0019] The present invention relates to a method for evaluating the layer-by-layer response and dynamic characteristics of a multi-layer structure using multiple unmanned aerial vehicles and image recognition technology, wherein the method comprises: a step of identifying a shooting area of ​​a target structure due to vibration or external force; a step of mounting imaging devices on multiple unmanned aerial vehicles and shooting the shooting area of ​​the target structure using multiple unmanned aerial vehicles; a step of analyzing the shooting data obtained by the imaging devices to extract layer-by-layer data of the structure over time; a step of analyzing the dynamic characteristics of the target structure based on the layer-by-layer displacement data of the structure over time; and a step of conducting an evaluation of the analyzed dynamic characteristics of the target structure.

[0021] The step of photographing the shooting area of ​​the above-mentioned target structure is characterized by the fact that when multiple unmanned aerial vehicles photograph the target structure layer by layer, the shooting areas are superimposed.

[0023] The step of extracting floor-by-floor data of the structure according to the above time is characterized by analyzing an image captured by an imaging device through image recognition technology to extract floor-by-floor data of the structure according to time regarding constant vibration and external forces of the structure.

[0025] The step of extracting layer-by-layer data of the structure according to the above time is characterized by including: a reference point being set on the ground when the image is secured from the imaging device according to the movement of a plurality of unmanned aerial vehicles and the image being captured based on the reference point; a step of correcting displacement information based on the captured reference point for the image secured from the imaging device according to the movement of the plurality of unmanned aerial vehicles; a step of removing the intrinsic displacement data of the plurality of unmanned aerial vehicles measured based on the reference point; and a step of extracting time-series response data of the structure corresponding to the reference point.

[0027] The step of correcting the displacement information is characterized by applying phase correction and data synchronization to the unique displacement and unique phase of each of the plurality of unmanned aerial vehicles to enable the extraction of structural response data at the same phase. Effects of the invention

[0029] The method for evaluating the layer-by-layer response and dynamic characteristics of a multilayer structure using multiple unmanned aerial vehicles and image recognition technology according to the present invention can provide the following effects.

[0031] First, by using a UAV to photograph and record the constant vibration of a structure, it becomes possible to measure all floors of the structure with minimal manpower. Consequently, the use of additional equipment or sensors is unnecessary, and the dynamic characteristics of the structure can be evaluated more easily and thoroughly. Furthermore, by providing this information to structural experts to assess the structural integrity and, if necessary, perform repairs and precise safety inspections, it can contribute to the maintenance of the structure.

[0033] Second, measurements can be taken on all floors of a structure by overlapping each floor of the structure being filmed and recorded. By analyzing the captured vibration images using image recognition technology, changes in the movement of the structure over time can be measured using only the images, and the dynamic characteristics of the structure can be evaluated by linking this with existing response-dependent analysis techniques or programs.

[0035] Third, when filming and recording video of constant vibrations or external forces on a structure using an unmanned aerial vehicle, it becomes possible to measure all floors of the structure with minimal manpower, and as a result, the use of additional equipment or sensors is unnecessary, so it can be reviewed more easily and simply than conventional methods.

[0037] Fourth, the structural integrity of existing completed structures, structures under construction, or structures damaged by remodeling, reinforcement, or natural disasters after construction can be efficiently and quickly analyzed and evaluated. This can be used to update computer models during the design phase, predict the future behavior of structures and evaluate their performance, and utilize the evaluated dynamic characteristics to apply to the usability evaluation of structural vibrations. Brief explanation of the drawing

[0039] FIG. 1 is a flowchart illustrating a method for evaluating the layer-by-layer response and dynamic characteristics of a multilayer structure using multiple unmanned aerial vehicles and image recognition technology according to an embodiment of the present invention. FIG. 2 is a diagram illustrating a method for evaluating the layer-by-layer response and dynamic characteristics of a multilayer structure using multiple unmanned aerial vehicles and image recognition technology according to an embodiment of the present invention. FIG. 3 is a diagram showing the overall data extraction process using image recognition technology in the method for evaluating layer-by-layer response and dynamic characteristics of a multilayer structure using multiple unmanned aerial vehicles and image recognition technology according to an embodiment of the present invention. FIG. 4 is a drawing showing an unmanned aerial vehicle that superimposes images of each layer of a multilayer structure in a method for evaluating the layer-by-layer response and dynamic characteristics of a multilayer structure using multiple unmanned aerial vehicles and image recognition technology according to an embodiment of the present invention. FIG. 5 is a drawing showing an unmanned aerial vehicle that superimposes images of each layer of a multilayer structure, including a reference point, in a method for evaluating the layer-by-layer response and dynamic characteristics of a multilayer structure using multiple unmanned aerial vehicles and image recognition technology according to an embodiment of the present invention. FIG. 6 is a graph exemplifying the displacement removal process in the method for evaluating the layer-by-layer response and dynamic characteristics of a multilayer structure using multiple unmanned aerial vehicles and image recognition technology according to an embodiment of the present invention. FIG. 7 is a graph exemplarily showing the extraction of self-vibration data of a structure in a method for evaluating the layer-by-layer response and dynamic characteristics of a multilayer structure using multiple unmanned aerial vehicles and image recognition technology according to an embodiment of the present invention. FIG. 8 is a graph exemplifying a simultaneous measurement process using multiple unmanned aerial vehicles in a method for evaluating layer-by-layer response and dynamic characteristics of a multilayer structure using multiple unmanned aerial vehicles and image recognition technology according to an embodiment of the present invention. FIG. 9 is a graph exemplifying phase correction and synchronization in a method for evaluating layer-by-layer response and dynamic characteristics of a multilayer structure using multiple unmanned aerial vehicles and image recognition technology according to an embodiment of the present invention. FIG. 10 is a graph exemplifying the results of dynamic extraction for vibration modes in a method for evaluating layer-by-layer response and dynamic characteristics of a multilayer structure using multiple unmanned aerial vehicles and image recognition technology according to an embodiment of the present invention. Specific details for implementing the invention

[0040] The present invention will be described in more detail below with reference to preferred embodiments to which it belongs.

[0042] FIG. 1 is a flowchart illustrating a method for evaluating the layer-by-layer response and dynamic characteristics of a multi-layer structure using multiple unmanned aerial vehicles and image recognition technology according to an embodiment of the present invention; FIG. 2 is a diagram illustrating a method for evaluating the layer-by-layer response and dynamic characteristics of a multi-layer structure using multiple unmanned aerial vehicles and image recognition technology according to an embodiment of the present invention; FIG. 3 is a diagram illustrating the overall data extraction process utilizing image recognition technology in the method for evaluating the layer-by-layer response and dynamic characteristics of a multi-layer structure using multiple unmanned aerial vehicles and image recognition technology according to an embodiment of the present invention; FIG. 4 is a diagram illustrating an unmanned aerial vehicle superimposing images of each layer of a multi-layer structure in the method for evaluating the layer-by-layer response and dynamic characteristics of a multi-layer structure using multiple unmanned aerial vehicles and image recognition technology according to an embodiment of the present invention; FIG. 5 is a diagram illustrating an unmanned aerial vehicle superimposing images of each layer of a multi-layer structure including a reference point in the method for evaluating the layer-by-layer response and dynamic characteristics of a multi-layer structure using multiple unmanned aerial vehicles and image recognition technology according to an embodiment of the present invention; FIG. 6 illustrates multiple unmanned aerial vehicles and image recognition technology according to an embodiment of the present invention Figure 7 is a graph showing an exemplary displacement removal process in a method for evaluating the layer-by-layer response and dynamic characteristics of a multi-layer structure using multiple unmanned aerial vehicles and image recognition technology according to an embodiment of the present invention; Figure 8 is a graph showing an exemplary simultaneous measurement process using multiple unmanned aerial vehicles in a method for evaluating the layer-by-layer response and dynamic characteristics of a multi-layer structure using multiple unmanned aerial vehicles and image recognition technology according to an embodiment of the present invention; Figure 9 is a graph showing an exemplary phase correction and synchronization in a method for evaluating the layer-by-layer response and dynamic characteristics of a multi-layer structure using multiple unmanned aerial vehicles and image recognition technology according to an embodiment of the present invention.FIG. 10 is a graph illustrating exemplary dynamic extraction results for vibration modes in a method for evaluating layer-by-layer response and dynamic characteristics of a multilayer structure using multiple unmanned aerial vehicles and image recognition technology according to an embodiment of the present invention.

[0044] A method for evaluating the layer-by-layer response and dynamic characteristics of a multi-layer structure using multiple unmanned aerial vehicles and image recognition technology according to an embodiment of the present invention is characterized by capturing images of vibrations of a structure (1) using multiple unmanned aerial vehicles (10), extracting time-displacement data from the captured vibration images using image recognition technology, and then evaluating dynamic characteristics through response-dependent analysis. (Hereinafter referred to as the 'method for evaluating dynamic characteristics of a structure')

[0046] As shown in FIGS. 1 to 3, a shooting area that can represent the movement of a constantly vibrating target structure (1) is first identified (S10). In this process, the structure, shape, size, and location of the target structure can be identified in advance to plan the flight and positioning of the unmanned aerial vehicle (10). After identifying the shooting area, a plurality of unmanned aerial vehicles (10) are equipped with imaging devices, and the shooting area of ​​the target structure (1) is captured using the mounted imaging devices (S20). Here, it is preferable to use a camera mounted on the unmanned aerial vehicle (10) as the imaging device, and any shape of camera capable of achieving the above purpose is possible.

[0048] Meanwhile, the photographing of the structure (1) using the unmanned aerial vehicle (10) is performed by flying multiple unmanned aerial vehicles (10) equipped with cameras based on a set position of the target structure and a flight path to reach it so that the movement of the target structure (1) can be observed, and then stopping the flight when it reaches the set position. At this time, video is acquired by taking and recording images using the cameras mounted on the unmanned aerial vehicles (10) in response to the movement of the structure (1).

[0050] At this time, multiple unmanned aerial vehicles (10) are made to fly simultaneously so that overlapping areas corresponding to the layer sections of the pre-set target structure (1) can be photographed. By synchronizing images of the overlapping areas through the overlapping photography, information about the overlapping areas can be obtained.

[0052] Additionally, multiple unmanned aerial vehicles (10) may be set up to be photographed simultaneously or at different times, and the flight of the multiple unmanned aerial vehicles (10) may be controlled by a control device that guides and controls the flight. The control device stores dynamic characteristic data values ​​of the multiple unmanned aerial vehicles (10) obtained during the flight process, and the flight may be controlled by adjusting the angle of attack of the wing portion of each of the multiple unmanned aerial vehicles (10) in response to a setting algorithm based thereon.

[0054] After photographing a target structure (1) by means of an imaging device embedded in a plurality of unmanned aerial vehicles (10), the image data obtained by the imaging device is analyzed to extract floor-by-floor displacement data of the structure over time (S30). At this time, the extraction of floor-by-floor displacement data of the structure over time can be performed by analyzing the image captured by the imaging device to detect movement within the image and extracting floor-by-floor displacement data of the structure over time in relation to constant vibration and external forces of the target structure (1). (The floor-by-floor displacement data of the structure over time is described below as time-displacement data.)

[0056] At this time, the captured time-displacement data can be extracted through image recognition technology, and it is preferable to apply Computer Vision for this image recognition technology. Image analysis can be performed on video recorded after capture, but in some cases, video can also be analyzed in real time. Furthermore, the extraction of time-displacement data can be carried out by including the following detailed processes.

[0058] As shown in FIGS. 4 and 5, image distortion caused by the movement of the plurality of unmanned aerial vehicles (10) can be corrected. The correction of image distortion caused by the movement of the unmanned aerial vehicle (10) can be performed by examining the degree of camera shake or distortion caused by the movement of the unmanned aerial vehicle (10) itself and then applying this to the image to correct the image distortion. To this end, distortion correction can be performed by setting a reference point (2) outside the target structure (1), then simultaneously photographing the target structure (1) and the reference point (2) while the unmanned aerial vehicle (10) is in a stationary flight state, and then analyzing the relationship with the target structure (1) based on the reference point (2) to correct the distortion.

[0060] In addition, referring to FIG. 4, it is possible to take superimposed shots by setting the layers of the target structure (1), and it is also possible to take shots of the reference point (2) set as in FIG. 5 simultaneously to analyze and correct the relationship with the target structure (1).

[0062] At this time, the reference point (2) may be set at a predetermined location on the structure (1) to be photographed or on the ground where the structure (1) is installed, and a fixed location where the reference point (2) and the structure (1) to be photographed are fixed together may be applied. In addition, it is also possible to set and apply the reference point (2) to an installation fixed to the structure outside the ground and in another area. Furthermore, such a reference point (2) may be set considering the characteristics of the ground of the structure (1) to be photographed.

[0064] As described above, once image correction is completed by correcting camera shake and distortion, a target point is set for the target structure (1) to enable object tracking. Object tracking can be performed by setting a region of interest as a target point for the target structure (1), extracting feature points within the region of interest, and then tracking the object based on the average of the feature points. A region that serves as a reference for the behavior of the target structure (1) and feature points within that region are extracted from the measured image, and by tracking this, displacement data (time-displacement data) of the structure (1) moving over time can be obtained.

[0066] Meanwhile, the object tracking described above can be performed by positioning multiple unmanned aerial vehicles (10) at a set location on the structure (1) and tracking the object by tracking the average points of feature points obtained from each unmanned aerial vehicle (10). Additionally, a reference point (2) located at a fixed position together with the target structure (1) is simultaneously photographed and utilized. Through this, the self-movement of multiple unmanned aerial vehicles (10) occurring while hovering in the air is measured and removed, thereby enabling image correction through camera distortion correction, and finally, time-displacement data of the target structure (1) itself can be extracted.

[0068] Subsequently, the dynamic characteristics of the target structure (1) are determined based on the extracted time-displacement data (S40). Here, the dynamic characteristics may include displacement and acceleration, natural period and natural frequency, damping ratio, mode shape, etc.

[0070] In the present invention, the dynamic characteristics of the structure (1) can be determined by performing response-dependent analysis (OMA, Operational Modal Analysis) based on time-displacement data.

[0073] As shown in FIGS. 6 to 9, the step (S30) of extracting layer-by-layer data of a structure over time may include setting a reference point on the ground when securing images from a video device based on the movement of multiple unmanned aerial vehicles and taking images based on the reference point, correcting displacement information based on the captured reference point for images secured from the video device based on the movement of multiple unmanned aerial vehicles, removing the unique displacement data of multiple unmanned aerial vehicles measured based on the reference point, and extracting time-series response data of the structure corresponding to the reference point.

[0075] In addition, the step of correcting the displacement information applies phase correction and data synchronization to the unique displacement and unique phase of each of the multiple unmanned aerial vehicles, thereby enabling the extraction of structural response data at the same phase. As shown in FIGS. 8 and 9, multiple unmanned aerial vehicles (10) hover in the air, and the generated unique displacement is removed and the data is synchronized to measure the relative vibration of the structure with respect to the ground floor. The independent measurement data of the multiple unmanned aerial vehicles (10) is synchronized to finally enable the extraction of layer-by-layer time-series response data of the target structure (10) formed in multiple layers.

[0077] Afterwards, an evaluation of the dynamic characteristics of the analyzed target structure is performed (S50). As shown in Fig. 10, it is configured to enable the extraction of dynamic characteristics for not only vibration modes but also higher-order vibration modes, and values ​​for higher-order vibrations can be calculated based on a typical mass participation rate of 90%, which is a major vibration mode.

[0079] As described above, unlike conventional methods that identify dynamic characteristics by installing measurement-related equipment (sensors) on a structure, the present invention utilizes an unmanned aerial vehicle (10) equipped with a camera to record images of a structure (1) in its natural state without installing separate measurement sensors on the structure. After recording these images, the response to constant vibration of the structure (1) can be analyzed through response-dependent analysis based on the analysis of the images, thereby enabling easier evaluation of the dynamic characteristics of the structure. Furthermore, since there is no need for maintenance of related measurement equipment installed on the structure, it can provide a reduction in manpower and costs. Additionally, because the unmanned aerial vehicle (10) simply needs to be flown to photograph the structure, rapid application to the target structure is possible.

[0081] Meanwhile, since the description of the technology disclosed in this specification is merely an example for structural or functional explanation, the scope of the disclosed technology should not be interpreted as being limited by the examples described in the text. That is, since the examples are subject to various modifications and may take various forms, the scope of the disclosed technology should be understood to include equivalents capable of realizing the technical concept. Furthermore, since the purposes or effects presented in the disclosed technology do not imply that a specific example must include all of them or only such effects, the scope of the disclosed technology should not be understood as being limited by them.

[0083] Furthermore, the meaning of the terms described in this invention should be understood as follows. Terms such as “first,” “second,” etc., are intended to distinguish one component from another, and the scope of rights shall not be limited by these terms. For example, the first component may be named the second component, and similarly, the second component may be named the first component.

[0085] Furthermore, when it is stated that one component is “connected” to another component, it should be understood that while it may be directly connected to that other component, there may also be other components in between. Conversely, when it is stated that one component is “directly connected” to another component, it should be understood that there are no other components in between. Meanwhile, other expressions describing the relationships between components, such as “between” and “between” or “adjacent to” and “directly adjacent to,” should be interpreted in the same way.

[0087] A singular expression should be understood to include a plural expression unless the context clearly indicates otherwise, and terms such as “include” or “have” are intended to specify the existence of the set-up features, numbers, steps, actions, components, parts, or combinations thereof, and should be understood not to preclude the existence or addition of one or more other features, numbers, steps, actions, components, parts, or combinations thereof. Explanation of the symbols

[0089] 1: Structure to be photographed 2: Reference point 10 : Unmanned aerial vehicle

Claims

Claim 1 A method for evaluating the layer-by-layer response and dynamic characteristics of a multi-layer structure using multiple unmanned aerial vehicles and image recognition technology, wherein the method comprises: a step of identifying a shooting area of ​​a target structure due to vibration or external force; a step of mounting imaging devices on multiple unmanned aerial vehicles and shooting the shooting area of ​​the target structure using multiple unmanned aerial vehicles; a step of analyzing the shooting data obtained by the imaging devices to extract layer-by-layer displacement data of the structure over time; and a step of analyzing the dynamic characteristics of the target structure based on the layer-by-layer data of the structure over time. A method for evaluating the layer-by-layer response and dynamic characteristics of a multi-layer structure using multiple unmanned aerial vehicles and image recognition technology, comprising: a step of conducting an evaluation of the dynamic characteristics of the analyzed target structure; wherein the step of extracting layer-by-layer data of the structure over time includes setting a reference point on the ground when securing images from a video device based on the movement of multiple unmanned aerial vehicles and taking images based on the reference point, a step of correcting displacement information based on the captured reference point for images secured from the video device based on the movement of multiple unmanned aerial vehicles, a step of removing the intrinsic displacement data of multiple unmanned aerial vehicles measured based on the reference point, and a step of extracting time-series response data of the structure corresponding to the reference point. Claim 2 A method for evaluating the layer-by-layer response and dynamic characteristics of a multi-layer structure using multiple unmanned aerial vehicles and image recognition technology, wherein the step of photographing the shooting area of ​​the target structure is characterized by the shooting area being superimposed when multiple unmanned aerial vehicles photograph the target structure layer by layer. Claim 3 A method for evaluating the layer-by-layer response and dynamic characteristics of a multi-layer structure using multiple unmanned aerial vehicles and image recognition technology, wherein the step of extracting layer-by-layer data of the structure according to time is characterized by analyzing an image captured by an image device through image recognition technology to extract layer-by-layer data of the structure according to time regarding constant vibration and external force of the structure. Claim 4 A method for evaluating layer-by-layer response and dynamic characteristics of a multilayer structure using multiple unmanned aerial vehicles and image recognition technology, wherein the step of correcting displacement information is characterized by applying phase correction and data synchronization to the unique displacement and unique phase of each of the multiple unmanned aerial vehicles to enable the extraction of structural response data at the same phase. Claim 5 In claim 4, the step of correcting the displacement information comprises a plurality of unmanned aerial vehicles hovering in the air, removing the generated natural displacement and synchronizing the data to measure the relative vibration of the structure with respect to the ground floor, and synchronizing the independent measurement data of the plurality of unmanned aerial vehicles to finally extract layer-by-layer time-series response data of the target structure formed in multiple layers, thereby evaluating the layer-by-layer response and dynamic characteristics of the multi-layer structure.

Images