Building leakage point detecting, identifying and positioning method for constructional engineering

Abstract

The invention relates to the technical field of building nondestructive testing, and discloses a building leakage point detecting, identifying and positioning method for building engineering, which comprises the following steps: firstly, adaptively identifying a resonant frequency set of a to-be-detected building structure; then, on the basis of the resonant frequency set, normal excitation and tangential excitation are sequentially applied through a composite orthogonal bimodal excitation unit, and surface micro-motion video data of the structure are synchronously collected by a high-frame-rate optical imaging unit; thirdly, resolving the video data to generate a P-wave velocity field and an S-wave velocity field, and constructing a wave velocity ratio field sensitive to leakage characteristics; and finally, multi-dimensional features are fused to construct a high-dimensional feature vector, an unsupervised learning algorithm is utilized to calculate an abnormal score, and automatic identification and accurate positioning of the leakage point are realized. According to the method, the sensitivity and the accuracy of detection are remarkably improved by utilizing a resonance enhancement mechanism and a wave velocity ratio diagnosis basis sensitive to a water medium, and automation of a whole detection process and objectification of result interpretation are realized by introducing unsupervised learning.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of building non-destructive testing, in particular to a building leakage point detection, identification and positioning method for building engineering. BACKGROUND

[0002] Building leakage is a common and highly dangerous problem in the field of building engineering. It not only damages the decoration and living environment of the building, but also erodes the structural steel, affects the structural safety and service life of the building, and causes significant economic losses. Therefore, timely and accurate detection and positioning of building leakage points are crucial for the maintenance and repair of buildings.

[0003] Currently, there are various technical methods for building leakage detection, mainly including infrared thermal imaging method, acoustic emission detection method, resistivity method and microwave method, etc. The infrared thermal imaging method identifies leakage by detecting the surface temperature anomalies caused by water evaporation or specific heat capacity differences in the leakage area; the acoustic emission detection method locates the leakage by capturing weak acoustic signals generated by the flow or dripping of water; the resistivity method and the microwave method use the changes in the conductivity and dielectric constant of the water-containing area for detection.

[0004] However, these conventional detection methods have inherent limitations in practical application. On the one hand, the detection sensitivity of these methods is limited, especially for early-stage small leaks or internal leaks with complex penetration paths, the physical signals (such as temperature difference, acoustic signal) generated are extremely weak, and are easily overwhelmed by the complex environmental noise and structural background signals on site, resulting in low signal-to-noise ratio and easy missed detection.

[0005] On the other hand, many existing diagnostic criteria have ambiguity and indirectness, which affects the accuracy of identification. For example, the temperature anomalies detected by the infrared thermal imaging method have multiple causes, such as uneven structure materials, internal cavities or thermal bridge effects, etc. non-leakage factors; the source of acoustic emission signals can also be confused with other friction and vibration sources. The non-uniqueness of such diagnostic criteria makes the detection results have a high misdiagnosis rate, and it is difficult to provide conclusive evidence of leakage.

[0006] In addition, the implementation process and result interpretation of existing detection technologies largely depend on the professional knowledge and subjective experience of the detection personnel. The operator needs to manually set complex instrument parameters according to the on-site situation, and interpret the detection graph or signal through naked eye observation and experience judgment, which not only leads to low detection efficiency and poor result repeatability, but also lacks an objective and standardized evaluation system, making it difficult to guarantee the reliability of the detection results. Therefore, developing a building leakage detection method with high sensitivity, high accuracy and automatic identification has become a technical problem to be solved in the field. SUMMARY

[0007] The technical problem to be solved by the present application is that the existing building leakage detection technology has the problems of being insensitive to early micro leakage, being easily disturbed by environmental noise, being dependent on manual experience for detection results, and being difficult to accurately locate and visualize the leakage area.

[0008] To solve the above technical problems, the first aspect of the present application provides a building leakage point detection, identification and positioning method for building engineering. The core innovation of the method is that the micro-motion response of the structure is amplified by orthogonal dual-mode resonance excitation, and the difference in propagation speed of different mechanical waves (P waves and S waves) in water-containing medium is used as a high-sensitivity diagnostic basis, and finally combined with an unsupervised machine learning algorithm to realize automatic and high-precision identification of the leakage area.

[0009] The method comprises the following steps:

[0010] S1, system layout and parameter initialization: a high-frame-rate optical imaging unit is laid out to cover the region of interest of the building structure to be tested, and a composite orthogonal dual-mode excitation unit is coupled to the building structure to be tested.

[0011] S2, adaptive resonance frequency identification: identify the resonance frequency set of the building structure to be tested. The significance of this step is that applying excitation at the structural resonance frequency can obtain the maximum structural response with the minimum energy input, thereby greatly improving the signal-to-noise ratio of the detection.

[0012] S3, orthogonal dual-mode resonance data acquisition: based on the resonance frequency set, normal excitation and tangential excitation are applied to the building structure to be tested in turn, and the high-frame-rate optical imaging unit is used to synchronously collect the surface micro-motion video data of the building structure to be tested under the normal excitation and tangential excitation. Normal excitation mainly excites P waves, while tangential excitation mainly excites S waves, providing mutually orthogonal original data for subsequent multi-dimensional feature calculation.

[0013] S4, multi-dimensional micro-motion feature field calculation: calculate the surface micro-motion video data to generate P wave velocity field and S wave velocity field corresponding to the normal excitation and tangential excitation respectively, and calculate the wave velocity ratio field based on the P wave velocity field and S wave velocity field. Due to the incompressibility of water, the bulk modulus of the medium in the leakage area increases significantly while the shear modulus remains unchanged, resulting in a significant increase in P wave velocity and a slight decrease in S wave velocity. Therefore, the ratio of the two, i.e. the wave velocity ratio field, can greatly amplify the abnormal features of the leakage area.

[0014] S5, feature fusion and leakage area identification: based on the wave velocity ratio field, a high-dimensional feature vector representing the physical state of each pixel point is constructed, and an unsupervised learning anomaly detection algorithm is used to calculate the anomaly score of each pixel point to identify and locate the leakage point. This method does not require pre-labeled samples, can automatically discover abnormal patterns from high-dimensional data, and realizes the intelligentization and objectivity of leakage identification.

[0015] In one specific embodiment, S2 specifically comprises: applying a broadband sweep excitation to the building structure to be tested and collecting its dynamic response, constructing a structure transfer function by Fourier transform and spatial accumulation of the dynamic response, and applying a peak search algorithm to the structure transfer function to automatically and accurately identify the set of resonance frequencies.

[0016] Preferably, in S4, the generation of the P-wave velocity field and the S-wave velocity field specifically comprises:

[0017] The P-wave phase field and the S-wave phase field are calculated from the surface micro-motion video data and The P-wave wave number field and the S-wave wave number field are obtained by calculating the spatial gradient of the phase field and The P-wave velocity field and the S-wave velocity field are calculated according to the following formula and :

[0018] ;

[0019] ;

[0020] wherein, is the resonance frequency used for excitation.

[0021] Further, the P-wave phase field and the S-wave phase field are obtained by quadrature demodulation of the time-domain displacement signal extracted from the surface micro-motion video data.

[0022] Preferably, in S5, the construction of the high-dimensional feature vector specifically comprises:

[0023] For each pixel point in the area to be tested, the wave velocity ratio field feature value calculated at each frequency in the set of resonance frequencies and the P-wave amplitude field and S-wave amplitude field feature value are concatenated to construct the high-dimensional feature vector. This fusion of multiple frequencies and multiple physical dimensions enhances the robustness and accuracy of identification.

[0024] In one specific embodiment, the unsupervised learning anomaly detection algorithm is an isolation forest algorithm.

[0025] Preferably, the identification and positioning of the leakage point in S5 specifically comprises: generating an anomaly score atlas according to the anomaly scores of the respective pixel points, and performing adaptive threshold segmentation on the anomaly score atlas to generate a binary mask image identifying the leakage point.

[0026] Further, after generating the binary mask image, it further comprises a morphological processing operation to eliminate noise and optimize the integrity of the leakage area.

[0027] Preferably, after identifying and positioning the leakage point, it further comprises a step of superimposing the identified leakage point area in pseudo-color on the original optical image for visual presentation, providing an intuitive diagnostic result for the user.

[0028] The second aspect of the present application provides a building leakage point detection, identification and positioning system for construction engineering, which comprises:

[0029] a composite orthogonal dual-mode excitation unit;

[0030] a high-frame-rate optical imaging unit;

[0031] and an information processing and control unit.

[0032] The present application provides a building leakage point detection, identification and positioning method for construction engineering. It has the following beneficial effects:

[0033] 1. The present application can obtain maximum structure surface micro-motion response with extremely low excitation energy by adaptive identification and excitation at the structural resonance frequency, thereby significantly improving the signal-to-noise ratio. This resonance enhancement mechanism enables the capture of early and weak leakage signals far superior to traditional non-resonant detection methods, and has strong suppression ability to environmental vibration noise.

[0034] 2. The present application does not rely on a single amplitude or phase anomaly, but innovatively utilizes the differential response of P-wave and S-wave wave velocity to the change of physical parameters of water-containing medium. By constructing a "wave velocity ratio field" which is extremely sensitive to leakage characteristics as the core diagnostic index, it can effectively exclude the interference of non-leakage factors such as material non-uniformity and internal small cavities, thereby fundamentally ensuring the reliability and accuracy of the diagnostic results.

[0035] 3. The present application constructs a high-dimensional feature vector by fusing multi-dimensional feature information, and introduces an unsupervised learning algorithm for anomaly detection, which eliminates the serious dependence of traditional detection on professional experience of operating personnel. The entire identification process does not require human intervention and preset threshold, and can automatically and objectively locate the leakage area, greatly improving the detection efficiency and standardization level, and finally presenting the results in a visual manner, which is intuitive and easy to understand. BRIEF DESCRIPTION OF DRAWINGS

[0036] Figure 1 This is a schematic diagram of a building leakage point detection, identification and location system for building engineering according to an embodiment of the present invention;

[0037] Figure 2 This is a flowchart illustrating a method for detecting, identifying, and locating building leakage points in a construction project, according to an embodiment of the present invention.

[0038] Figure 3 This is a schematic diagram of the adaptive resonant frequency identification process according to an embodiment of the present invention;

[0039] Figure 4 This is a schematic diagram of the orthogonal dual-mode resonance data acquisition process according to an embodiment of the present invention;

[0040] Figure 5 This is a schematic diagram illustrating the process of solving a multidimensional micro-motion feature field according to an embodiment of the present invention.

[0041] Figure 6 This is a schematic diagram illustrating the feature fusion and leakage area identification process according to an embodiment of the present invention.

[0042] Among them, 100 is a composite orthogonal dual-modal excitation unit; 110 is a normal excitation component; 120 is a tangential excitation component; 130 is a coupling interface; 200 is a high frame rate optical imaging unit; 210 is an image sensor; 220 is an optical lens; 230 is a synchronous trigger interface; 300 is an information processing and control unit; 310 is a signal generation and control module; 320 is a data acquisition and synchronization module; 330 is a feature calculation module; and 340 is a fusion recognition and visualization module. Detailed Implementation

[0043] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0044] See attached document Figure 1 The present invention provides a building leakage point detection, identification and location system for building engineering. The system includes: a composite orthogonal dual-modal excitation unit 100, a high frame rate optical imaging unit 200 and an information processing and control unit 300.

[0045] The information processing and control unit 300 is electrically connected or in signal communication with the composite orthogonal dual-mode excitation unit 100 and the high-frame-rate optical imaging unit 200. In operation, the information processing and control unit 300 sends excitation control signals to the composite orthogonal dual-mode excitation unit 100 and sends synchronization triggering and acquisition instructions to the high-frame-rate optical imaging unit 200. The image sequence data collected by the high-frame-rate optical imaging unit 200 is transmitted to the information processing and control unit 300 for subsequent processing.

[0046] Specifically, the composite orthogonal dual-mode excitation unit 100 functions to apply controllable, normal and tangential orthogonal micro-vibrations to the surface of the measured building structure. The composite orthogonal dual-mode excitation unit 100 can include a normal excitation assembly 110, a tangential excitation assembly 120, and a coupling interface 130.

[0047] The normal excitation assembly 110 is used to generate vibrations perpendicular to the surface of the measured structure, which can be composed of a piezoelectric ceramic stack;

[0048] The tangential excitation assembly 120 is used to generate vibrations parallel to the surface of the measured structure, which can be composed of a shear mode piezoelectric sheet or an electromagnetic exciter;

[0049] The coupling interface 130 is used to ensure good mechanical energy transmission between the excitation unit and the surface of the measured structure.

[0050] The composite orthogonal dual-mode excitation unit 100 can selectively output wideband excitation signals or single-frequency harmonic excitation signals of specific frequencies in response to instructions from the information processing and control unit 300.

[0051] The high-frame-rate optical imaging unit 200 functions to non-contactly collect micro-motion response video sequences of the surface of the measured building structure under excitation. The high-frame-rate optical imaging unit 200 includes an image sensor 210, an optical lens 220, and a synchronization triggering interface 230.

[0052] The image sensor 210 is an industrial-grade high-speed CMOS or CCD sensor with sufficient frame rate and spatial resolution to capture the high-frequency vibration response of the measured structure under excitation;

[0053] The optical lens 220 is a high-resolution lens for clearly imaging the measured area on the image sensor 210;

[0054] The synchronization triggering interface 230 is connected to the information processing and control unit 300 for receiving synchronization instructions to ensure that the starting time of image sequence acquisition is accurately aligned with the application of vibration excitation. The digital video sequence collected by the unit can be represented as wherein is the pixel coordinate of the image, is the time variable.

[0055] The information processing and control unit 300, as the central core of the system, is responsible for performing all the controls, data processing, algorithm operations and result presentations. The unit is a hardware computing platform, and its internal functions are realized by software modules. These modules include a signal generation and control module 310, a data acquisition and synchronization module 320, a feature solving module 330 and a fusion recognition and visualization module 340.

[0056] The signal generation and control module 310 is responsible for generating the digital waveform signals required to drive the composite orthogonal dual-mode excitation unit 100, including wide-band signals for pre-scanning and single-frequency sinusoidal signals for main detection, and controlling the switching of excitation modes (normal or tangential).

[0057] The data acquisition and synchronization module 320 is responsible for sending acquisition instructions to the high-frame-rate optical imaging unit 200 through the synchronous trigger interface 230, and receiving and caching the video sequences transmitted by the unit .

[0058] The feature solving module 330 is responsible for executing core algorithms to process the cached video sequences and solve multi-dimensional physical feature fields, including but not limited to amplitude field, phase field, P-wave velocity field, S-wave velocity field and wave velocity ratio field.

[0059] The fusion recognition and visualization module 340 is responsible for fusing the multiple feature field data output by the feature solving module 330, constructing a high-dimensional feature vector, and using an unsupervised learning algorithm to analyze the feature vector to identify abnormal areas. Finally, the position information of the identified leakage area is superimposed on the original optical image to generate a visualization result.

[0060] Referring to the accompanying Figure 2 , the present application provides a building leakage point detection, recognition and positioning method for construction engineering, which can include the following steps:

[0061] S201, system layout and parameter initialization: the layout includes arranging the composite orthogonal dual-mode excitation unit 100 and the high-frame-rate optical imaging unit 200 on the surface area of the building structure to be measured, and establishing the mapping relationship between the pixel coordinate system and the physical coordinate system;

[0062] S202, adaptive resonant frequency identification: by applying a wide-band excitation signal and collecting the preliminary dynamic response of the structure surface, the structure transfer function of the measured area is calculated and constructed, so as to identify one or more resonant frequencies;

[0063] S203, orthogonal bimodal resonance data acquisition: that is, in each identified resonance frequency, the normal vibration mode and the tangential vibration mode are excited respectively, and the corresponding micro-motion response video sequence is synchronously collected by the high-frame-rate optical imaging unit 200;

[0064] S204, multi-dimensional micro-motion characteristic field calculation: processing each group of collected video sequences to calculate and construct a plurality of characteristic field maps including amplitude field, phase field, P-wave velocity field, S-wave velocity field and wave velocity ratio field;

[0065] S205, feature fusion and leakage area identification: fusing the data of the plurality of characteristic field maps into a high-dimensional feature vector, analyzing the feature vector by using an unsupervised anomaly detection algorithm to locate a feature anomaly area, and finally generating a visualization result map superimposed with leakage position information.

[0066] The above-mentioned steps in the method of the embodiment of the application will be described in detail below

[0067] S201, system layout and parameter initialization:

[0068] When the detection and positioning method of the application is executed, system layout and parameter initialization are first performed. This step provides a basis for subsequent accurate measurement. This step involves on-site arrangement of the composite orthogonal bimodal excitation unit 100 and the high-frame-rate optical imaging unit 200, and necessary calibration of the system.

[0069] Specifically, a region of interest (ROI) is selected on the surface of the building structure to be measured, which is the range suspected to have leakage or needing screening. The composite orthogonal bimodal excitation unit 100 is arranged at the edge or center of the region of interest. In order to ensure that the excitation energy can be efficiently and stably transmitted to the building structure, an acoustic coupling agent, which can be an ultrasonic coupling agent, glycerol or high-viscosity grease, is applied between the coupling interface 130 of the excitation unit and the structure surface to fill the small gap and achieve good acoustic impedance matching. At the same time, the normal excitation direction of the excitation unit is perpendicular to the structure surface, and the tangential excitation direction is parallel to the structure surface.

[0070] The high-frame-rate optical imaging unit 200 is arranged opposite the region of interest, and the focal length of the optical lens 220 is adjusted so that the natural texture or artificial marker points of the structure surface are clearly imaged on the image sensor 210. The distance between the high-frame-rate optical imaging unit 200 and the structure surface and the lens viewing angle are adjusted to ensure that the field of view can completely cover the entire region of interest. In the case of unstable or insufficient ambient light, an auxiliary light source can be enabled to uniformly and constantly illuminate the region of interest, so as to eliminate the disturbance of ambient light fluctuations on image acquisition.

[0071] To ensure the accuracy of the subsequent physical quantity calculation, the mapping relationship between the image pixel coordinate system and the physical world coordinate system of the measured structure surface needs to be established. This process is achieved through camera calibration. A specific embodiment is to tightly attach a planar calibration board with a known size array pattern (such as a checkerboard or a circular dot array) to the surface of the measured region of interest, and capture a static image of the calibration board by the high-frame-rate optical imaging unit 200.

[0072] The information processing and control unit 300 then calls the camera calibration algorithm, identifies the positions of the corner points or the centers of the calibration pattern in the image in the pixel coordinate system, and calculates the transformation matrix between the two coordinate systems using the known accurate coordinates of these points in the physical world coordinate system. This transformation relationship can be described by a 3x3 homography matrix , which establishes the mapping between the physical world coordinates and the corresponding image pixel coordinates . This mapping relationship can be expressed in homogeneous coordinates as:

[0073] ;

[0074] is the physical coordinate of a point on the measured structure surface, with units of meters or millimeters.

[0075] is the corresponding pixel coordinate of the point on the image sensor.

[0076] is a non-zero scale factor.

[0077] is the homography transformation matrix calculated through the calibration process.

[0078] After calibration, the calibration board is removed. The calculated transformation matrix is stored in the information processing and control unit 300, which is used in subsequent S204 to convert the spatial gradient calculated in pixels to a spatial gradient expressed in physical units, thereby ensuring that the calculation results of physical characteristic quantities such as wave velocity field have accurate physical meaning.

[0079] After the system setup and parameter initialization are completed, adaptive resonant frequency identification is entered. This step is an important technical feature of the present application, and its purpose is to intelligently determine the optimal excitation frequency by preliminary exploration of the dynamic characteristics of the measured structure, thereby replacing the traditional fixed frequency or blind frequency sweep detection method.

[0080] The physical basis of this step is that any building structure has its inherent vibration characteristics, which are represented by a series of natural frequencies or resonance frequencies. When the frequency of external excitation is consistent with the resonance frequency of the structure, resonance phenomenon occurs, at which time the structure only needs a small excitation energy to produce a vibration response much larger than that at non-resonance frequencies. The existence of the leakage area changes the overall or local resonance characteristics of the structure, or generates new local resonance modes related to the leakage defect, because it changes the local mass, stiffness and damping. Therefore, by identifying and utilizing these resonance frequencies most sensitive to changes in the internal medium of the structure for excitation, the response difference between the leakage area and the intact area can be greatly amplified, thereby significantly improving the signal-to-noise ratio of the detection signal and the sensitivity to weak leakage defects.

[0081] Referring to the drawings Figure 3 To realize automatic identification of resonance frequencies, the embodiment first applies a wideband excitation signal to the structure and collects the preliminary response. The signal generation and control module 310 of the information processing and control unit 300 generates a preset wideband excitation signal and drives the composite orthogonal dual-mode excitation unit 100 to excite the structure. The excitation mode here can be a normal excitation mode with uniform energy distribution.

[0082] The wideband excitation signal is a linear frequency modulation signal, i.e. a Chirp signal. The instantaneous frequency of the signal changes linearly with time. The signal In the time interval It can be expressed as:

[0083] ;

[0084] Where:

[0085] is the amplitude of the excitation signal, which is controlled in a small range that will not cause any damage to the building structure;

[0086] is the starting frequency of the scan;

[0087] is the termination frequency of the scan;

[0088] The frequency range is selected according to the general response characteristics of the structure material (such as concrete, brick wall) and thickness, for example, it can be set to 10 Hz to 1000 Hz.

[0089] is the duration of the linear frequency modulation signal.

[0090] While the composite ortho-dual-mode excitation unit 100 is applying the wideband excitation signal, the data acquisition and synchronization module 320 of the information processing and control unit 300 synchronously triggers the high-frame-rate optical imaging unit 200 to continuously capture the dynamic response video sequence of the measured region during the entire excitation duration and stores this sequence in the buffer for subsequent use in the structural transfer function construction step.

[0091] After the preliminary dynamic response video sequence is captured, the feature calculation module 330 of the information processing and control unit 300 processes it to construct the structural transfer function and identify the resonant frequency.

[0092] First, a micro-motion analysis algorithm is called to extract the time-domain displacement signal of each pixel from the video sequence . A specific embodiment is that the micro-motion analysis algorithm is a phase-based spatial optical flow field algorithm. This algorithm estimates the sub-pixel level displacement of each pixel by analyzing the change in the local phase information of its neighborhood in the video sequence over time, and it has high robustness to light changes. The output of this step is the time-domain displacement signal of each pixel in the entire region of interest . To analyze the frequency response characteristics of the measured region, the feature calculation module 330 applies the fast Fourier transform (FFT) to the time-domain displacement signal of each pixel to convert it from the time domain to the frequency domain and obtain the complex frequency response of each pixel . This transformation can be represented as:

[0093] ;

[0094] where: is the frequency variable; is the time variable; is the imaginary unit; is the duration of the wideband excitation signal.

[0095] To obtain the structural transfer function characterizing the overall dynamic characteristics of the measured region, the frequency response amplitudes of all pixels need to be spatially accumulated. This structural transfer function is constructed by integrating or summing the frequency response modulus of all pixels in the entire region of interest (ROI):

[0096] ;

[0097] This structural transfer function is a curve describing the amplitude of the response of the structure to different frequency excitations. The local peak points on this curve correspond to the resonant frequency points of the structure.

[0098] Through this algorithm, a set of one or more resonant frequencies is finally obtained . This set is passed to the signal generation and control module 310 for guiding the subsequent orthogonal bimodal resonant excitation main detection step.

[0099] Referring to the accompanying Figure 4 , after the adaptive resonant frequency identification determines the resonant frequency set , the method enters the orthogonal bimodal resonant data acquisition. In this step, the signal generation and control module 310 of the information processing and control unit 300 automatically generates and executes a predetermined excitation control sequence based on the identified resonant frequency set , to carry out the main data acquisition.

[0100] The generation and execution process of the excitation control sequence is as follows: the information processing and control unit 300 first sets the excitation mode to the normal excitation mode. Then, each frequency in the resonant frequency set is traversed. For the first resonant frequency in the set, the signal generation and control module 310 generates a steady-state single-frequency sinusoidal harmonic signal with a frequency of , and outputs it to the composite orthogonal bimodal excitation unit 100 through the digital analog converter, while issuing a control instruction to activate only the normal excitation component 110 of the composite orthogonal bimodal excitation unit 100. The single-frequency harmonic signal can be represented as:

[0101] ;

[0102] where:

[0103] is the amplitude of the resonant excitation, which is set to ensure that the structure surface produces a micro-motion response that can be clearly captured by the high-frame-rate optical imaging unit 200, but is far below any level that may cause structural fatigue or damage;

[0104] is the current resonant frequency selected from the set ;

[0105] is the time variable.

[0106] After normal excitation at frequency and completion of data acquisition, the signal generation and control module 310 continues to select the next resonant frequency in the set, generates the corresponding sinusoidal harmonic signal and performs excitation, until the set All the frequencies are used for normal excitation.

[0107] After the normal excitation of all frequencies is completed, the signal generation and control module 310 switches the excitation mode to tangential excitation mode. This switching is achieved by sending a specific control signal to the compound orthogonal dual-mode excitation unit 100, which deactivates the normal excitation assembly 110 and activates the tangential excitation assembly 120. Then, the set of resonance frequencies is traversed again in the same order as the normal excitation, using each resonance frequency in turn to generate a single-frequency harmonic signal and apply tangential excitation.

[0108] In this way, the excitation control sequence ensures that, at each resonance frequency point sensitive to structural defects, dynamic data caused by normal excitation (which produces a response dominated by compression waves) and tangential excitation (which produces a response dominated by shear waves) are collected separately, providing a complete data basis for subsequent comprehensive calculation of multi-dimensional feature fields.

[0109] For each single-frequency resonance excitation event in the aforementioned excitation control sequence, the data acquisition and synchronization module 320 of the information processing and control unit 300 performs a data acquisition operation that is precisely synchronized with it, to ensure that the video data collected has a determined time and phase relationship with the excitation signal.

[0110] Specifically, at the start of an excitation event, for example when the compound orthogonal dual-mode excitation unit 100 starts to perform normal excitation at the resonance frequency , the system waits for a preset stabilization time . The stabilization time is set to ensure that the forced vibration of the structure surface reaches a steady state, eliminating the influence of transient response.

[0111] After the stabilization time ends, the data acquisition and synchronization module 320 generates a digital trigger pulse signal through a hardware trigger circuit. The trigger pulse signal is a TTL level signal. This trigger pulse signal is sent to two targets at the same time:

[0112] First, it is sent to the high-frame-rate optical imaging unit 200 through the synchronization trigger interface 230. The rising edge or falling edge of the pulse serves as the starting instruction for image acquisition, causing it to immediately start capturing a video sequence according to the preset frame rate and total number of frames.

[0113] Second, the pulse signal also serves as a reference signal for time zero recorded by the signal generation and control module 310, which is used to mark the initial phase of the excitation sinusoidal waveform .

[0114] The hardware trigger mechanism, rather than a software instruction trigger, can control the time delay jitter between the excitation and the acquisition to the nanosecond or microsecond level, thereby providing a reliable time reference for subsequent high-precision phase field calculation.

[0115] The high-frame-rate optical imaging unit 200 continuously acquires video data for a preset time period after receiving the trigger signal. To ensure the accuracy of subsequent Fourier analysis, the acquisition time is set to an integer multiple of the excitation signal period This can effectively suppress spectral leakage and improve signal-to-noise ratio.

[0116] This synchronous data acquisition process is repeated for each excitation event (i.e., each resonant frequency under each excitation mode) in the excitation control sequence.

[0117] Finally, the information processing and control unit 300 obtains and stores a series of video sequence data corresponding to each excitation event, denoted as the normal excitation video set and the tangential excitation video set where traverses all the identified resonant frequencies. These video sequences are the direct input for the subsequent multi-dimensional feature field calculation step.

[0118] Referring to the accompanying drawings Figure 5 After completing data acquisition, the multi-dimensional micro-motion feature field is calculated. The goal of this step is to accurately extract multi-dimensional features that can represent the local physical properties of the structure from the raw video data collected. This step is performed by the feature calculation module 330 of the information processing and control unit 300. First, a micro-motion displacement extraction operation is performed on each video sequence in the normal excitation video set and the tangential excitation video set This operation is consistent with the phase-based spatial optical flow field algorithm used in S202, which converts the video sequence into a pixel-level time-domain displacement signal .

[0119] Since the excitation is a single-frequency steady-state harmonic signal, the displacement response of any point on the structure surface can also be modeled as a harmonic function with the same frequency , which has the general form:

[0120] ;

[0121] where:

[0122] ​ is the amplitude of vibration at this point, which reflects the distribution of vibration energy at this point.

[0123] is the phase delay of vibration response at this point relative to the excitation source, whose spatial distribution reflects the delay characteristics of wave propagation in the medium.

[0124] To accurately solve the amplitude and phase from the time-domain signal , the feature calculation module 330 adopts an orthogonal demodulation algorithm. This algorithm can effectively suppress the noise of non-excitation frequencies and directly solve the amplitude and phase information by operating the measured signal with a pair of reference signals with the same frequency but orthogonal phase.

[0125] Specifically, for the displacement signal of each pixel point, the algorithm first calculates its in-phase component and quadrature component . This calculation is achieved by multiplying with reference signals and respectively, and integrating over the entire acquisition duration :

[0126] ;

[0127] ;

[0128] According to the integral and difference formulas of trigonometric functions, the above integral operation can filter out high-frequency components with frequency when is an integer multiple of the excitation period, and only retain the direct current components related to amplitude and phase. The obtained in-phase component and quadrature component are related to amplitude and phase as and . Therefore, the final amplitude and phase can be calculated as follows:

[0129] ;

[0130] ;

[0131] where is the four-quadrant arctangent function, and its calculation result can clearly determine the phase angle in ​​the quadrant ambiguity problem brought by single arctangent function is avoided.

[0132] The feature calculation module 330 performs the above-mentioned quadrature demodulation calculation for all pixel points in the normal excitation video and the tangential excitation video under each excitation frequency . The final output of this process is that, for each , four two-dimensional feature field maps are generated: P-wave amplitude field , P-wave phase field , S-wave amplitude field and S-wave phase field . These feature fields are the basis for constructing subsequent higher-order diagnostic features.

[0133] After the amplitude field and the phase field of the P-wave and the S-wave are generated, the feature calculation module 330 continues to process the phase field to estimate and construct a dynamic wave velocity field. The physical principle of this estimation is based on wave theory, that is, the propagation velocity of a plane harmonic wave is the ratio of its angular frequency to its wave number . Among them, the angular frequency is a known excitation parameter, and the wave number , as a measure of the phase change of the wave in a unit physical distance, is equal to the modulus of the spatial gradient of the phase field .

[0134] Since the phase field is two-dimensional data in the form of a discrete pixel grid, its spatial gradient needs to be calculated by numerical methods. One specific implementation is to use the central difference method to calculate the partial derivatives at each pixel point . For the P-wave phase field , its partial derivatives in the and directions can be approximated as:

[0135] ;

[0136] ;

[0137] where and represent the horizontal and vertical distances of the pixel point in the physical world coordinate system, which have been obtained through the camera calibration process of the preparation and S201. To enhance the robustness of the calculation to noise, another implementation can use an image gradient operator containing a smoothing filter, such as the Sobel operator, to calculate the two partial derivative components. After calculating the partial derivative components of the phase field, the local P-wave wave number at the point can be obtained, which is the modulus of the phase gradient vector:

[0138] ;

[0139] Similarly, the phase field of S-wave can be obtained by performing the same gradient calculation: Finally, according to the relationship between angular frequency and wave number, the P-wave velocity and S-wave velocity at each pixel point can be calculated:

[0140] ;

[0141] ;

[0142] For each harmonic frequency , the above calculations are performed on both the P-wave phase field and the S-wave phase field, resulting in the P-wave velocity field and the S-wave velocity field . These two velocity fields are key intermediate physical quantities for constructing the final diagnostic features.

[0143] After obtaining the P-wave velocity field and the S-wave velocity field , the feature calculation module 330 generates a key diagnostic feature field with higher sensitivity and stronger indication for leakage areas by combining these two physical quantities.

[0144] A core technical solution of the present application is to construct and utilize the wave velocity ratio field as a core indicator for identifying leakage. This wave velocity ratio field is calculated by point-by-point division of the P-wave velocity and the S-wave velocity at each pixel point:

[0145] ;

[0146] This wave velocity ratio field has a clear and sensitive physical meaning, and an abnormal increase in its value directly indicates an increase in the water content in the pores of building materials. This physical meaning is rooted in the basic principles of elastic wave propagation in porous media (such as concrete and masonry structures).

[0147] Specifically, the propagation speed of P-wave (compressional wave) is mainly determined by the bulk modulus and shear modulus of the medium, which reflects the medium's ability to resist volume changes. Water, as a nearly incompressible fluid, has a high bulk modulus. When water seeps into and fills the internal pores of materials such as concrete, it significantly enhances the overall bulk modulus of the water-solid composite medium. Therefore, in water-containing areas, the propagation speed of P-wave usually does not decrease significantly, and may even remain at a high level.

[0148] In contrast, the propagation speed of S-wave (shear wave) is only determined by the shear modulus of the medium, which reflects the medium's ability to resist shape changes. As a fluid, water has no ability to resist shear deformation, and its shear modulus is zero. When water fills the pores of the material, it not only cannot contribute to the shear stiffness of the composite medium, but also acts as a lubricant between solid particles, significantly reducing the overall shear modulus of the region. The reduction of shear modulus directly leads to the sharp drop of the propagation speed of S-wave in the water-containing region.

[0149] In summary, in the water-containing region caused by leakage, due to the relative stability or slight decrease in the significant decrease in the two effects superimposed on each other, resulting in their ratio local, large-scale increase. Therefore, the wave velocity ratio field converts the physical and chemical changes of leakage into a two-dimensional physical field feature with high contrast and high signal-to-noise ratio, which can be accurately measured.

[0150] In addition to the wave velocity ratio field, in this embodiment, the previously calculated P-wave amplitude field and S-wave amplitude field are also retained as auxiliary diagnostic features. The water-containing region will increase the internal damping of the material, causing the energy attenuation of the wave during propagation to intensify, which will be reflected in the local

[0151] rapid attenuation of the amplitude field. The feature calculation module 330 generates a set of feature fields for each resonant frequency , providing a rich and complementary multi-dimensional data set for the final feature fusion and recognition step.

[0152] Referring to the accompanying Figure 6 , after completing the calculation of all feature fields, the method of the present application enters S205. This step is performed by the fusion recognition and visualization module 340 of the information processing and control unit 300, which is primarily responsible for integrating the multiple, discrete feature field map data generated in the previous step into a unified, high-dimensional feature space, providing input for the subsequent intelligent recognition algorithm. This integration process is carried out at the pixel level. For any pixel point in the measured region of interest (ROI), the system will collect multiple feature values generated by different physical mechanisms (i.e. P-wave and S-wave) under different excitation conditions (i.e. different resonant frequencies ), and concatenate them in a predetermined order to construct a multi-dimensional feature vector specific to that pixel point.

[0153] Specifically, for each pixel point high-dimensional eigenvector of the pixel point is constructed as follows:

[0154] The vector is formed by concatenating the eigenvalues calculated at all resonant frequencies . In one embodiment, the eigenvalues selected at each frequency include the P-wave amplitude , the S-wave amplitude , and the wave velocity ratio . The high-dimensional eigenvector of the pixel point can then be expressed as:

[0155] where: is the high-dimensional eigenvector of the pixel point. is the P-wave amplitude of the point at the th resonant frequency. is the S-wave amplitude of the point at the th resonant frequency. is the wave velocity ratio of the point at the th resonant frequency. denotes concatenation of all elements into a column vector.

[0156] If resonant frequencies and types of eigenvalues are used (in this example = 3), each pixel point will be mapped to a -dimensional eigenvector. Before constructing the eigenvector, to eliminate the effects of different physical eigenvalues due to differences in dimension and numerical range, the fusion recognition and visualization module 340 will normalize each eigenvalue field map. One specific embodiment is the min-max normalization, which linearly scales all pixel values of each eigenvalue field to the interval. For any eigenvalue , its normalized value is calculated as follows:

[0157]

[0158] where and are the maximum and minimum values in the eigenvalue field, respectively.

[0159] ​​​By constructing high-dimensional feature vectors, the method of the present application effectively fuses originally isolated, single-dimensional information. The feature vector of each pixel point provides a comprehensive, multi-angle physical state description for it, and this rich information expression greatly enhances the distinguishability between normal regions and abnormal regions caused by leakage, laying a data foundation for subsequent high-precision, automated recognition using unsupervised learning algorithms.

[0160] After constructing high-dimensional feature vectors for each pixel point in the measured region, the fusion recognition and visualization module 340 uses an unsupervised learning anomaly detection algorithm to analyze these feature vectors to identify abnormal regions related to leakage. The fundamental reason for using unsupervised learning is that in actual detection scenarios, the patterns, sizes, and locations of leakage regions are unknown, and usually only occupy a small part of the entire detection region, which fully meets the typical characteristics of "few and unknown abnormal samples" in anomaly detection problems.

[0161] In this embodiment, the unsupervised anomaly detection algorithm is an isolation forest algorithm. The core idea of this algorithm is that abnormal points are more easily isolated in the feature space due to their "few and different" characteristics. Isolation forest achieves this idea by constructing an integrated model composed of multiple "isolation trees" (iTree).

[0162] The specific execution process of this algorithm is as follows: First, a subsample is randomly extracted from the set of high-dimensional feature vectors of all pixel points to construct an isolation tree. In the process of constructing this tree, the algorithm randomly selects a feature dimension from the data points contained in the current node at each node of the tree, and then randomly selects a split value between the maximum and minimum values of this dimension to divide the data points into two parts, which enter the left and right child nodes, respectively. This process is recursively executed until each data point is isolated in a leaf node or the depth of the tree reaches the preset upper limit.

[0163] Since the pixel points of the leakage region, especially the wave velocity ratio component, will be significantly higher than the dense data clusters composed of pixel points of normal, intact structure regions. Therefore, in the above random partitioning process, these abnormal feature vectors are likely to be separated from the main data group with fewer partitioning times. This is reflected in the structure of the isolation tree, that is, the leaf nodes corresponding to abnormal points are usually located near the root node, with a shorter path length.

[0164] The algorithm constructs multiple such isolation trees to form a "forest". For the feature vector of each pixel point to be evaluated, the algorithm calculates the average path length of it in all trees in the forest.

[0165] Finally, based on the average path length, the anomaly score of each pixel is calculated . The score is normalized to between 0 and 1, and the formula is:

[0166] ;

[0167] Where:

[0168] is the feature vector The average path length on all isolated trees in the forest.

[0169] is the number of sample points used to build the tree.

[0170] is a normalization factor representing the average path length of a binary search tree under the given number of samples , used to normalize the result. The formula is:

[0171] ;

[0172] Where is the harmonic series.

[0173] The value of the anomaly score has a clear indication:

[0174] When the average path length of a point is much smaller than , its score will tend to 1, indicating that the point is very likely to be an anomaly point;

[0175] When the average path length is close to , the score tends to 0.5;

[0176] When the average path length is much larger than , the score tends to 0, indicating that the point is very likely to be a normal point.

[0177] By performing the above calculation on all pixel points in the measured area, the fusion recognition and visualization module 340 finally generates a two-dimensional anomaly score map. In this map, the gray value or pseudo-color of each pixel directly corresponds to its anomaly score, and the area with higher score is the suspected leakage area.

[0178] The conversion process is achieved by threshold segmentation of the anomaly score map. The system sets an anomaly score threshold convert the score map to a binary mask image For each pixel in the score map its value in the binary mask image is determined according to the following rule:

[0179] ;

[0180] wherein, is the anomaly score of the pixel. The pixel with value 1 is preliminarily determined as belonging to the leakage region, and the pixel with value 0 is determined as normal region.

[0181] The threshold value can be determined in various ways. In one specific implementation, the threshold value can be pre-set by the operator according to experience, for example, 0.75, that is, the pixel with anomaly score higher than 0.75 is considered as a leakage point.

[0182] To realize the automation of the process, another preferred implementation is to use an adaptive threshold segmentation algorithm, such as Otsu method. This algorithm can automatically find a best threshold value in the histogram of the anomaly score map that can best distinguish the two classes of pixels "abnormal" and "normal", thereby avoiding the subjectivity of manual setting.

[0183] For the binary mask image obtained by the preliminary segmentation To eliminate the isolated misjudgment pixels caused by noise or fill the possible small holes in the internal region, a series of morphological image processing operations can be performed. The morphological operation can include an opening operation to remove small noise points, followed by a closing operation to fill the holes in the internal region, so that the boundary of the finally recognized leakage region is smoother and the region is more complete.

[0184] After the above processing, the system obtains a final mask image accurately representing the contour of the leakage region. The last step is to fuse the recognition result with an original optical reference image (for example, the first frame of static image collected by the high-frame-rate optical imaging unit 200) to realize the visual presentation of the result. One specific presentation method is to superimpose the region marked as leakage in the final mask image on the original optical image in a prominent, semi-transparent pseudo-color (for example, red). Alternatively, the contour line of the leakage region can also be extracted and drawn on the original image in a highlighted color.

[0185] The final visual result is presented on the user interface of the information processing and control unit 300, providing the user with intuitive and clear indications about the location, shape and range of the leakage.

[0186] To further illustrate the practical application effect of the present application, the following will be described in detail in combination with a leakage detection project of an outer wall of a high-rise residential building in a certain city.

[0187] 1. Engineering background and detection object: The measured object is a region at the 5th floor of the east facade of the residential building, with an area of about 2m x 2m. The wall structure is a reinforced concrete shear wall, and the outer facing is glazed tile. There is a moisture return phenomenon at the corresponding indoor position of the region, and the outer wall leakage point needs to be accurately positioned.

[0188] 2. System layout:

[0189] Excitation unit arrangement: The composite orthogonal dual-mode excitation unit 100 is installed at the edge of the window sill below the measured region, and is fixed to the wall base through high-viscosity vacuum grease as a coupling agent, to ensure that the normal and tangential vibrations can be effectively transmitted.

[0190] Imaging unit arrangement: The high-frame-rate optical imaging unit 200 is erected on the ground lifting platform about 8 meters away from the wall surface. An industrial lens with a focal length of 50mm is selected, and the field of view completely covers the 2m x 2m region of interest.

[0191] Environmental assistance: Since the detection time is in the afternoon, the natural light is sufficient, and no auxiliary light source is used. The camera and the physical plane are calibrated using a checkerboard calibration board to obtain the homography matrix.

[0192] 3. Detection process:

[0193] Frequency identification: The system first transmits a linear frequency modulation signal from 100Hz to 800Hz. After fast Fourier transform analysis, it is found that there is a significant local resonance frequency in the facing layer of the region .

[0194] Data acquisition: The system automatically locks at 420Hz, first controls the normal excitation component to work for 0.5 seconds, and then switches to the tangential excitation component to work for 0.5 seconds. The camera synchronously collects two groups of video sequences at a rate of 2000 frames / second.

[0195] Characteristic solution: The information processing and control unit 300 processes the video data using orthogonal demodulation algorithm. It is found that near the left upper corner coordinate (x=0.5m, y=1.8m) of the region, the S wave velocity field shows a significant low value area (about 40% reduction), while the P wave velocity field changes little.

[0196] Wave velocity ratio calculation: The wave velocity ratio field generated by the system shows that the wave velocity ratio value of the left upper corner region jumps from 1.6 of the background value to 2.1, which strongly indicates that there is moisture filling pores at this position.

[0197] 4. Identification result and verification:

[0198] Smart decision: The anomaly score map output by the Isolation Forest algorithm highlights the upper-left region (score > 0.85). The red semi-transparent mask generated by the system accurately covers this irregular area.

[0199] Engineering verification: Based on the visualization results output by the system, the engineering personnel conducted a brick-by-brick inspection of the marked area and found that the adhesive mortar layer behind the tiles was completely saturated with water and there were obvious hollow water channels, which was highly consistent with the detection conclusion of the system.

[0200] Differentiation of micro-crack leakage and dry crack in basement concrete shear wall:

[0201] This embodiment aims to demonstrate the ability of the present application to distinguish between "leakage (water)" and "dry crack (gas)".

[0202] 1. Engineering background and detection object: The measured object is a concrete shear wall in the underground second floor parking lot of a commercial center, which has visible fine cracks on the surface, and it is necessary to determine whether the cracks contain underground water infiltration.

[0203] 2. System layout: Since the basement is dark, a high-frequency non-flickering LED surface light source is installed beside the imaging unit for supplementary lighting. The excitation unit is directly attached to the top of the concrete wall.

[0204] 3. Detection process:

[0205] Frequency identification: The frequency range is set to 500Hz to 2000Hz (the stiffness of the concrete structure is large, and the resonant frequency is usually high). The system automatically identifies the main resonant frequency as 1150Hz.

[0206] Dual-mode imaging: At a frequency of 1150Hz, micro-motion videos under normal and tangential excitation are collected respectively.

[0207] Physical field analysis:

[0208] Region A (upper section of the crack): The characteristic solution shows that the P-wave amplitude and S-wave amplitude both show significant attenuation (as the crack blocks the propagation of the wave), and the wave velocity ratio remains in the normal range of concrete (about 1.5). This indicates that this is a dry medium (air-filled crack).

[0209] Region B (lower section of the crack): The characteristic solution shows that the S-wave velocity drops sharply, but the P-wave can still propagate well due to the incompressibility of water (with less attenuation), resulting in an abnormally high wave velocity ratio above 1.9.

[0210] 4. Identification result and verification: the system fusion identification module only marks "region B" as a leakage abnormal area according to the wave velocity ratio feature, and excludes "region A" (judged as a common structure damage).

[0211] Verification: field drilling sampling verification, the powder drilled in region A is dry, and a small amount of water seeps out after drilling in region B. It is proved that the present application can effectively exclude the interference of dry cracks by using the orthogonal dual-mode wave velocity ratio field, and realize accurate identification of true and false leakage.

Claims

1. A method for detecting, identifying, and locating building leakage points in construction engineering, characterized in that, Includes the following steps: S1. System setup and parameter initialization: Deploy a high frame rate optical imaging unit to cover the building structure under test with its field of view, and couple a composite orthogonal dual-mode excitation unit to the building structure under test. S2, Adaptive Resonance Frequency Identification: Identifying the set of resonance frequencies of the building structure under test; S3. Orthogonal dual-mode resonance data acquisition: Based on the set of resonant frequencies, normal excitation and tangential excitation are applied to the building structure under test in sequence, and the high frame rate optical imaging unit is used to synchronously acquire surface micro-motion video data of the building structure under test under the normal excitation and tangential excitation. S4. Solve the multidimensional micro-motion feature field: Solve the surface micro-motion video data to generate P-wave velocity field and S-wave velocity field corresponding to the normal excitation and tangential excitation respectively, and calculate the wave velocity ratio field based on the P-wave velocity field and S-wave velocity field. S5. Feature Fusion and Leakage Area Identification: Based on the wave velocity ratio field, a high-dimensional feature vector representing the physical state of each pixel is constructed, and the anomaly score of each pixel is calculated through an unsupervised learning anomaly detection algorithm to identify and locate the leakage point.

2. The method for detecting, identifying, and locating building leakage points in construction engineering according to claim 1, characterized in that, S2 specifically includes: A broadband sweep frequency excitation is applied to the building structure under test and its dynamic response is collected. The structure transfer function is constructed by performing Fourier transform and spatial accumulation on the dynamic response, and a peak search algorithm is applied to the structure transfer function to identify the set of resonant frequencies.

3. The method for detecting, identifying, and locating building leakage points in construction engineering according to claim 1, characterized in that, The generation of the P-wave velocity field and S-wave velocity field in S4 specifically includes: The P-wave phase field and S-wave phase field are calculated from the surface micro-motion video data. The P-wave wavenumber field and S-wave wavenumber field are obtained by calculating the spatial gradient of the P-wave phase field and S-wave phase field. The P-wave velocity field and S-wave velocity field are calculated according to the relationship between wave velocity and wave number.

4. The method for detecting, identifying, and locating building leakage points in construction engineering according to claim 3, characterized in that, The P-wave phase field and S-wave phase field are obtained by orthogonal demodulation of the time-domain displacement signal extracted from the surface micro-motion video data.

5. The method for detecting, identifying, and locating building leakage points in construction engineering according to claim 1, characterized in that, The construction of the high-dimensional feature vector in S5 specifically includes: For each pixel within the test area, the wave velocity ratio field characteristic value, P-wave amplitude field characteristic value, and S-wave amplitude field characteristic value calculated at each frequency of the set of resonant frequencies are concatenated to construct the high-dimensional feature vector.

6. The method for detecting, identifying, and locating building leakage points in construction engineering according to claim 1, characterized in that, The unsupervised learning anomaly detection algorithm is the Isolation Forest algorithm.

7. The method for detecting, identifying, and locating building leakage points in construction engineering according to claim 1, characterized in that, The identification and location of leak points in S5 specifically includes: An anomaly score map is generated based on the anomaly scores of each pixel, and adaptive threshold segmentation is performed on the anomaly score map to generate a binary mask image that identifies the leakage point.

8. The method for detecting, identifying, and locating building leakage points in construction engineering according to claim 7, characterized in that, After generating the binary mask image, morphological processing operations are also performed on it to eliminate noise and optimize the integrity of the leakage area.

9. A method for detecting, identifying, and locating building leakage points in construction engineering according to claim 1, characterized in that, After identifying and locating the leak points, the process also includes a step of visualizing the identified leak point areas by overlaying them in pseudo-color onto the original optical image.

10. A building leakage point detection, identification, and location system for building engineering, applied to the building leakage point detection, identification, and location method for building engineering as described in any one of claims 1-9, characterized in that, include: A composite orthogonal bimodal excitation unit; A high frame rate optical imaging unit; And an information processing and control unit.

Images