CN-121558748-B - Building exterior wall defect detection method and system based on acousto-optic-thermal multi-physical field coupling

CN121558748BCN 121558748 BCN121558748 BCN 121558748BCN-121558748-B

Abstract

The invention provides a building outer wall defect detection method and system based on acousto-optic-thermal multi-physical field coupling, and relates to the technical field of intelligent nondestructive detection. The method comprises the steps of screening defective candidate areas from a plurality of areas of a building exterior wall image according to comparison results between optical indexes and index conditions of the areas in the building exterior wall image, obtaining a temperature change sequence of each pixel point acquired during acoustic excitation and thermal excitation applied to the exterior wall area corresponding to the candidate areas, carrying out time-frequency conversion and phase-locking analysis on the temperature change sequence of each pixel point to obtain thermoacoustic coupling parameters and resonance intensity parameters of each pixel point, determining defect types of each pixel point, and obtaining the empty depth and the bonding stiffness according to formant characteristic inversion of the plurality of pixel points according to preset calibration relations between formant characteristics and empty depth and bonding stiffness of the plurality of pixel points, wherein the defect types of the pixel points are empty.

Inventors

DIAO YU
HUANG JIATAO
HUANG JIANYOU
TANG SHENGMING
PANG JIEWEN
LIU CHENGJUN
XU YONGXIN

Assignees

天津大学

Dates

Publication Date: 20260512
Application Date: 20260122

Claims (9)

1. The method for detecting the defects of the outer wall of the building based on the acousto-optic-thermal multi-physical field coupling is characterized by comprising the following steps of: Screening defective candidate areas from a plurality of areas of the building exterior wall image according to a comparison result between optical indexes and index conditions of each area in the building exterior wall image, wherein the building exterior wall image is acquired by an optical acquisition module; Acquiring a temperature change sequence of each pixel point acquired during the period of applying acoustic excitation and thermal excitation to the outer wall area corresponding to the candidate area; Performing time-frequency conversion and phase-locking analysis on the temperature change sequence of each pixel point to obtain a thermoacoustic coupling parameter and a resonance intensity parameter of each pixel point; Determining the defect type of each pixel point according to the thermoacoustic coupling parameter, the resonance intensity parameter and the geometric reference parameter of each pixel point determined based on the building exterior wall image; Aiming at a plurality of pixel points with the defect type of empty drum in the candidate region, obtaining the empty drum depth and the bonding rigidity according to the inversion of the formant characteristics of the pixel points based on the preset calibration relation between the formant characteristics and the empty drum depth and the bonding rigidity respectively; Performing time-frequency conversion and phase-locking analysis on the temperature change sequence of each pixel point to obtain thermoacoustic coupling parameters and resonance intensity parameters of each pixel point comprises the following steps: performing discrete Fourier transform on the temperature change sequence of each pixel point to obtain a frequency domain representation of the temperature change sequence; performing digital phase locking operation on the temperature change sequence of each pixel point to obtain a temperature spectrum amplitude of each pixel point; determining a first noise power of a dominant frequency symmetric neighborhood according to the symmetric neighborhood half-width of the acoustic excitation and the total power of the acoustic excitation frequency band determined based on the frequency domain representation; determining the thermoacoustic coupling parameter using the first noise power and the temperature spectrum amplitude, and And determining the temperature spectrum amplitude corresponding to the formant frequency in the process of applying the acoustic excitation according to a preset frequency interval as the resonance intensity parameter.
2. The method of claim 1, wherein said determining said thermo-acoustic coupling parameter using said first noise power and said temperature spectrum magnitude comprises: A ratio of the temperature spectrum amplitude to a square root of the first noise power is determined as the thermo-acoustic coupling parameter.
3. The method of claim 1, wherein the determining the defect type to which each pixel belongs from the thermo-acoustic coupling parameter, the resonance intensity parameter, and the geometric reference parameter of each pixel determined based on the building exterior wall image comprises: For each of the pixel points detected by signal reliability: determining respective corresponding self-adaptive weights according to respective signal-to-noise ratios of the thermoacoustic coupling parameter, the resonance intensity parameter and the geometric reference parameter; Weighting and summing the normalized thermoacoustic coupling parameter, the normalized resonance intensity parameter and the normalized geometric reference parameter with the self-adaptive weights corresponding to the normalized thermoacoustic coupling parameter and the normalized geometric reference parameter to obtain a defect comprehensive confidence coefficient, and And determining the defect type of each pixel point according to a first comparison relation, a second comparison relation, a third comparison relation and/or a fourth comparison relation, wherein the first comparison relation, the second comparison relation, the third comparison relation and the fourth comparison relation are respectively comparison relations among the defect comprehensive confidence coefficient, the thermoacoustic coupling parameter, the resonance intensity parameter, the geometric reference parameter and the respective corresponding threshold value.
4. A method according to claim 3, wherein said determining respective corresponding adaptive weights from respective signal-to-noise ratios of said thermo-acoustic coupling parameter, said resonance intensity parameter and said geometric reference parameter comprises: Summing the signal to noise ratios of the thermoacoustic coupling parameter, the resonance intensity parameter and the geometric reference parameter to obtain a comprehensive signal to noise ratio; And determining the ratio of the signal to noise ratio of each of the thermoacoustic coupling parameter, the resonance intensity parameter and the geometric reference parameter to the comprehensive signal to noise ratio as the self-adaptive weight corresponding to each of the thermoacoustic coupling parameter, the resonance intensity parameter and the geometric reference parameter.
5. The method of claim 4, wherein the signal-to-noise ratio of the thermo-acoustic coupling parameter is a ratio of a square of the temperature spectrum magnitude to the first noise power; The signal to noise ratio of the resonance intensity parameter is the ratio of the square of the temperature spectrum amplitude to the second noise power of the formant frequency in the symmetrical neighborhood, wherein the second noise power is the ratio of the total power of the acoustic excitation frequency band to the total number of frequencies in the preset frequency interval; The signal to noise ratio of the geometric reference parameter is the ratio of the geometric reference parameter to the geometric reference parameter distribution of the same-material non-defective outer wall.
6. A method according to claim 3, wherein determining the defect type to which each pixel belongs according to the first comparison relation, the second comparison relation, the third comparison relation and/or the fourth comparison relation comprises: For each of the pixel points detected by signal reliability: Determining that the defect type of the pixel point is empty under the conditions that the defect comprehensive confidence coefficient is larger than or equal to a defect judging threshold value, the thermo-acoustic coupling parameter is larger than or equal to a thermo-acoustic coupling threshold value and the resonance intensity parameter is larger than or equal to a resonance intensity threshold value; And determining the defect type of the pixel point as a surface crack under the conditions that the geometric reference parameter is larger than or equal to a geometric reference threshold, the thermo-acoustic coupling parameter is smaller than the thermo-acoustic coupling threshold and the resonance intensity parameter is smaller than the resonance intensity threshold.
7. The method according to claim 1, wherein the inverting the formant characteristics of the pixel points to obtain the empty depth and the bonding stiffness based on the preset calibration relation between the formant characteristics and the empty depth and the bonding stiffness respectively for the pixel points with the empty defect types in the candidate region comprises: determining at least one empty region according to a plurality of pixel points with the defect type being empty based on a connected domain algorithm; Determining an average value of formant quality parameters of each empty area according to respective formant quality parameters of a plurality of pixel points in each empty area, wherein the formant quality parameters are determined according to formant characteristics of formant frequencies of each pixel point and half-power bandwidths of formants; screening the empty areas to be inverted from at least one empty area according to the average value of the formant quality parameters of each empty area and the area of the empty area; And aiming at the blank area to be inverted, inverting to obtain the blank depth and the bonding rigidity of each blank area to be inverted according to the formant characteristics of each pixel point of the blank area to be inverted based on the preset calibration relation between the formant characteristics and the blank depth and the bonding rigidity respectively.
8. The method of claim 7, wherein the method further comprises: Generating a visualized defect type image, a hollowing depth map aiming at the hollowing area to be inverted and/or, And generating a structured detection report according to intermediate calculation parameters of each pixel point, wherein the intermediate calculation parameters comprise at least one of the bonding rigidity, the formant quality parameters and the thermo-acoustic coupling parameters.
9. An architectural exterior wall defect detection system based on acousto-optic thermal multi-physical field coupling, the system comprising: The unmanned aerial vehicle platform is configured to fly to a position which is separated from an outer wall of a building to be detected by a preset distance in response to a control instruction, and comprises an optical acquisition module, an acoustic excitation module, a thermal excitation module, an infrared imaging module and an edge calculation module; the optical acquisition module is configured to project structural light to the building exterior wall and acquire a building exterior wall image after the structural light is projected; The acoustic excitation module is configured to apply acoustic excitation to an outer wall region corresponding to a candidate region in the building outer wall image, wherein the candidate region is obtained by screening a plurality of regions in the building outer wall image according to a comparison result between optical indexes and index conditions of each region in the building outer wall image by the edge calculation module; the thermal stimulation module is configured to apply thermal stimulation to the exterior wall region; the infrared imaging module is configured to acquire a temperature change sequence of each pixel point in the candidate region during the process of applying acoustic excitation and thermal excitation to the outer wall region corresponding to the candidate region; The edge calculation module is configured to perform time-frequency conversion and phase-locked analysis on the temperature change sequence of each pixel point to obtain a thermoacoustic coupling parameter and a resonance intensity parameter of each pixel point, determine a defect type of each pixel point according to the thermoacoustic coupling parameter of each pixel point, the resonance intensity parameter and a geometric reference parameter of each pixel point determined based on the building exterior wall image, wherein the defect type comprises a blank; the thermoacoustic coupling parameter and the resonance intensity parameter are determined by: performing discrete Fourier transform on the temperature change sequence of each pixel point to obtain a frequency domain representation of the temperature change sequence; performing digital phase locking operation on the temperature change sequence of each pixel point to obtain a temperature spectrum amplitude of each pixel point; determining a first noise power of a dominant frequency symmetric neighborhood according to the symmetric neighborhood half-width of the acoustic excitation and the total power of the acoustic excitation frequency band determined based on the frequency domain representation; Determining the thermo-acoustic coupling parameter using the first noise power and the temperature spectrum amplitude; And determining the temperature spectrum amplitude corresponding to the formant frequency in the process of applying the acoustic excitation according to a preset frequency interval as the resonance intensity parameter.

Description

Building exterior wall defect detection method and system based on acousto-optic-thermal multi-physical field coupling Technical Field The invention relates to the technical field of intelligent nondestructive testing, in particular to a building exterior wall defect detection method and system based on acousto-optic-thermal multi-physical field coupling. Background Building exterior wall defects are key factors affecting building safety and service life. Currently, building exterior wall defect detection techniques include relying on infrared thermal imaging and acoustic excitation. The infrared thermal imaging technology indirectly captures the temperature difference abnormal characteristics of the internal structure by receiving the thermal radiation signals of the wall body, so that the identification and judgment of potential defects are realized. However, on one hand, passive infrared imaging relies on natural temperature differences only, detection effect is extremely poor at night or in low-temperature environments, and a system is extremely easy to misjudge background hot spots as empty and drum defect outer walls. On the other hand, the surface temperature is easy to be interfered by factors such as solar radiation, shadow, wind speed, humidity and the like, and particularly, under an unstable thermal environment, a large number of background hot spots are easy to appear on an acquired thermal image, so that the temperature difference contrast between a defect area and a normal area is reduced, and the accuracy is influenced. Furthermore, the effective depth of thermal imaging detection is typically only a few millimeters to 1 centimeter of the surface layer, and the thermal response signal is significantly reduced when the defect is located deeper below the finish or mortar layer. The acoustic wave excitation technique analyzes the reflected information to determine if a defect exists by applying excitation, such as a tap or acoustic wave emission, to the wall surface. The method can detect the empty drum below the facing layer in theory, but the traditional manual knocking depends on manual experience judgment, and has low detection speed and strong subjectivity. The effective signal can be obtained by tightly attaching the acoustic wave sensor to the wall surface, and is easily affected by background noise, wind noise, equipment vibration, material density, thickness, surface roughness and the like. In summary, the single-mode detection scheme is difficult to meet the requirements of accurate identification, depth quantification and the like of the defects of the outer wall, so that a multi-mode fusion detection technical scheme is needed to be constructed, and the technical bottleneck of the single mode is broken through the cooperative utilization of information. Disclosure of Invention In view of the above, the invention provides a method and a system for detecting defects of an outer wall of a building based on acousto-optic-thermal multi-physical field coupling. The invention provides a building exterior wall defect detection method based on acousto-optic-thermal multi-physical field coupling, which comprises the steps of screening defective candidate areas from a plurality of areas of a building exterior wall image according to comparison results between optical indexes and index conditions of the areas in the building exterior wall image, acquiring temperature change sequences of each pixel point acquired during acoustic excitation and thermal excitation applied to the exterior wall area corresponding to the candidate areas, performing time-frequency conversion and phase-locked analysis on the temperature change sequences of each pixel point to obtain thermoacoustic coupling parameters and resonance intensity parameters of each pixel point, determining defect types of each pixel point according to the thermoacoustic coupling parameters and the resonance intensity parameters of each pixel point and geometric reference parameters of each pixel point determined based on the building exterior wall image, and obtaining the drum depth and the bonding stiffness according to the peak characteristic inversion of each pixel point according to the preset relation between the resonance peak characteristic and the drum depth and the bonding stiffness of each pixel point in the candidate areas. According to the embodiment of the invention, time-frequency conversion and phase-locking analysis are carried out on the temperature change sequence of each pixel point to obtain the thermoacoustic coupling parameter and the resonance intensity parameter of each pixel point, wherein the discrete Fourier transformation is carried out on the temperature change sequence of each pixel point to obtain the frequency domain representation of the temperature change sequence, the digital phase-locking operation is carried out on the temperature change sequence of each pixel point to obtain the temperature spectrum ampl