CN-121994817-A - Building outer wall defect monitoring method, system, device and electronic equipment

CN121994817ACN 121994817 ACN121994817 ACN 121994817ACN-121994817-A

Abstract

The invention relates to the technical field of building engineering detection and discloses a method, a system, a device and electronic equipment for monitoring defects of an outer wall of a building, wherein the method comprises the steps of acquiring multi-source sensing data of each area to be detected of the outer wall of the building in a non-contact automatic inspection mode; the method comprises the steps of carrying out environment correction and image fusion on multisource perception data to generate a correction fusion image, adopting a target detection model to automatically identify and classify the defects of the outer wall based on the correction fusion image to obtain a detection result, carrying out time sequence analysis on the defects of the outer wall detected by the same detection area for the past time to obtain the development trend of the defects of the outer wall, evaluating the risk level of the defects of the outer wall based on the development trend and the detection result, and carrying out early warning notification operation based on the risk level.

Inventors

YANG WEIJUN
LI XISONG
CHEN MUHE
WANG SHAO
YI XIAOMING
KANG RONG
ZHANG YUHUA
GU BAOCHENG

Assignees

广西壮族自治区建筑工程质量检测中心有限公司
湖南建研信息技术股份有限公司

Dates

Publication Date: 20260508
Application Date: 20260302

Claims (10)

1. A method for monitoring defects of an exterior wall of a building, the method comprising: Acquiring multisource sensing data of each area to be detected of the building outer wall in a non-contact automatic inspection mode; Performing environment correction and image fusion on the multi-source perception data to generate a correction fusion image; Based on the corrected fusion image, automatically identifying and classifying the defects of the outer wall by adopting a target detection model to obtain a detection result; performing time sequence analysis on the defects of the outer wall detected by the same detection area for the past time to acquire the development trend of the defects of the outer wall; And evaluating the risk level of the outer wall defect based on the development trend and the detection result, and carrying out early warning notification operation based on the risk level.
2. The method according to claim 1, wherein the method further comprises: Acquiring the structural characteristics and defect distribution priori knowledge of a building to be detected; based on the structural characteristics of the building to be detected and the prior knowledge of defect distribution, dividing the outer wall of the building into a plurality of areas to be detected with different detection priorities, wherein the areas to be detected comprise key detection areas and common detection areas.
3. The method of claim 1, wherein the multi-source perception data comprises an infrared thermal imaging image, a visible light image, an environmental parameter, and laser ranging data, and wherein performing environmental correction and image fusion on the multi-source perception data to generate a corrected fused image comprises: performing format standardization, invalid data rejection and missing data interpolation complement preprocessing on the multi-source perception data; Based on the environmental parameters and the laser ranging data, calling a preset environmental correction model to carry out environmental correction on the infrared thermal imaging image; and carrying out image fusion on the infrared thermal imaging image subjected to the environment correction and the corresponding visible light image to generate a corrected fusion image.
4. A method according to claim 3, wherein image blending the environmentally corrected infrared thermographic image with the corresponding visible light image to generate a corrected blended image comprises: And carrying out feature level fusion on the infrared thermal imaging image after the environmental correction and the corresponding visible light image by adopting a double-branch fusion algorithm based on an attention mechanism, and generating a correction fusion image which simultaneously retains temperature anomaly information and texture details.
5. The method of claim 1, wherein the target detection model is a trained YOLO-series target detection model comprising a multi-scale adaptive fusion network, a dual-path fusion pyramid network, and a context-aware fusion module; Based on the corrected fusion image, automatically identifying and classifying the defects of the outer wall by adopting a target detection model to obtain a detection result, wherein the method comprises the following steps: Performing size standardization and normalization preprocessing on the correction fusion image; And inputting the preprocessed images into a trained improved YOLO series target detection model, and respectively passing through a multi-scale self-adaptive fusion network, a dual-path fusion pyramid network and a context sensing fusion module to generate detection results comprising defect types, defect positions, defect areas and defect severity.
6. The method of claim 1, wherein the performing a time sequence analysis on the defects of the exterior wall detected by the same detection area for a long time to obtain the development trend of the defects of the exterior wall comprises: Taking the latest detected visible light image as a reference, and performing cross-period pixel level registration on the latest detected visible light image and the historical region image by utilizing a feature matching algorithm based on deep learning; Calculating time sequence change parameters for the same defect instance identified in the past detection based on the registration result, wherein the time sequence change parameters comprise defect area, defect position, contour shape and characteristic intensity; Based on the time sequence variation parameters, constructing a multi-parameter time sequence of each outer wall defect, and fitting the evolution trend of each outer wall defect by adopting a time sequence analysis model; And predicting the state change of the external wall defect in a future specified period according to the evolution trend, and generating an external wall defect development situation for risk assessment.
7. The method according to claim 1, wherein evaluating the risk level of the exterior wall defect based on the development trend and the detection result, and performing the early warning notification operation based on the risk level, comprises: acquiring environmental risk factors of a detection area where the outer wall defect is located; inputting the detection result and the environmental risk factors into a preset multi-level risk assessment system to generate a risk level of the current external wall defect, wherein the risk level comprises low risk, medium risk and high risk; And when the risk level reaches or exceeds a preset corresponding early warning trigger threshold value, automatically executing early warning notification operation corresponding to the corresponding risk level.
8. A building exterior wall defect monitoring system, the system comprising: The unmanned aerial vehicle flight platform is provided with a multi-sensor detection module and is used for acquiring multi-source sensing data of each area to be detected of the outer wall of the building in a non-contact automatic inspection mode; The ground processing subsystem is communicated with the multi-sensor detection module through the data transmission module, and is used for carrying out environment correction and image fusion on multi-source perception data to generate a correction fusion image, automatically identifying and classifying the defects of the outer wall by adopting a target detection model based on the correction fusion image to obtain a detection result, carrying out time sequence analysis on the defects of the outer wall detected by the same detection area for the past time to obtain the development trend of the defects of the outer wall, evaluating the risk level of the defects of the outer wall based on the development trend and the detection result, and carrying out early warning notification operation based on the risk level.
9. A building exterior wall defect monitoring device, the device comprising: the image acquisition module is used for acquiring multi-source perception data of each area to be detected of the outer wall of the building in a non-contact automatic inspection mode; The correction fusion module is used for carrying out environment correction and image fusion on the multi-source perception data to generate a correction fusion image; The target detection module is used for automatically identifying and classifying the defects of the outer wall by adopting a target detection model based on the corrected fusion image to obtain a detection result; The development trend analysis module is used for carrying out time sequence analysis on the outer wall defects detected in the same detection area for the past time to acquire the development trend of the outer wall defects; And the evaluation early warning module is used for evaluating the risk level of the outer wall defect based on the development trend and the detection result and carrying out early warning notification operation based on the risk level.
10. An electronic device, comprising: A memory and a processor, the memory and the processor being communicatively connected to each other, the memory having stored therein computer instructions, the processor executing the computer instructions to perform the method of building exterior wall defect monitoring of any one of claims 1 to 7.

Description

Building outer wall defect monitoring method, system, device and electronic equipment Technical Field The invention relates to the technical field of building engineering detection, in particular to a method, a system and a device for monitoring defects of an outer wall of a building and electronic equipment. Background Along with the prolongation of the service life of the building and the long-term erosion of natural environments such as blowing, sun, rain, temperature difference circulation and the like, structural defects such as hollowness, cracking, leakage and the like are easy to occur in an external wall heat insulation system, and the phenomena of hollowness and even falling off of an external wall finish layer (such as face bricks, paint, stone and the like) are also often caused. Such defects not only affect the appearance and durability of the building, but also are more likely to cause high-altitude falling of the components, thus constituting a potential threat to public safety. Currently, external wall defect detection mainly depends on traditional methods (such as visual inspection, knocking and bridging contact detection) and novel technologies (such as manual hand-held infrared thermal imaging). The traditional method has the problems of low efficiency, experience dependence on precision, high risk of high-altitude operation, difficult coverage of high-rise and special-shaped structures and the like. While the infrared thermal imaging technology can identify internal defects, the manual operation mode of the infrared thermal imaging technology leads to limited detection range, unstable efficiency, difficult long-term dynamic monitoring, and the result is easily interfered by environmental factors, so that the infrared thermal imaging technology cannot meet the requirements of large-scale, long-acting and high-precision monitoring and early warning. Disclosure of Invention The invention provides a method, a system, a device and electronic equipment for monitoring defects of an outer wall of a building, and aims to solve the problems that in the prior art, the precision is poor, the safety is insufficient, and long-acting dynamic monitoring and detection are difficult to realize in a building outer wall heat insulation system and in a hollowing drop monitoring mode. In a first aspect, the present invention provides a method for monitoring defects of an exterior wall of a building, the method comprising: Acquiring multisource sensing data of each area to be detected of the building outer wall in a non-contact automatic inspection mode; performing environment correction and image fusion on the multi-source perception data to generate a correction fusion image; based on the corrected fusion image, automatically identifying and classifying the defects of the outer wall by adopting a target detection model to obtain a detection result; performing time sequence analysis on the defects of the outer wall detected by the same detection area for the past time to acquire the development trend of the defects of the outer wall; And evaluating the risk level of the outer wall defect based on the development trend and the detection result, and carrying out early warning notification operation based on the risk level. According to the method for monitoring the defects of the building outer wall, provided by the invention, the accuracy and the anti-interference capability of defect identification are improved through multi-source sensing data, environment correction and image fusion are carried out to generate a correction fusion image, defects are automatically identified based on the image, time sequence analysis is carried out on the defects in the past to obtain a development trend, and finally risk grade and early warning are estimated based on the trend and the result, so that the full-flow closed-loop monitoring from data acquisition to risk early warning is realized. In an alternative embodiment, the method further comprises: Acquiring the structural characteristics and defect distribution priori knowledge of a building to be detected; based on the structural characteristics of the building to be detected and the prior knowledge of defect distribution, dividing the outer wall of the building into a plurality of areas to be detected with different detection priorities, wherein the areas to be detected comprise key detection areas and common detection areas. The method for monitoring the defects of the building outer wall is based on the self structural characteristics of the building to be detected and the prior knowledge of the historical defect distribution, and scientifically divides the building outer wall into detection areas with different priorities, such as emphasis, common and the like, according to the knowledge. The process enables the subsequent monitoring resources to be optimally configured, can be focused on the high-altitude areas of window angles and joints in a targeted manner, and