Intelligent safety monitoring method and system for glass curtain wall, electronic equipment and storage medium

Abstract

The invention discloses an intelligent safety monitoring method and system for a glass curtain wall, electronic equipment and a storage medium, and relates to the technical field of curtain wall detection. The method comprises the following steps: acquiring multi-source image data of an external facade of the glass curtain wall, wherein the multi-source image data comprises a front view image, an infrared thermal imaging image and an ultraviolet spectrum image of the glass curtain wall, which are acquired at different time periods and different angles; performing registration fusion and identification analysis on the multi-source image data to obtain crack characteristics, deformation characteristics and temperature distribution characteristics of the surface of the glass curtain wall; establishing an abnormal characteristic spectrum of the glass curtain wall based on the crack characteristics, the deformation characteristics and the temperature distribution characteristics; performing spatio-temporal evolution analysis on the abnormal characteristic spectrum to determine a multi-level risk area of the glass curtain wall; and generating a monitoring report of the glass curtain wall according to the multi-level risk area. By implementing the technical scheme provided by the invention, comprehensive monitoring of the glass curtain wall can be realized, so that the monitoring accuracy of the glass curtain wall is improved.

Description

Technical Field

[0001] The present application relates to the field of curtain wall detection technology, and in particular to a glass curtain wall intelligent safety monitoring method, system, electronic equipment and storage medium. Background Art

[0002] As modern architecture develops towards higher-rises and larger buildings, glass curtain walls are widely used due to their aesthetics, transparency, and energy efficiency. However, due to their long-term exposure to outdoor environments, glass curtain walls are susceptible to various factors, including temperature fluctuations, wind loads, and building settlement, which can easily lead to safety hazards.

[0003] Currently, glass curtain wall safety monitoring primarily relies on fixed monitoring equipment. Strain sensors, displacement sensors, and other sensors are installed on the glass curtain wall surface to monitor its stress and deformation in real time. However, this monitoring method has limitations: the installation of fixed monitoring equipment can affect the building's aesthetics, and the limited number of monitoring points makes comprehensive monitoring of the glass curtain wall difficult, resulting in low monitoring accuracy. Summary of the Invention

[0004] The present application provides a glass curtain wall intelligent safety monitoring method, system, electronic equipment and storage medium, which can realize comprehensive monitoring of the glass curtain wall, thereby improving the monitoring accuracy of the glass curtain wall.

[0005] In a first aspect, the present application provides a glass curtain wall intelligent safety monitoring method, the method comprising: Acquire multi-source image data of the glass curtain wall facade, wherein the multi-source image data includes front view images, infrared thermal imaging images, and ultraviolet spectrum images of the glass curtain wall collected at different time periods and different angles; Performing registration, fusion, and recognition analysis on the multi-source image data to obtain crack characteristics, deformation characteristics, and temperature distribution characteristics on the surface of the glass curtain wall; Establishing an abnormal characteristic map of the glass curtain wall based on the crack characteristics, deformation characteristics and temperature distribution characteristics; Performing a spatiotemporal evolution analysis on the abnormal characteristic map to determine the multi-level risk areas of the glass curtain wall; A monitoring report of the glass curtain wall is generated according to the multi-level risk areas.

[0006] By adopting the above technical solution, multi-source image data of the glass curtain wall facade is obtained. The multi-source image data includes front view images, infrared thermal imaging images and ultraviolet spectrum images collected at different time periods and angles, so that the surface state information of the glass curtain wall can be comprehensively collected; the multi-source image data is aligned, fused and identified and analyzed, so that the crack characteristics, deformation characteristics and temperature distribution characteristics of the glass curtain wall surface can be accurately extracted; then, based on these characteristics, an abnormal feature map is established, and multi-level risk areas are determined through spatiotemporal evolution analysis, and finally a monitoring report is generated, thereby realizing comprehensive monitoring of the glass curtain wall. There is no need to install fixed monitoring equipment on the surface of the glass curtain wall, which not only avoids the impact on the aesthetics of the building, but also overcomes the problem of limited monitoring points, thereby realizing comprehensive and accurate monitoring of the glass curtain wall.

[0007] In a second aspect of the present application, a glass curtain wall intelligent safety monitoring system is provided, the system comprising: An image data acquisition module is used to acquire multi-source image data of the glass curtain wall facade, wherein the multi-source image data includes front view images, infrared thermal imaging images, and ultraviolet spectrum images of the glass curtain wall collected at different time periods and different angles; A feature recognition module is used to perform registration, fusion, and recognition analysis on the multi-source image data to obtain crack features, deformation features, and temperature distribution features on the surface of the glass curtain wall; A feature map generation module, configured to establish an abnormal feature map of the glass curtain wall based on the crack features, deformation features, and temperature distribution features; a risk area determination module, configured to perform spatiotemporal evolution analysis on the abnormal feature map to determine the multi-level risk areas of the glass curtain wall; A monitoring report generating module is used to generate a monitoring report of the glass curtain wall according to the multi-level risk areas.

[0008] In a third aspect of the present application, a computer storage medium is provided. The computer storage medium stores a plurality of instructions, and the instructions are suitable for being loaded by a processor and executing the above method steps.

[0009] In a fourth aspect of the present application, an electronic device is provided, comprising: a processor and a memory; wherein the memory stores a computer program, and the computer program is suitable for being loaded by the processor and executing the above-mentioned method steps.

[0010] In summary, one or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: The present application obtains multi-source image data of the glass curtain wall facade, which includes front view images, infrared thermal imaging images and ultraviolet spectrum images collected at different time periods and angles, and can comprehensively collect the surface state information of the glass curtain wall; the multi-source image data are aligned, fused and identified and analyzed to accurately extract the crack characteristics, deformation characteristics and temperature distribution characteristics of the glass curtain wall surface; then, based on these characteristics, an abnormal feature map is established, and multi-level risk areas are determined through spatiotemporal evolution analysis, and finally a monitoring report is generated, thereby realizing comprehensive monitoring of the glass curtain wall. There is no need to install fixed monitoring equipment on the surface of the glass curtain wall, which not only avoids the impact on the aesthetics of the building, but also overcomes the problem of limited monitoring points, thereby realizing comprehensive and accurate monitoring of the glass curtain wall. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 This is a flow chart of a glass curtain wall intelligent safety monitoring method provided in an embodiment of the present application; Figure 2 This is a module diagram of a glass curtain wall intelligent safety monitoring system provided by an embodiment of the present application; Figure 3 This is a structural diagram of an electronic device provided in an embodiment of the present application.

[0012] Description of reference numerals: 300, electronic device; 301, processor; 302, communication bus; 303, user interface; 304, network interface; 305, memory. DETAILED DESCRIPTION

[0013] In order to enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this application, not all of the embodiments.

[0014] In the description of the embodiments of this application, words such as "for example" or "for instance" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as "for example" or "for instance" in the embodiments of this application should not be construed as being preferred or advantageous over other embodiments or designs. Rather, the use of words such as "for example" or "for instance" is intended to present the relevant concepts in a concrete manner.

[0015] In the description of the embodiments of the present application, the term "multiple" means two or more. For example, multiple systems refer to two or more systems, and multiple screen terminals refer to two or more screen terminals. In addition, the terms "first" and "second" are used for descriptive purposes only and are not to be understood as indicating or implying relative importance or implicitly indicating the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized.

[0016] The following will provide a clear and complete description of the technical solutions in the embodiments of the present application in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments.

[0017] Please refer to Figure 1 , a flowchart of a glass curtain wall intelligent safety monitoring method is proposed. The method can be implemented by a computer program, a single-chip microcomputer, or run on a glass curtain wall intelligent safety monitoring system. The computer program can be integrated into a computer device or run as an independent tool application. Specifically, the method includes steps 10 to 50, which are as follows: Step 10: Acquire multi-source image data of the glass curtain wall facade, which includes front view images, infrared thermal imaging images, and ultraviolet spectrum images of the glass curtain wall collected at different time periods and angles.

[0018] Among them, in the embodiment of the present application, multi-source image data refers to a set of image data obtained by photographing the facade of the glass curtain wall through multiple image acquisition devices carried by a drone, including front view images, infrared thermal imaging images and ultraviolet spectrum images.

[0019] The front view image refers to the image of the glass curtain wall facade collected by the visible light camera carried by the drone. This image can record the intuitively visible defect features such as cracks and breakage on the surface of the glass curtain wall.

[0020] Infrared thermal imaging images refer to the temperature distribution images of glass curtain walls collected by infrared thermal imagers carried by drones. This image can reflect the temperature field distribution on the surface of the glass curtain wall and help to discover stress concentration areas inside the glass curtain wall.

[0021] Ultraviolet spectral image refers to the spectral characteristic image of the glass curtain wall surface collected by the ultraviolet spectral camera carried by the drone. This image can reflect the structural changes and material performance degradation of the glass curtain wall surface.

[0022] Specifically, the drone is controlled to cruise along a preset flight path on the exterior facade of the glass curtain wall, and multi-source image data of the glass curtain wall is collected through the visible light camera, infrared thermal imager, and ultraviolet spectrum camera carried by the drone. Specifically, under different weather conditions such as sunny and cloudy days, the drone is controlled to shoot the glass curtain wall at three angles: vertical, 30 degrees to the left, and 30 degrees to the right, in the morning, noon, and afternoon. Among them, the visible light camera collects the front view image of the glass curtain wall to record visible defects such as surface cracks; the infrared thermal imager collects infrared thermal imaging images of the glass curtain wall to detect temperature abnormalities; and the ultraviolet spectrum camera collects ultraviolet spectrum images of the glass curtain wall to identify material degradation. By collecting multi-source image data at different times and angles, the surface status information of the glass curtain wall can be fully reflected, providing a complete data foundation for subsequent analysis.

[0023] Based on the above embodiment, as an optional embodiment, the step of acquiring multi-source image data of the glass curtain wall facade may further include the following steps: Step 101: Divide the glass curtain wall into multiple monitoring areas, each monitoring area including an image acquisition reference point.

[0024] Specifically, in this embodiment, the design drawings of the glass curtain wall are first obtained and divided into several 10-meter x 10-meter square monitoring areas based on the overall dimensions of the glass curtain wall. Image acquisition reference points are set at the four vertices and the center point of each monitoring area, and the spatial coordinate information of each reference point is recorded. This division method ensures that the monitoring area size is suitable for the drone's shooting range and that there is a certain overlap between adjacent areas, which is conducive to subsequent image stitching and analysis. By rationally dividing the monitoring area and setting reference points, a spatial reference basis is provided for subsequent route planning and image acquisition.

[0025] Step 102: Determine a monitoring route based on each image acquisition reference point, and control the UAV to fly along the monitoring route.

[0026] Specifically, a shortest path algorithm was used to connect all reference points to form an S-shaped round-trip route, ensuring that the drone covered the entire monitoring area. During route planning, the vertical distance between the drone and the curtain wall was set at 5 meters, the flight speed was set at 2 meters per second, and hovering points were preset at each reference point. Furthermore, obstacle avoidance buffer zones were incorporated into the route to account for the distribution of obstacles around the building. This scientifically planned monitoring route ensured the safety and integrity of the data collection process while improving efficiency.

[0027] Step 103: On the monitoring route, obtain frontal images of the glass curtain wall taken by the drone at preset distances, infrared thermal imaging images of the glass curtain wall taken at preset intervals, and ultraviolet spectrum images of the glass curtain wall taken at different angles, wherein the ultraviolet spectrum images are taken when the ambient light intensity is greater than a light intensity threshold.

[0028] Specifically, differentiated acquisition strategies are designed based on the acquisition characteristics and data requirements of different types of images. The acquisition of orthographic images is based on spatial dimension control, and it is necessary to ensure that there is a certain degree of overlap between images for later stitching; the acquisition of infrared thermal imaging is based on time dimension control, and it is necessary to consider the dynamic change characteristics of the temperature field; the acquisition of ultraviolet spectrum images needs to consider the influence of lighting conditions and shooting angles at the same time. In specific implementation, the drone is controlled to fly according to the preset route, and the visible light camera is used to collect orthographic images at fixed distance intervals. The infrared thermal imager is used to collect thermal imaging images at fixed time intervals. When the lighting conditions are met, the ultraviolet spectrum camera is controlled to collect spectral images at multiple angles. During the acquisition process, the attitude control system of the drone is used to ensure that the camera and the curtain wall surface maintain the correct shooting angle, and the image stabilization system is used to ensure the quality of the collected images.

[0029] For example, when the drone flies along the monitoring route, the following strategy is adopted to collect images: obtain the front view image of the glass curtain wall taken by the drone at a preset distance, and the visible light camera takes a front view image every 2 meters; the infrared thermal imager collects a thermal imaging image every 30 seconds; when the light sensor detects that the ambient light intensity exceeds 1000 lux, the ultraviolet spectrum camera is controlled to take ultraviolet spectrum images at three angles: vertical, 15 degrees to the left, and 15 degrees to the right.

[0030] Step 104: Integrate the front view image, infrared thermal imaging image, and ultraviolet spectrum image of the glass curtain wall into multi-source image data.

[0031] Specifically, a multi-level data organization approach is used to integrate different types of image data. First, a unified data standard is established, adding metadata tags containing temporal information, spatial information, and acquisition parameters to each image. A hierarchical data structure is then constructed to systematically manage the data based on dimensions such as monitoring area, image type, and acquisition time. During the data integration process, associations between different image types are established based on spatiotemporal characteristics, forming a structured multi-source image dataset.

[0032] Step 20: Perform registration, fusion, and recognition analysis on the multi-source image data to obtain the crack characteristics, deformation characteristics, and temperature distribution characteristics of the glass curtain wall surface.

[0033] Among them, in the embodiment of the present application, crack characteristics refer to various types of damage that appear on the surface of the glass curtain wall, including surface damage such as cracks, scratches, and notches on the glass panel. These characteristics can be directly observed and identified through front-view images and are important indicators for evaluating the safety status of the glass curtain wall.

[0034] Deformation characteristics refer to the structural deformation of the glass curtain wall, including the concave and convex deformation, tilt, misalignment of the panel, and deformation of the supporting components. These characteristics reflect the overall stress state and structural stability of the glass curtain wall and can be obtained through analysis of images taken from multiple angles.

[0035] Temperature distribution characteristics refer to the temperature field distribution on the surface of the glass curtain wall, including temperature anomaly areas, temperature gradient changes, and thermal stress concentration areas. These characteristics can reflect the internal stress state and potential structural hazards of the glass curtain wall, and can be quantitatively analyzed through infrared thermal imaging images.

[0036] Specifically, the SIFT feature point matching algorithm is used to spatially register the front view image, infrared thermal image, and ultraviolet spectrum image, establishing positional correspondence between the images. A deep learning network is then used to extract features from the registered images and perform fusion analysis. A convolutional neural network is used to identify crack features in the front view image; a 3D model is reconstructed from the multi-view data obtained from the image registration to analyze the deformation characteristics of the glass curtain wall; and a temperature field model is established based on the infrared thermal image to extract temperature distribution characteristics. Through the collaborative analysis of multi-source image data, a comprehensive characterization of the glass curtain wall surface condition is achieved.

[0037] Based on the above embodiment, as an optional embodiment, the step of performing registration, fusion, and recognition analysis on multi-source image data to obtain crack characteristics, deformation characteristics, and temperature distribution characteristics on the glass curtain wall surface may further include the following steps: Step 201: extracting feature points of the front view image from the multi-source image data and establishing a corresponding relationship between the feature points.

[0038] Specifically, to achieve accurate registration of multi-source image data, it is first necessary to extract stable and reliable feature points from the orthographic image. An improved SIFT (Scale Invariant Feature Transform) algorithm is used to extract feature points from the orthographic image. This algorithm constructs a Gaussian difference pyramid to detect local extreme points in different scale spaces and extract feature points with rotation invariance and scale invariance. For the extracted feature points, their local gradient direction histogram is calculated as a feature descriptor. Then, the nearest neighbor matching algorithm is used to match the feature points in different images and establish a correspondence between the feature points. To improve matching accuracy, the RANSAC algorithm is introduced to eliminate erroneous matching points, and finally a stable and reliable feature point correspondence is obtained.

[0039] Step 202: Based on the correspondence between the feature points, spatial registration is performed on the front view image, the infrared thermal imaging image, and the ultraviolet spectrum image.

[0040] Specifically, based on the established feature point correspondences, a multi-step iterative approach is used to achieve image spatial registration. First, the homography matrix between the images is calculated based on the feature point correspondences. This matrix describes the spatial transformation relationship between different images. Then, the homography matrix parameters are optimized using gradient descent to minimize the registration error. During the registration process, the orthographic image and infrared thermal image are first registered, and the registration results are then registered with the ultraviolet spectrum image to ensure precise spatial correspondence between the three image types. Finally, the images are resampled using bilinear interpolation to eliminate geometric distortion during the registration process.

[0041] Step 203: performing edge enhancement processing on the registered front view image to extract crack features on the surface of the glass curtain wall.

[0042] Specifically, crack feature extraction is performed on the registered frontal image. First, adaptive histogram equalization is used to enhance image contrast and highlight crack edge information. Next, an improved Canny operator is designed for edge detection, which improves the detection of small cracks through adaptive threshold selection. Morphological processing is used to remove noise from the detected edge features, and connected domain analysis is used to extract the complete crack outline. Finally, a feature vector is constructed based on the crack's geometric characteristics (such as length, width, and direction) to characterize the crack characteristics on the glass curtain wall surface.

[0043] Step 204: Divide the registered infrared thermal imaging image into temperature distribution areas to obtain temperature distribution characteristics of the glass curtain wall.

[0044] Specifically, a multi-level temperature analysis method is used to process infrared thermal imaging images. First, the thermal images are temperature-corrected to eliminate the effects of ambient temperature and measurement distance. Then, a K-means clustering algorithm is used to adaptively partition the temperature field, dividing the temperature distribution into multiple hierarchical regions. Based on this, the temperature statistical characteristics of each region are calculated, including the average temperature, temperature gradient, and the distribution of temperature anomalies. By analyzing the temperature differences and temperature gradient changes between adjacent regions, temperature anomaly areas are identified, ultimately forming a characteristic description of the temperature distribution of the glass curtain wall.

[0045] Step 205: performing pixel-level comparison between the registered front view image and a preset standard glass curtain wall image, and calculating a deformation deviation value. When the deformation deviation value is greater than a preset threshold, it is determined to be a deformation feature.

[0046] Specifically, the registered front view image is calibrated against a preset standard glass curtain wall image to ensure precise spatial alignment between the two images. A phase correlation method is then used to calculate the pixel displacement field between the two images, yielding the deformation deviation value for each pixel. A deformation threshold based on engineering experience is set. When the deformation deviation value in a region exceeds the preset threshold, the region is marked as a deformation feature region. Finally, a region growing algorithm is used to determine the complete extent of the deformation region and calculate the direction and magnitude of the deformation.

[0047] Step 30: Establish an abnormal feature map of the glass curtain wall based on the crack characteristics, deformation characteristics, and temperature distribution characteristics.

[0048] In the embodiment of the present application, the abnormal feature map refers to a feature description model generated during the safety status assessment of the glass curtain wall.

[0049] Specifically, a multi-layer neural network is used to construct an abnormal feature map. The extracted crack, deformation, and temperature distribution features are respectively input into the feature encoding layer for dimensionality reduction, resulting in a low-dimensional representation of each feature. Then, an attention mechanism fusion layer adaptively assigns weights to the three features and fuses them, highlighting the contribution of key abnormal features. Finally, a decoder network is used to reconstruct the fused features into a structured abnormal feature map, which contains comprehensive information such as crack distribution, deformation degree, and temperature anomaly. This deep learning-based feature map construction method can automatically discover correlations between features and improve the accuracy of abnormal state representation.

[0050] Based on the above embodiment, as an optional embodiment, the step of establishing an abnormal characteristic map of the glass curtain wall based on the crack characteristics, deformation characteristics, and temperature distribution characteristics may further include the following steps: Step 301: Create a crack distribution layer based on the length, width and direction of the crack characteristics.

[0051] Specifically, to visualize crack distribution, a layered feature mapping method was used to create a crack distribution layer. The detected crack features were quantified according to geometric properties, including categorizing crack length into three levels: micro (less than 5mm), medium (5-20mm), and large (greater than 20mm). Crack width was also categorized into three levels: fine (less than 0.1mm), general (0.1-0.5mm), and severe (greater than 0.5mm). The crack strike angle was also recorded. Then, vector graphics were used to draw the cracks in a spatial coordinate system, using different line types and color codes to represent the crack levels and properties. Ultimately, a crack distribution layer was formed that characterized the spatial distribution characteristics of the cracks.

[0052] Step 302: Create a deformation distribution layer according to the deformation deviation value of the deformation feature.

[0053] Specifically, a deformation distribution layer is constructed by establishing a spatial distribution model of deformation characteristics. The calculated deformation deviation values ​​are normalized, and an interpolation algorithm is used to construct a continuous deformation field across the entire curtain wall surface. Then, based on the magnitude of the deformation deviation values, the degree of deformation is categorized into three levels: slight (deviation less than 5mm), moderate (deviation 5-15mm), and severe (deviation greater than 15mm). Using a pseudo-color mapping method, different shades of color are used to represent the degree of deformation, and contour lines are added to indicate deformation trends. Ultimately, a deformation distribution layer is generated that reflects the spatial distribution of deformation.

[0054] Step 303: Create a temperature anomaly layer based on the temperature gradient of the temperature distribution characteristics.

[0055] Specifically, a temperature anomaly layer is created based on temperature gradient analysis. The spatial gradient of the temperature field is calculated, including the horizontal and vertical temperature change rates. Then, based on the magnitude of the temperature gradient, the degree of temperature anomaly is divided into normal areas (gradient less than 2°C / m), caution areas (gradient 2-5°C / m), and warning areas (gradient greater than 5°C / m). Using a heat map, different temperature gradient areas are marked with gradient colors, and isotherms are added to the temperature anomaly areas to form a temperature anomaly layer that characterizes the temperature distribution.

[0056] Step 304: The crack distribution layer, the deformation distribution layer, and the temperature anomaly layer are superimposed and fused, and the superimposed and fused layers are graded and labeled based on a preset anomaly grade classification standard to generate an anomaly feature map.

[0057] Specifically, a multi-layer feature fusion method is used to generate an abnormal feature map. The three layers are aligned and superimposed in the spatial coordinate system, and the weighted superposition algorithm is used to determine the weight coefficient of each layer feature. Then, based on the preset abnormality level classification standard, the superimposed features are comprehensively evaluated into four levels: normal, mild abnormality, moderate abnormality and severe abnormality. Among them, when cracks and large deformations appear in the area at the same time, it is classified as severe abnormality; when moderate cracks or deformations appear in the area, and accompanied by temperature abnormalities, it is classified as moderate abnormality; when only a single slight abnormal feature appears in the area, it is classified as mild abnormality. Finally, a layered coloring scheme is used to mark the areas of different levels, and a legend is added to generate a complete abnormal feature map.

[0058] Step 40: Perform spatiotemporal evolution analysis on the abnormal feature map to determine the multi-level risk areas of the glass curtain wall.

[0059] Among them, the multi-level risk area in the embodiment of the present application refers to dividing the safety risks of the glass curtain wall into multiple risk areas of different levels according to the degree of superposition and distribution of various abnormal features in the abnormal feature map.

[0060] Specifically, a spatiotemporal sequence analysis method is used to perform evolutionary analysis on the abnormal feature maps, establish a time series database, and store abnormal feature maps collected at different time points. Then, the feature data at different time points are aligned through a time series registration algorithm, and the feature change rate and development trend are calculated. Based on the feature evolution law, a Markov prediction model is used to evaluate the development trend of abnormal features, and spatial cluster analysis is combined to determine the risk area. When there are no obvious abnormal features in a certain area, it is divided into a first-level risk area; when a single mild abnormal feature appears, it is divided into a second-level risk area; when multiple abnormal features appear and show an accelerating development trend, it is divided into a third-level risk area; when multiple serious abnormal features appear, it is divided into a fourth-level risk area.

[0061] Based on the above embodiment, as another optional embodiment, the step of performing spatiotemporal evolution analysis on the abnormal feature map to determine the multi-level risk areas of the glass curtain wall may further include the following steps: Step 401: establishing a multi-feature weight matrix based on the abnormal feature map, where the multi-feature weight matrix includes a crack weight coefficient, a deformation weight coefficient, and a temperature weight coefficient.

[0062] Specifically, the analytic hierarchy process (AHP) was used to establish a multi-feature weight matrix to quantify the impact of different abnormal features on the safety status of the glass curtain wall. First, a judgment matrix for the characteristic factors was constructed, and the importance ratios between the features were determined based on engineering experience and expert evaluation. Then, the maximum eigenvalue and corresponding eigenvector of the judgment matrix were calculated using the eigenvector method, and a consistency check was performed. For example, the final weight coefficient for cracks was 0.4 (because it directly reflects structural damage), the weight coefficient for deformation was 0.35 (because it represents overall stability), and the weight coefficient for temperature was 0.25 (because it reflects potential risks).

[0063] Step 402: Divide the abnormal feature map into a grid and calculate the abnormal feature comprehensive score of each grid unit. The abnormal feature comprehensive score is determined by the sum of the products of the crack feature, deformation feature and temperature distribution feature and the corresponding weight coefficient.

[0064] Specifically, the anomaly feature map is refined and scored. For example, a glass curtain wall is divided into uniform grid cells of 1 meter by 1 meter size, with each grid cell containing complete feature information. Each feature is then normalized: crack features are assigned a value of 0-1 based on the severity of their length and width; deformation features are assigned a value of 0-1 based on the magnitude of the deviation; and temperature distribution features are assigned a value of 0-1 based on the temperature gradient. Finally, the normalized feature values ​​are multiplied by the corresponding weight coefficients and summed to obtain a comprehensive score for the anomaly feature for each grid cell. This grid-based scoring method achieves a quantitative representation of the anomaly state.

[0065] Step 403: Determine the multi-level risk areas of the glass curtain wall based on the comprehensive score of abnormal features.

[0066] Specifically, to accurately identify areas with similar abnormal characteristics within glass curtain walls, spatial statistics and cluster analysis methods are used to process abnormal feature data. First, a spatial distribution function for the comprehensive score of the abnormal characteristics is constructed based on the kriging interpolation method. This function is expressed as F(x,y)=∑λiZ(xi,yi), where (x,y) represents the coordinates of the point to be estimated, λi is the weight coefficient, and Z(xi,yi) represents the comprehensive score of the abnormal characteristics of a given grid cell. This function can estimate the comprehensive score of the abnormal characteristics at any location, achieving a continuous spatial distribution of the score. Then, the abnormal feature correlation between grid cells is calculated, and the Moran's I index is used to assess the spatial correlation of the scores of adjacent grid cells. For any two grid cells i and j, the correlation is calculated as Rij = (Zi-Z̄)(Zj-Z̄) / σ², where Zi and Zj are the comprehensive scores of the abnormal characteristics of the corresponding grid cells, Z̄ is the mean, and σ² is the variance. When the correlation exceeds a set threshold (e.g., 0.7), the two grid cells are considered to have significant feature correlation.

[0067] Next, a cluster analysis of the grid cells was performed based on the DBSCAN density clustering algorithm. The algorithm uses the correlation degree of abnormal features as a distance metric and aggregates spatially adjacent grid cells with similar scoring features into a set of regions. Specifically, the minimum sample number MinPts = 4 and the neighborhood radius ε = 0.3 are set. Through iterative search, all density-reachable grid cells are classified into the same category, and finally multiple sets of regions with similar abnormal features are obtained. Finally, the average abnormal feature comprehensive score of each set of regions is calculated, and the risk level is divided based on the score range: the set of regions with a score less than 0.3 is divided into a first-level risk area, the range between 0.3-0.6 is divided into a second-level risk area, the range between 0.6-0.8 is divided into a third-level risk area, and the score greater than 0.8 is divided into a fourth-level risk area.

[0068] Step 50: Generate a monitoring report for the glass curtain wall based on the multi-level risk areas.

[0069] Specifically, the distribution of multi-level risk areas is plotted as a risk heat map, with different colors used to identify risk areas from level one to level four. Feature data is then extracted for each risk area, including the size parameters of crack features, the deviation values ​​of deformation features, and the gradient values ​​of temperature distribution features, to generate a quantitative indicator curve. Finally, based on the risk level assessment results, a monitoring report is automatically generated, including a risk distribution map, a statistical table of abnormal features, and treatment recommendations. Specifically, for level three and four risk areas, the specific manifestations and development trends of abnormal features are described, and corresponding maintenance recommendations are given. This visual report format intuitively demonstrates the safety status of the glass curtain wall.

[0070] Based on the above embodiment, as another optional embodiment, the step of generating a monitoring report of a glass curtain wall according to the multi-level risk areas may further include the following steps: Step 501: Calculate the characteristic combination strength of each risk area, where the characteristic combination strength is the superposition value of the crack feature, deformation feature, and temperature distribution feature.

[0071] Specifically, a weighted fusion method is used to calculate the feature combination strength for each risk area. First, each feature is standardized: the crack feature is calculated by calculating the ratio of the crack length to the standard length to obtain the crack feature value Fc (0-1); the deformation feature is calculated by calculating the ratio of the actual displacement to the allowable displacement to obtain the deformation feature value Fd (0-1); and the temperature distribution feature is calculated by calculating the ratio of the temperature difference to the allowable temperature difference to obtain the temperature feature value Ft (0-1). Then, considering the impact of each feature on safety, weight coefficients are set: crack feature weight Wc = 0.4, deformation feature weight Wd = 0.35, and temperature feature weight Wt = 0.25. The final feature combination strength S is obtained through weighted summation: S = Fc × Wc + Fd × Wd + Ft × Wt. This calculation method not only ensures the reasonable contribution of each feature but also achieves the effective fusion of multi-dimensional features.

[0072] Step 502: Determine the risk status coefficient of each risk area based on the level and feature combination strength of each risk area.

[0073] Specifically, a two-factor risk status assessment model is constructed based on the risk level and the strength of the feature combination. First, the basic coefficient α is determined according to the risk area level: for the first-level risk area, α = 0.2; for the second-level risk area, α = 0.4; for the third-level risk area, α = 0.6; for the fourth-level risk area, α = 0.8. Then, the adjustment factor β of the feature combination strength is introduced. When the feature combination strength S exceeds the warning value, the risk status coefficient is correspondingly increased. The specific calculation formula is: the risk status coefficient R = α×(1 + β×S), where the value range of β is 0.1 - 0.5 and increases with the increase of the feature combination strength. When S ≤ 0.3, β = 0.1; when 0.3 < S ≤ 0.6, β = 0.3; when S > 0.6, β = 0.5.

[0074] Step 503: Generate a monitoring report including the risk status coefficients of each risk area and the corresponding warning information.

[0075] Specifically, the embodiment of the present application adopts an intelligent monitoring report and designs a hierarchical report template, including three parts: general overview, detailed analysis, and warning suggestions. In the general overview, a risk area distribution map is generated, and each level of risk area is marked with different colors; in the detailed analysis part, the risk status coefficients of each risk area and their constituent elements are listed, including the basic coefficient, the feature combination strength, and the adjustment factor; in the warning suggestions part, the warning level is set according to the risk status coefficient: when R < 0.3, it is in a normal state and a green prompt is displayed; when 0.3 ≤ R < 0.5, it is in a concerned state and a yellow prompt is displayed; when 0.5 ≤ R < 0.7, it is in a warning state and an orange warning is displayed; when R ≥ 0.7, it is in a dangerous state and a red warning is displayed. At the same time, the system automatically generates targeted processing suggestions, such as "It is recommended to check regularly", "Need to strengthen monitoring", "Immediately carry out maintenance", etc.

[0076] Please refer to Figure 2 , which is a schematic diagram of the modules of an intelligent safety monitoring system for glass curtain walls provided by the embodiment of the present application. Among them, the system includes: An image data acquisition module, which is used to acquire multi-source image data of the outer facade of the glass curtain wall. The multi-source image data includes frontal images, infrared thermal imaging images, and ultraviolet spectral images of the glass curtain wall collected at different times and from different angles; A feature recognition module, which is used to perform registration fusion and recognition analysis on the multi-source image data to obtain crack features, deformation features, and temperature distribution features on the surface of the glass curtain wall; A feature map generation module, which is used to establish an abnormal feature map of the glass curtain wall based on the crack features, deformation features, and temperature distribution features; A risk area determination module, which is used to perform spatio-temporal evolution analysis on the abnormal feature map to determine multiple levels of risk areas of the glass curtain wall; A monitoring report generating module is used to generate a monitoring report of the glass curtain wall according to the multi-level risk areas.

[0077] Optionally, the image data acquisition module is further configured to divide the glass curtain wall into a plurality of monitoring areas, each of the monitoring areas including an image acquisition reference point; Determining a monitoring route based on each of the image acquisition reference points, and controlling the drone to fly along the monitoring route; On the monitoring route, obtaining frontal images of the glass curtain wall taken by the drone at preset distances, infrared thermal imaging images of the glass curtain wall taken at preset intervals, and ultraviolet spectrum images of the glass curtain wall taken at different angles, wherein the ultraviolet spectrum images are taken when the ambient light intensity is greater than a light intensity threshold; The front view image, infrared thermal imaging image and ultraviolet spectrum image of the glass curtain wall are integrated into multi-source image data.

[0078] Optionally, the feature recognition module is further configured to extract feature points of the front view image in the multi-source image data and establish a corresponding relationship between the feature points; Based on the correspondence between the feature points, spatially registering the front view image, the infrared thermal imaging image, and the ultraviolet spectrum image; performing edge enhancement processing on the registered front view image to extract crack features on the surface of the glass curtain wall; Dividing the registered infrared thermal imaging image into temperature distribution areas to obtain the temperature distribution characteristics of the glass curtain wall; The registered front view image is compared with the preset standard glass curtain wall image at the pixel level, and the deformation deviation value is calculated. When the deformation deviation value is greater than a preset threshold, it is determined as a deformation feature.

[0079] Optionally, the feature map generation module is further configured to establish a crack distribution layer according to the length, width and direction of the crack features; Establishing a deformation distribution layer according to the deformation deviation value of the deformation feature; Establishing a temperature anomaly layer according to the temperature gradient of the temperature distribution characteristics; The crack distribution layer, deformation distribution layer and temperature anomaly layer are superimposed and fused, and based on a preset anomaly grade classification standard, the superimposed and fused layers are graded and labeled to generate an anomaly feature map.

[0080] Optionally, the risk area determination module is further configured to establish a multi-feature weight matrix based on the abnormal feature map, wherein the multi-feature weight matrix includes a crack weight coefficient, a deformation weight coefficient, and a temperature weight coefficient; Dividing the abnormal feature map into a grid and calculating a comprehensive abnormal feature score for each grid unit, wherein the comprehensive abnormal feature score is determined by the sum of the products of the crack feature, the deformation feature, and the temperature distribution feature with corresponding weight coefficients; The multi-level risk areas of the glass curtain wall are determined according to the comprehensive scores of the abnormal features.

[0081] Optionally, the risk area determination module is further configured to construct a spatial distribution function of the comprehensive score of the abnormal features, and calculate the abnormal feature correlation between grid cells based on the spatial distribution function; Performing cluster analysis on the grid cells according to the abnormal feature correlation to obtain a set of regions with similar abnormal features; The glass curtain wall is divided into multiple risk areas according to the comprehensive score of abnormal characteristics of the set of areas.

[0082] Optionally, the monitoring report generation module is further configured to calculate a characteristic combination strength of each risk area, where the characteristic combination strength is a superposition value of crack characteristics, deformation characteristics, and temperature distribution characteristics; Determining a risk status coefficient for each risk area based on the level of each risk area and the strength of the characteristic combination; Generate a monitoring report containing the risk status coefficient of each risk area and corresponding early warning information.

[0083] It should be noted that the above embodiments provide systems that implement their functions using only the division of the above functional modules as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the system and method embodiments provided in the above embodiments are based on the same concept. The specific implementation process is detailed in the method embodiment and will not be repeated here.

[0084] An embodiment of the present application also provides a computer storage medium, which can store multiple instructions. The instructions are suitable for being loaded by a processor and executing a glass curtain wall intelligent safety monitoring method of the above embodiment. The specific execution process can be found in the specific description of the above embodiment and will not be repeated here.

[0085] Please refer to Figure 3 The present application also discloses an electronic device. Figure 3 The electronic device 300 may include: at least one processor 301 , at least one network interface 304 , a user interface 303 , a memory 305 , and at least one communication bus 302 .

[0086] The communication bus 302 is used to implement the connection and communication between these components.

[0087] The user interface 303 may include a display screen (Display) and a camera (Camera). Optionally, the user interface 303 may also include a standard wired interface and a wireless interface.

[0088] The network interface 304 may optionally include a standard wired interface or a wireless interface (such as a WI-FI interface).

[0089] The processor 301 may include one or more processing cores. Using various interfaces and circuits, the processor 301 connects to various components within the server. It executes instructions, programs, code sets, or instruction sets stored in the memory 305, as well as accesses data stored in the memory 305, to perform various server functions and process data. Optionally, the processor 301 may be implemented using at least one of the following hardware forms: a digital signal processing (DSP), a field-programmable gate array (FPGA), or a programmable logic array (PLA). The processor 301 may integrate one or a combination of a central processing unit (CPU), a graphics processing unit (GPU), and a modem. The CPU primarily processes the operating system, user interface, and application programs; the GPU is responsible for rendering and drawing content displayed on the display screen; and the modem handles wireless communications. It is understood that the modem may not be integrated into the processor 301 but implemented as a separate chip.

[0090] Among them, the memory 305 may include a random access memory (RAM) or a read-only memory (Read-Only Memory). Optionally, the memory 305 includes a non-transitory computer-readable storage medium. The memory 305 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 305 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data involved in the above-mentioned various method embodiments, etc. The memory 305 may also optionally be at least one storage device located away from the aforementioned processor 301. Refer to Figure 3 , as a computer storage medium, the memory 305 may include an operating system, a network communication module, a user interface module, and an application program of a glass curtain wall intelligent safety monitoring method.

[0091] exist Figure 3 In the electronic device 300 shown, the user interface 303 is mainly used to provide an input interface for the user and obtain the data input by the user; and the processor 301 can be used to call an application program for storing a glass curtain wall intelligent safety monitoring method in the memory 305. When executed by one or more processors 301, the electronic device 300 executes one or more methods in the above-mentioned embodiments. It should be noted that for the aforementioned method embodiments, for the sake of simplicity of description, they are all expressed as a series of action combinations, but those skilled in the art should know that this application is not limited to the described order of actions, because according to this application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required for this application.

[0092] In the above embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0093] In the several embodiments provided in this application, it should be understood that the disclosed devices can be implemented in other ways. For example, the device embodiments described above are merely schematic, such as the division of units, which is only a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some service interface, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

[0094] Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0095] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0096] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of this application, or the portion that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions for causing a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the various embodiments of the method of this application. The aforementioned memory includes various media that can store program code, such as USB flash drives, mobile hard drives, magnetic disks, or optical disks.

[0097] The above are merely exemplary embodiments of the present disclosure and are not intended to limit the scope of the present disclosure. In other words, any equivalent variations and modifications made in accordance with the teachings of the present disclosure are still within the scope of the present disclosure. Those skilled in the art will readily conceive of other embodiments of the present disclosure after considering the disclosure and the practical implications thereof.

[0098] This application is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art not described herein. The description and examples are to be considered as exemplary only, and the scope and spirit of the present disclosure are to be defined by the claims.

Claims

1. A glass curtain wall intelligent safety monitoring method, characterized in that: The method comprises: Acquire multi-source image data of the glass curtain wall facade, wherein the multi-source image data includes front view images, infrared thermal imaging images, and ultraviolet spectrum images of the glass curtain wall collected at different time periods and different angles; Performing registration, fusion, and recognition analysis on the multi-source image data to obtain crack characteristics, deformation characteristics, and temperature distribution characteristics on the surface of the glass curtain wall; Establishing an abnormal characteristic map of the glass curtain wall based on the crack characteristics, deformation characteristics and temperature distribution characteristics; Performing a spatiotemporal evolution analysis on the abnormal characteristic map to determine the multi-level risk areas of the glass curtain wall; A monitoring report of the glass curtain wall is generated according to the multi-level risk areas.

2. The glass curtain wall intelligent safety monitoring method according to claim 1, characterized in that: The step of acquiring multi-source image data of the glass curtain wall facade includes: Dividing the glass curtain wall into a plurality of monitoring areas, each of the monitoring areas includes an image acquisition reference point; Determining a monitoring route based on each of the image acquisition reference points, and controlling the drone to fly along the monitoring route; On the monitoring route, obtaining frontal images of the glass curtain wall taken by the drone at preset distances, infrared thermal imaging images of the glass curtain wall taken at preset intervals, and ultraviolet spectrum images of the glass curtain wall taken at different angles, wherein the ultraviolet spectrum images are taken when the ambient light intensity is greater than a light intensity threshold; The front view image, infrared thermal imaging image and ultraviolet spectrum image of the glass curtain wall are integrated into multi-source image data.

3. The glass curtain wall intelligent safety monitoring method according to claim 1, characterized in that: The registering, fusing and identifying analysis of the multi-source image data to obtain crack characteristics, deformation characteristics and temperature distribution characteristics of the glass curtain wall surface includes: Extracting feature points of the front view image from the multi-source image data and establishing a corresponding relationship between the feature points; Based on the corresponding relationship of the feature points, spatially registering the front view image, the infrared thermal imaging image, and the ultraviolet spectrum image; performing edge enhancement processing on the registered front view image to extract crack features on the surface of the glass curtain wall; Dividing the registered infrared thermal imaging image into temperature distribution areas to obtain the temperature distribution characteristics of the glass curtain wall; The registered front view image is compared with the preset standard glass curtain wall image at the pixel level, and the deformation deviation value is calculated. When the deformation deviation value is greater than a preset threshold, it is determined as a deformation feature.

4. The glass curtain wall intelligent safety monitoring method according to claim 1, characterized in that: The establishing of the abnormal characteristic map of the glass curtain wall based on the crack characteristics, deformation characteristics and temperature distribution characteristics includes: Establishing a crack distribution layer according to the length, width and direction of the crack characteristics; Establishing a deformation distribution layer according to the deformation deviation value of the deformation feature; Establishing a temperature anomaly layer according to the temperature gradient of the temperature distribution characteristics; The crack distribution layer, deformation distribution layer and temperature anomaly layer are superimposed and fused, and based on a preset anomaly grade classification standard, the superimposed and fused layers are graded and labeled to generate an anomaly feature map.

5. The glass curtain wall intelligent safety monitoring method according to claim 1, characterized in that: The performing of spatiotemporal evolution analysis on the abnormal characteristic map to determine the multi-level risk areas of the glass curtain wall includes: Establishing a multi-feature weight matrix based on the abnormal feature map, wherein the multi-feature weight matrix includes a crack weight coefficient, a deformation weight coefficient, and a temperature weight coefficient; Dividing the abnormal feature map into a grid and calculating a comprehensive abnormal feature score for each grid unit, wherein the comprehensive abnormal feature score is determined by the sum of the products of the crack feature, the deformation feature, and the temperature distribution feature with corresponding weight coefficients; The multi-level risk areas of the glass curtain wall are determined according to the comprehensive scores of the abnormal features.

6. The glass curtain wall intelligent safety monitoring method according to claim 5, characterized in that: The determining of the multi-level risk areas of the glass curtain wall according to the comprehensive scoring of the abnormal features includes: Constructing a spatial distribution function of the comprehensive score of the abnormal features, and calculating the abnormal feature correlation between grid cells based on the spatial distribution function; Performing cluster analysis on the grid cells according to the abnormal feature correlation to obtain a set of regions with similar abnormal features; The glass curtain wall is divided into multiple risk areas according to the comprehensive score of abnormal characteristics of the set of areas.

7. The glass curtain wall intelligent safety monitoring method according to claim 1, characterized in that: Generating a monitoring report of the glass curtain wall according to the multi-level risk areas includes: Calculating the characteristic combination strength of each risk area, where the characteristic combination strength is the superposition value of crack characteristics, deformation characteristics, and temperature distribution characteristics; Determining a risk status coefficient for each risk area based on the level of each risk area and the strength of the characteristic combination; Generate a monitoring report containing the risk status coefficient of each risk area and corresponding early warning information.

8. A glass curtain wall intelligent safety monitoring system, characterized in that: The system comprises: An image data acquisition module is used to acquire multi-source image data of the glass curtain wall facade, wherein the multi-source image data includes front view images, infrared thermal imaging images, and ultraviolet spectrum images of the glass curtain wall collected at different time periods and different angles; A feature recognition module is used to perform registration, fusion, and recognition analysis on the multi-source image data to obtain crack features, deformation features, and temperature distribution features on the surface of the glass curtain wall; A feature map generation module, configured to establish an abnormal feature map of the glass curtain wall based on the crack features, deformation features, and temperature distribution features; a risk area determination module, configured to perform spatiotemporal evolution analysis on the abnormal feature map to determine the multi-level risk areas of the glass curtain wall; A monitoring report generating module is used to generate a monitoring report of the glass curtain wall according to the multi-level risk areas.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a plurality of instructions, and the instructions are suitable for being loaded by a processor and executing the method according to any one of claims 1 to 7.

10. An electronic device, characterized in that: The electronic device comprises a processor, a memory, a user interface and a network interface, wherein the memory is used to store instructions, the user interface and the network interface are used for communication, and the processor is used to execute the instructions stored in the memory so that the electronic device executes the method according to any one of claims 1 to 7.

Images