Computer vision-based safety monitoring method and system for housing status

Abstract

The present invention relates to the field of intelligent monitoring, and proposes a safety monitoring method and system for house status based on computer vision, the method comprising: obtaining a high-definition image of a target house, performing image enhancement and edge corner point detection to identify key corner points, and then performing structural analysis to construct a three-dimensional house model, then identifying the house structure facilities in the model and meshing, calculating the house surface thickness value, finding abnormal thickness points and abnormal areas, performing risk marking and regional force analysis to obtain a critical force value, and then determining the safety level of the house structure, combining the preset safety standard monitoring and evaluation to obtain evaluation parameters, and finally generating a safety monitoring report. The present invention can improve the monitoring efficiency and accuracy of the house status.

Description

Technical Field

[0001] The present invention relates to the field of intelligent monitoring, and in particular to a computer vision-based method and system for realizing safe monitoring of house status. Background Art

[0002] In the field of smart homes, methods for security monitoring of house status are gaining more and more attention, among which computer vision technology plays an important role in house security monitoring.

[0003] At present, the methods for safety monitoring of house status mainly include traditional means such as manual inspection and infrared sensing. Among them, manual inspection mainly relies on manpower for regular inspection and maintenance, which is inefficient and easy to miss potential dangers. Infrared sensing can detect moving objects, but it is easily affected by environmental factors, resulting in a high false alarm rate. Therefore, a safety monitoring method for house status based on computer vision is needed to improve the efficiency and accuracy of house status monitoring. Summary of the invention

[0004] The present invention provides a computer vision-based safety monitoring method and system for house status, the main purpose of which is to improve the monitoring efficiency and accuracy of house status.

[0005] Acquire a house panoramic image corresponding to a target house, perform image enhancement on the house panoramic image to obtain a house enhanced image, perform edge corner point detection on the house enhanced image to obtain edge corner point features, and identify key corner points in the house enhanced image based on the edge corner point features;

[0006] Based on the key corner points, a structural analysis is performed on the enhanced image of the house to obtain structural features of the house, and based on the structural features of the house, a three-dimensional construction is performed on the surface of the target house to obtain a three-dimensional house model;

[0007] Identify the house structure facilities in the three-dimensional house model, mesh the house structure facilities to obtain structural mesh units, and calculate the house surface thickness value corresponding to the target house based on the structural mesh units;

[0008] Identify the abnormal thickness point corresponding to the thickness value of the housing surface, determine the abnormal area corresponding to the abnormal thickness point, perform risk marking on the abnormal area to obtain the risk marked area, perform regional force analysis on the risk marked area to obtain regional force data, and calculate the force critical value corresponding to the regional force data;

[0009] Based on the stress critical value, the structural safety level corresponding to the target house is determined; based on the structural safety level and preset safety standards, the target house is monitored and evaluated to obtain evaluation parameters; based on the evaluation parameters, a safety monitoring report corresponding to the target house is generated.

[0010] Optionally, the identifying key corner points in the house enhanced image based on the edge corner point features includes:

[0011] Smoothing the edge corner point features to obtain smoothed corner point features;

[0012] Calculating the gradient amplitude corresponding to the smoothed corner feature;

[0013] Based on the gradient magnitude, marking the corner point candidate points in the house enhanced image;

[0014] Querying the window response value corresponding to the corner point candidate point;

[0015] Based on the window response value, key corner points in the house enhanced image are identified.

[0016] Optionally, the three-dimensionally constructing the surface of the target house based on the house structure feature to obtain a three-dimensional house model includes:

[0017] Performing point cloud collection on the surface of the target house to obtain point cloud data;

[0018] Performing filtering on the point cloud data to obtain a filtered point cloud;

[0019] Perform feature mapping on the filtered point cloud and the building structure features to obtain point cloud feature data;

[0020] Based on the point cloud feature data, a three-dimensional house model corresponding to the house surface of the target house is constructed.

[0021] Optionally, the calculating, based on the structural grid unit, a house surface thickness value corresponding to the target house includes:

[0022] The house surface thickness value corresponding to the target house is calculated using the following formula:

[0023]

[0024] in, represents the house surface thickness value corresponding to the target house, represents the index of the structural grid cell, represents the total number of structural grid cells, Indicates The actual thickness value measured by each grid unit is Indicates The area of ​​a grid cell, Indicates The weight coefficient of each grid cell.

[0025] Optionally, the identifying the abnormal thickness point corresponding to the building surface thickness value includes:

[0026] Extracting thickness distribution points corresponding to the building surface thickness values;

[0027] generating a thickness distribution map corresponding to the thickness distribution point;

[0028] Performing statistical analysis on the thickness distribution diagram to obtain statistical analysis indicators;

[0029] Based on the statistical analysis indicator, setting an abnormal threshold corresponding to the target house;

[0030] Based on the abnormal threshold, the thickness abnormal point corresponding to the building surface thickness value is identified.

[0031] Optionally, the risk marking the abnormal area to obtain the risk marked area includes:

[0032] Extracting the boundary of the abnormal area to obtain the area boundary points;

[0033] Based on the boundary points of the area, the risk level corresponding to the abnormal area is divided;

[0034] Based on the risk level, setting a risk label corresponding to the abnormal area;

[0035] Based on the risk label, the abnormal area is risk marked to obtain a risk marked area.

[0036] Optionally, performing regional force analysis on the risk marked area to obtain regional force data includes:

[0037] Identify the stress factors corresponding to the risk area;

[0038] Based on the stress factors, construct a stress framework corresponding to the risk area;

[0039] Based on the force-bearing frame, load simulation is performed on the risk area to obtain load simulation parameters;

[0040] Based on the load simulation parameters, regional force analysis is performed on the risk area to obtain regional force data.

[0041] Optionally, the calculating the critical force value corresponding to the regional force data includes:

[0042] The critical force value corresponding to the force data of the area is calculated using the following formula:

[0043]

[0044] in, Indicates the critical force value corresponding to the force data of the area, Represents the element index in the force data of the area, Indicates the number of elements in the force data of the region, Indicates the force data of the area The area of ​​an element, Indicates The material strength of each element, Indicates The force on an element, Indicates The length of the elements, Represents the index of the external load in the force data of the region, represents the amount of external load in the force data of the area, Indicates The size of the external load, Indicates The direction of external load action, Indicates The element and The matrix relationship between the external loads.

[0045] Optionally, determining the structural safety level corresponding to the target house based on the stress critical value includes:

[0046] Determine a standard force range corresponding to the critical force value;

[0047] Based on the standard stress range, identifying the overall stress structure corresponding to the target house;

[0048] Analyze the load-bearing ratio corresponding to the overall force-bearing structure;

[0049] Based on the load-bearing ratio, a structural safety level corresponding to the target house is determined.

[0050] In order to solve the above problems, the present invention also provides a computer vision-based security monitoring system for housing status, the system comprising:

[0051] A corner point recognition module is used to obtain a panoramic image of a house corresponding to a target house, perform image enhancement on the panoramic image of the house to obtain an enhanced image of the house, perform edge corner point detection on the enhanced image of the house to obtain edge corner point features, and identify key corner points in the enhanced image of the house based on the edge corner point features;

[0052] A three-dimensional construction module is used to perform structural analysis on the house enhanced image based on the key corner points to obtain house structural features, and based on the house structural features, perform three-dimensional construction on the house surface of the target house to obtain a three-dimensional house model;

[0053] A thickness value calculation module is used to identify the house structure facilities in the three-dimensional house model, grid the house structure facilities to obtain structural grid units, and calculate the house surface thickness value corresponding to the target house based on the structural grid units;

[0054] A critical value calculation module is used to identify the thickness abnormal point corresponding to the thickness value of the housing surface, determine the abnormal area corresponding to the thickness abnormal point, perform risk marking on the abnormal area to obtain the risk marked area, perform regional force analysis on the risk marked area to obtain regional force data, and calculate the force critical value corresponding to the regional force data;

[0055] A report generation module is used to determine the structural safety level corresponding to the target house based on the stress critical value, monitor and evaluate the target house based on the structural safety level and preset safety standards, obtain evaluation parameters, and generate a safety monitoring report corresponding to the target house based on the evaluation parameters.

[0056] Firstly, the present invention acquires a panoramic image of a house corresponding to a target house, and performs image enhancement on the panoramic image of the house to obtain an enhanced image of the house, which can not only improve the clarity and detail of the image, but also effectively reduce noise and blur, so that subsequent edge corner point detection and structural analysis are more accurate. At the same time, the present invention performs structural analysis on the enhanced image of the house based on the key corner points to obtain the structural features of the house, and can accurately extract the structural features of the house, thereby improving the accuracy and reliability of the monitoring system, and enhancing the ability to identify potential risks of the house, thereby ensuring the effectiveness and practicality of house safety monitoring. The present invention identifies the house structural facilities in the three-dimensional house model, grids the house structural facilities, and obtains the structural Grid units can break down complex house structures into smaller, more manageable units, which not only helps to accurately calculate the physical properties of each unit, such as stress, deformation, and material properties, but also improves the efficiency and accuracy of structural analysis. The present invention helps to timely discover and evaluate the impact of these abnormal points on the overall safety of the house by identifying the thickness anomalies corresponding to the thickness value of the house surface, which can effectively prevent potential structural failure risks and extend the service life of the house. Furthermore, the present invention determines the structural safety level corresponding to the target house based on the stress critical value, which can clarify the performance of the house under current and possible future load conditions, thereby identifying potential risk points and predicting their development trends, thereby improving the accuracy and reliability of safety assessment. Therefore, the safety monitoring method and system for house status based on computer vision proposed in the present invention can improve the efficiency and accuracy of house status monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 A schematic diagram of a flow chart of a method for realizing safety monitoring of housing status based on computer vision provided by an embodiment of the present invention;

[0058] Figure 2 A schematic diagram of a module of a computer vision-based security monitoring system for housing status provided in one embodiment of the present invention.

[0059] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0060] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0061] The embodiment of the present application provides a method for implementing a security monitoring method for the state of a house based on computer vision. The execution subject of the method for implementing a security monitoring method for the state of a house based on computer vision includes but is not limited to at least one of the electronic devices such as a server, a terminal, etc. that can be configured to execute the method provided by the embodiment of the present application. In other words, the method for implementing a security monitoring method for the state of a house based on computer vision can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc. The server can be an independent server, or it can be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (Content Delivery Network, CDN), and big data and artificial intelligence platforms.

[0062] Reference Figure 1 FIG. 1 is a flow chart of a method for implementing a safety monitoring method of a house status based on computer vision according to an embodiment of the present invention. In this embodiment, the method for implementing a safety monitoring method of a house status based on computer vision includes:

[0063] S1. Acquire a panoramic image of a house corresponding to a target house, perform image enhancement on the panoramic image of the house to obtain an enhanced image of the house, perform edge corner point detection on the enhanced image of the house to obtain edge corner point features, and identify key corner points in the enhanced image of the house based on the edge corner point features.

[0064] The present invention obtains a house enhanced image by acquiring a house panoramic image corresponding to a target house and performing image enhancement on the house panoramic image. This can not only improve the clarity and detail of the image, but also effectively reduce noise and blur, making subsequent edge corner point detection and structure analysis more accurate.

[0065] Among them, the target house refers to a specific house building that needs to be monitored and evaluated for safety, which can be various types of buildings such as residential houses, commercial buildings, industrial plants, etc.; the house panoramic image refers to the actual scene structure used to display the house from various angles. Engineers in this field can obtain the house panoramic image through existing technical means (such as: using a panoramic camera for surround shooting, and synthesizing a panoramic image by stitching multiple photos from different angles; using a drone equipped with high-definition camera equipment to shoot the house at different heights and angles, and then processing and integrating the captured image data to obtain a panoramic image; using professional 3D scanning equipment to perform an all-round scan of the house, generate corresponding three-dimensional point cloud data, and then convert it into a panoramic image; setting multiple cameras at a fixed position, and obtaining the house panoramic image through synchronous shooting and post-image stitching technology); the house enhanced image refers to the image obtained after a series of image processing technologies such as contrast adjustment, brightness enhancement, denoising, etc. are performed on the original house panoramic image.

[0066] Optionally, the panoramic image of the house corresponding to the target house can be obtained by a camera device, such as a digital camera, a smart phone camera, a camera mounted on a drone, or a fixed surveillance camera; the image enhancement of the panoramic image of the house can be obtained by an image enhancement tool, such as Adobe Photoshop, GIMP, or MATLAB.

[0067] Furthermore, the present invention performs edge corner point detection on the enhanced image of the house to obtain edge corner point features, thereby extracting key information of the boundaries and structural details of objects in the image. These edge features are the basis for understanding and analyzing the structure of the house, while ensuring that the monitoring system can focus on the most critical visual information, thereby improving the overall performance and response speed of house safety monitoring.

[0068] Among them, the edge corner point features refer to the unique information representation of the edges and corner points of objects in the house image obtained after the house enhanced image is processed by a specific algorithm. These features specifically include the turning points of the object contour in the image, the intersection points of lines, and the position points where the shape of the object changes significantly, etc. Optionally, the edge corner point detection of the house enhanced image can be obtained by an edge detection algorithm, such as: Canny edge detection, Sobel operator, Prewitt operator and other algorithms.

[0069] Furthermore, the present invention identifies key corner points in the enhanced image of the house based on the edge corner point features, can accurately capture important turning points and details of the house structure, can ensure that the monitoring system focuses on the key parts of the house, and effectively improves the accuracy and efficiency of structural analysis.

[0070] The key corner points refer to pixel points that are confirmed as real and significant corner points after a series of corner point detection and verification steps.

[0071] As an embodiment of the present invention, the identifying key corner points in the house enhanced image based on the edge corner point features includes: smoothing the edge corner point features to obtain smoothed corner point features; calculating the gradient amplitude corresponding to the smoothed corner point features; marking the corner point candidate points in the house enhanced image based on the gradient amplitude; querying the window response values ​​corresponding to the corner point candidate points; and identifying the key corner points in the house enhanced image based on the window response values.

[0072] Among them, the smoothed corner feature refers to the feature obtained after smoothing the original edge corner feature; the gradient amplitude refers to the drastic degree of pixel value change in the image, which reflects the rate of change of grayscale values ​​in the image and is used to measure the strength of the edge in the image; the corner candidate point refers to the pixel point that is preliminarily identified as a possible corner point position in the house enhanced image according to the gradient amplitude; the window response value refers to the response value calculated by a specific corner detection algorithm (such as Harris corner detection) within a local window defined around the corner candidate point.

[0073] Furthermore, the smoothing of the edge corner feature can be achieved through image processing tools, such as Adobe Photoshop, GIMP and other tools; the calculation of the gradient amplitude corresponding to the smoothed corner feature can be achieved through the following gradient algorithm; the marking of the corner candidate points in the house enhanced image can be achieved through a feature marking method, such as data mining, interactive marking and other methods; the query of the window response value corresponding to the corner candidate point can be achieved through a window detection algorithm, such as Harris corner detection, Shi-Tomasi corner detection and other algorithms; the identification of key corner points in the house enhanced image can be achieved through a corner detection algorithm, such as FAST, BRIEF, ORB and other algorithms.

[0074] As an embodiment of the present invention, the calculating the gradient amplitude corresponding to the smoothed corner feature includes:

[0075] The gradient amplitude corresponding to the smooth edge feature is calculated using the following formula:

[0076]

[0077] in, Indicates that the smooth corner feature is at the feature point The corresponding gradient amplitude is Indicates the size of the neighborhood corresponding to the smooth corner feature, and Feature points in the neighborhood relative to the central feature point The offset of Represents the weight matrix, which is used to weight the feature points in the neighborhood. Indicates that the smooth corner feature is at the feature point The intensity value at Indicates that the smooth corner feature is at the feature point The intensity value at .

[0078] S2. Based on the key corner points, a structural analysis is performed on the enhanced image of the house to obtain structural features of the house, and based on the structural features of the house, a three-dimensional construction is performed on the surface of the target house to obtain a three-dimensional house model.

[0079] The present invention performs structural analysis on the enhanced image of the house based on the key corner points to obtain the structural features of the house, and can accurately extract the structural features of the house, thereby improving the accuracy and reliability of the monitoring system and enhancing the ability to identify potential risks of the house, thereby ensuring the effectiveness and practicality of house safety monitoring.

[0080] Among them, the structural features of the house refer to a series of characteristics and descriptions about the building structure of the house obtained by analyzing the enhanced image of the house based on key corner points, which covers the geometric shape information of the house, such as the overall outline of the house, the layout and size of each room; the characteristics of the structural components, including the position, thickness and connection method of beams and columns, the thickness, direction and material distribution of the walls; and the mechanical characteristics of the house structure, such as the force transmission path, possible weak points and the distribution of bearing capacity, etc. Optionally, the structural analysis of the enhanced image of the house can be obtained through structural analysis methods, such as static analysis, dynamic analysis, finite element analysis and other methods.

[0081] Furthermore, based on the structural features of the house, the present invention constructs the house surface of the target house in three dimensions to obtain a three-dimensional house model, which not only enhances the depth and breadth of house safety monitoring, but also enables potential safety hazards such as cracks, deformation or material degradation to be intuitively displayed, thereby providing strong data support for maintenance decisions.

[0082] The three-dimensional model of the house refers to a digital model constructed based on point cloud feature data that can comprehensively and three-dimensionally present the surface shape, structure and details of the target house.

[0083] As an embodiment of the present invention, the house surface of the target house is three-dimensionally constructed based on the house structure characteristics to obtain a three-dimensional house model, including: performing point cloud acquisition on the house surface of the target house to obtain point cloud data; filtering the point cloud data to obtain a filtered point cloud; performing feature mapping on the filtered point cloud and the house structure characteristics to obtain point cloud feature data; and constructing a three-dimensional house model corresponding to the house surface of the target house based on the point cloud feature data.

[0084] Among them, the point cloud data refers to a collection of a large number of spatial points obtained by collecting data on the surface of the target house, and each point contains its coordinate position in three-dimensional space and possible other attribute information, such as color, reflection intensity, etc.; the filtered point cloud refers to the point cloud data after filtering, by removing interference information such as noise points and outliers; the point cloud feature data refers to the data obtained after associating and mapping the filtered point cloud with the structural features of the house, which contains key information that can describe the surface characteristics and structure of the house.

[0085] Furthermore, the point cloud collection of the surface of the target house can be obtained through point cloud collection tools, such as laser scanners, structured light scanners and other tools; the filtering of the point cloud data can be obtained through filtering algorithms, such as SOR, ROR, MLS and other algorithms; the feature mapping of the filtered point cloud and the house structure features can be obtained through mapping algorithms, such as UV mapping, spherical mapping and other algorithms; the construction of a three-dimensional house model corresponding to the surface of the target house can be obtained through a three-dimensional modeling tool, such as AutoCAD, SketchUp and other tools.

[0086] S3. Identify the house structure facilities in the three-dimensional house model, grid the house structure facilities to obtain structural grid units, and calculate the house surface thickness value corresponding to the target house based on the structural grid units.

[0087] The present invention identifies the house structure facilities in the three-dimensional house model, grids the house structure facilities, and obtains structural grid units, which can refine the complex house structure into smaller and more manageable units, which not only helps to accurately calculate the physical properties of each unit, such as stress, deformation and material characteristics, but also improves the efficiency and accuracy of structural analysis.

[0088] Among them, the house structure facilities refer to various components and systems that constitute the house, including but not limited to walls, beams, floors, stairs, doors and windows, roof structures, and various pipes and lines; the structural grid units refer to a large number of small units into which the house structure facilities are divided according to certain rules and precisions. These units may be uniform or uneven in shape and size. Their function is to discretize the complex house structure. For example, a rectangular wall can be divided into several small rectangular or triangular grid units. Optionally, the identification of the house structure facilities in the three-dimensional house model can be achieved through a geometric measurement algorithm, such as: distance measurement, angle measurement, volume calculation and other algorithms; the grid division of the house structure facilities can be achieved through a grid division algorithm, such as: quadrilateral grid division, Delaunay triangulation, adaptive grid refinement and other algorithms.

[0089] The present invention calculates the house surface thickness value corresponding to the target house based on the structural grid unit, which not only improves the accuracy of thickness measurement, but also helps to identify potential structural defects and material losses, such as wall cracks, corrosion or wear, etc., and can also timely discover and locate areas that need repair or reinforcement, thereby ensuring the safety and durability of the house.

[0090] The building surface thickness value refers to the dimension measurement value of each surface part of the target building, such as the wall, roof, floor, etc., in the vertical direction at a specific position.

[0091] As an embodiment of the present invention, the calculating, based on the structural grid unit, the house surface thickness value corresponding to the target house includes:

[0092] The house surface thickness value corresponding to the target house is calculated using the following formula:

[0093]

[0094] in, represents the house surface thickness value corresponding to the target house, represents the index of the structural grid cell, represents the total number of structural grid cells, Indicates The actual thickness value measured by each grid unit is Indicates The area of ​​a grid cell, Indicates The weight coefficient of each grid cell.

[0095] S4. Identify the thickness anomaly points corresponding to the thickness values ​​of the housing surface, determine the abnormal areas corresponding to the thickness anomaly points, perform risk marking on the abnormal areas to obtain risk marked areas, perform regional force analysis on the risk marked areas to obtain regional force data, and calculate the critical force value corresponding to the regional force data.

[0096] The present invention helps to timely discover and evaluate the impact of these abnormal points on the overall safety of the house by identifying the abnormal thickness points corresponding to the surface thickness values ​​of the house, which can effectively prevent potential structural failure risks and extend the service life of the house.

[0097] The abnormal thickness points refer to those specific distribution points where the thickness values ​​on the surface of the house obviously deviate from the normal range and exceed or fall below the abnormal threshold, which usually indicate locations where there are structural problems or construction defects.

[0098] As an embodiment of the present invention, the identifying of thickness anomaly points corresponding to the house surface thickness values ​​includes: extracting thickness distribution points corresponding to the house surface thickness values; generating a thickness distribution map corresponding to the thickness distribution points; performing statistical analysis on the thickness distribution map to obtain statistical analysis indicators; based on the statistical analysis indicators, setting an abnormal threshold value corresponding to the target house; based on the abnormal threshold value, identifying thickness anomaly points corresponding to the house surface thickness values.

[0099] Among them, the thickness distribution point refers to the specific coordinate point in space of the thickness value corresponding to each measurement position on the surface of the house; the thickness distribution map refers to the display of these thickness distribution points in the form of a chart, usually including the horizontal axis representing the position information and the vertical axis representing the thickness value, so as to intuitively present the distribution of the thickness of the surface of the house; the statistical analysis index refers to a series of descriptive statistics obtained by performing data analysis on the thickness distribution map, such as the mean, standard deviation, median, maximum value, minimum value, etc.; the abnormal threshold refers to a boundary value set according to the statistical analysis index for judging whether the thickness is abnormal.

[0100] Furthermore, the extraction of thickness distribution points corresponding to the house surface thickness values ​​can be obtained through data visualization tools, such as Tableau, Power BI and other tools; the generation of thickness distribution maps corresponding to the thickness distribution points can be obtained through statistical analysis tools, such as SPSS, SAS and other tools; the statistical analysis of the thickness distribution map can be obtained through statistical-based methods, such as Z-score, IQR method and other methods; the setting of abnormal thresholds corresponding to the target house can be obtained through threshold setting algorithms, such as standard deviation method, percentile method and other algorithms; the identification of thickness abnormal points corresponding to the house surface thickness values ​​can be obtained through feature extraction methods, such as threshold segmentation, edge detection, morphological operations and other methods.

[0101] By determining the abnormal area corresponding to the abnormal thickness point, the present invention can accurately locate the damaged or degraded area that may exist in the house structure, thereby improving the efficiency of maintenance work, reducing unnecessary inspection and repair costs, ensuring the safety and durability of the house, and extending its service life.

[0102] Among them, the abnormal area refers to an area in the surface material of the house, which is defined by abnormal thickness change points. The material thickness in these areas deviates significantly from the normal range. Optionally, the abnormal area corresponding to the abnormal thickness change point can be obtained through a machine learning model, such as: CNN, SVM and other models.

[0103] Furthermore, the present invention can prevent the occurrence of structural failures, reduce accident risks, ensure the safety of occupants, and extend the service life of buildings by risk marking the abnormal areas to obtain risk marking areas, thereby helping to quickly identify and evaluate potential safety hazards.

[0104] The risk marking area refers to a clearly identified area in the building structure that is formed by marking abnormal areas with risk tags.

[0105] As an embodiment of the present invention, the risk marking of the abnormal area to obtain the risk marked area includes: extracting the boundary of the abnormal area to obtain area boundary points; based on the area boundary points, dividing the risk level corresponding to the abnormal area; based on the risk level, setting the risk label corresponding to the abnormal area; based on the risk label, risk marking the abnormal area to obtain the risk marked area.

[0106] Among them, the area boundary points refer to the points at the edge of the abnormal area, and these points together outline the range of the abnormal area; the risk level refers to the different levels divided according to factors such as the characteristics of the abnormal area, the degree of harm that may be caused, and the degree of impact on the overall safety of the building structure, such as high risk, medium risk, and low risk levels; the risk label refers to the specific symbols, text or codes used to identify different risk levels, so as to clearly indicate the risk level of the abnormal area.

[0107] Furthermore, the boundary extraction of the abnormal area can be achieved through a deep learning model, such as CNN, U-Net and other models; the risk level corresponding to the division of the abnormal area can be achieved through the analytic hierarchy process; the setting of the risk label corresponding to the abnormal area can be achieved through a markup language tool, such as SVG and other tools; the risk marking of the abnormal area can be achieved through a GIS tool, such as ArcGIS, QGIS and other tools.

[0108] The present invention obtains regional stress data by performing regional stress analysis on the risk marked area, thereby identifying potential structural weaknesses and stress concentration points, which helps to optimize subsequent design solutions, improve the durability and safety of the structure, and ensure the stable operation of the house under various load conditions.

[0109] The regional force data refers to the stress, strain, displacement and other data of each point in the risk area obtained through force analysis.

[0110] As an embodiment of the present invention, performing regional force analysis on the risk marked area to obtain regional force data includes: identifying force factors corresponding to the risk area; constructing a force framework corresponding to the risk area based on the force factors; performing load simulation on the risk area based on the force framework to obtain load simulation parameters; performing regional force analysis on the risk area based on the load simulation parameters to obtain regional force data.

[0111] Among them, the stress factors refer to various specific conditions that affect the stress analysis of the risk area, including the geometric shape of the risk area, material properties (such as elastic modulus, tensile strength, etc.), load types (such as static load, dynamic load, wind load, seismic load), boundary conditions (such as support method, constraint conditions) and environmental factors (such as temperature, humidity changes), etc.; the stress frame refers to the digital model established in the finite element analysis software based on the stress factors, which accurately reflects the geometric structure and material properties of the risk area; the load simulation parameters refer to the specific parameters used in load simulation, including the size, distribution, action position and action time of the load.

[0112] Furthermore, the identification of the stress factors corresponding to the risk area can be obtained through data identification tools, such as MySQL, PostgreSQL and other tools; the construction of the stress framework corresponding to the risk area can be obtained through finite element analysis tools, such as ANSYS, ABAQUS and other tools; the load simulation of the risk area can be obtained through simulation models, such as load models, dynamic response models and other models; the regional stress analysis of the risk area can be obtained through analysis algorithms, such as linear / nonlinear static analysis, modal analysis, transient dynamics analysis and other algorithms.

[0113] The present invention helps to identify potential structural weaknesses and guide the implementation of maintenance and reinforcement measures by calculating the stress critical value corresponding to the stress data of the area, and can also predict the structural response under extreme load conditions, providing a scientific basis for preventive maintenance and ensuring the long-term stability of the house and the safety of the occupants.

[0114] The stress critical value refers to the stress threshold in a structure or material that will cause structural failure or material damage when the stress reaches a certain level.

[0115] As an embodiment of the present invention, the calculating of the critical force value corresponding to the regional force data includes:

[0116] The critical force value corresponding to the force data of the area is calculated using the following formula:

[0117]

[0118] in, Indicates the critical force value corresponding to the force data of the area, Represents the element index in the force data of the area, Indicates the number of elements in the force data of the region, Indicates the force data of the area The area of ​​an element, Indicates The material strength of each element, Indicates The force on an element, Indicates The length of the elements, Represents the index of the external load in the force data of the region, represents the amount of external load in the force data of the area, Indicates The size of the external load, Indicates The direction of external load action, Indicates The element and The matrix relationship between the external loads.

[0119] S5. Based on the stress critical value, determine the structural safety level corresponding to the target house, monitor and evaluate the target house based on the structural safety level and preset safety standards to obtain evaluation parameters, and generate a safety monitoring report corresponding to the target house based on the evaluation parameters.

[0120] The present invention determines the structural safety level corresponding to the target house based on the stress critical value, and can clarify the performance of the house under current and possible future load conditions, thereby identifying potential risk points and predicting their development trends, thereby improving the accuracy and reliability of safety assessment.

[0121] The structural safety level refers to different levels divided according to factors such as the stress condition, structural integrity and potential risks of the house, which is used to characterize the safety and reliability of the house structure.

[0122] As an embodiment of the present invention, determining the structural safety level corresponding to the target house based on the stress critical value includes: determining a standard stress range corresponding to the stress critical value; identifying the overall stress structure corresponding to the target house based on the standard stress range; analyzing the bearing ratio corresponding to the overall stress structure; and determining the structural safety level corresponding to the target house based on the bearing ratio.

[0123] Among them, the standard stress range refers to the reasonable range of forces that different structural components can withstand under normal use and expected load conditions, as specified in the house structure design and relevant building specifications; the overall stress structure refers to the system that jointly bears external loads formed by the interaction of the main load-bearing components (such as beams, columns, walls, floor slabs, etc.) in the target house; the load-bearing ratio refers to the ratio between the actual load borne in the overall stress structure of the target house and the maximum load that the structure can withstand.

[0124] Furthermore, the determination of the standard stress range corresponding to the stress critical value can be obtained through structural analysis tools, such as SAP2000, ETABS and other tools; the identification of the overall stress structure corresponding to the target house can be obtained through BIM tools, such as Revit and other tools; the analysis of the bearing ratio corresponding to the overall stress structure can be obtained through structural assessment tools, such as Dlubal RFEM, SAFE and other tools; the determination of the structural safety level corresponding to the target house can be obtained through risk assessment tools, such as RiskScape, @Risk and other tools.

[0125] The present invention monitors and evaluates the target house based on the structural safety level and preset safety standards to obtain evaluation parameters, which can timely discover and locate potential safety hazards, ensure the stable operation of the house within its design life, improve the accuracy and efficiency of safety monitoring, and effectively prevent the risk of structural failure.

[0126] Among them, the preset safety standard refers to a series of specific requirements and indicators on the safety, stability, durability and other aspects of the house structure that are pre-set according to relevant building specifications, laws and regulations, industry experience and the specific purpose of the house before monitoring and evaluating the target house; the evaluation parameters refer to specific values, indicators or characteristics that can reflect the structural condition, performance and safety of the house, which are obtained and calculated during the monitoring and evaluation of the target house, such as the inclination of the house, the deformation of the components, the strength index of the material, the vibration frequency of the structure, etc. Optionally, the monitoring and evaluation of the target house can be achieved through simulation evaluation tools, such as: Dlubal RFEM, ETABS and other tools.

[0127] Furthermore, the present invention generates a security monitoring report corresponding to the target house based on the evaluation parameters, which can quickly identify key issues and potential risks of the house, thereby formulating effective prevention and response measures. The security monitoring report not only improves communication efficiency, but also ensures accurate transmission of information.

[0128] Among them, the safety monitoring report refers to a document that records in detail the safety status of the target house structure, which is generated based on evaluation parameters and related analysis results. The report usually contains the damage parameters of the house, safety standard comparison, risk assessment, maintenance recommendations and possible future predictions. Optionally, the safety monitoring report corresponding to the target house can be obtained through report automation tools, such as LaTeX, Pandoc and other tools.

[0129] Firstly, the present invention acquires a panoramic image of a house corresponding to a target house, and performs image enhancement on the panoramic image of the house to obtain an enhanced image of the house, which can not only improve the clarity and detail of the image, but also effectively reduce noise and blur, so that subsequent edge corner point detection and structural analysis are more accurate. At the same time, the present invention performs structural analysis on the enhanced image of the house based on the key corner points to obtain the structural features of the house, and can accurately extract the structural features of the house, thereby improving the accuracy and reliability of the monitoring system, and enhancing the ability to identify potential risks of the house, thereby ensuring the effectiveness and practicality of house safety monitoring. The present invention identifies the house structural facilities in the three-dimensional house model, grids the house structural facilities, and obtains the structural Grid units can break down complex house structures into smaller, more manageable units, which not only helps to accurately calculate the physical properties of each unit, such as stress, deformation, and material properties, but also improves the efficiency and accuracy of structural analysis. The present invention helps to timely discover and evaluate the impact of these abnormal points on the overall safety of the house by identifying the thickness anomalies corresponding to the thickness value of the house surface, which can effectively prevent potential structural failure risks and extend the service life of the house. Furthermore, the present invention determines the structural safety level corresponding to the target house based on the stress critical value, which can clarify the performance of the house under current and possible future load conditions, thereby identifying potential risk points and predicting their development trends, thereby improving the accuracy and reliability of safety assessment. Therefore, the safety monitoring method and system for house status based on computer vision proposed in the present invention can improve the efficiency and accuracy of house status monitoring.

[0130] like Figure 2 , which is a functional module diagram of a computer vision-based security monitoring system for house status provided by an embodiment of the present invention.

[0131] The computer vision-based security monitoring system 200 for house status of the present invention can be installed in an electronic device. According to the functions to be implemented, the computer vision-based security monitoring system 200 for house status can include a corner point recognition module 201, a three-dimensional construction module 202, a thickness value calculation module 203, a critical value calculation module 204 and a report generation module 205. The module of the present invention can also be called a unit, which refers to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions, which are stored in the memory of the electronic device.

[0132] In this embodiment, the functions of each module / unit are as follows:

[0133] A corner point recognition module 201 is used to obtain a panoramic image of a house corresponding to a target house, perform image enhancement on the panoramic image of the house to obtain an enhanced image of the house, perform edge corner point detection on the enhanced image of the house to obtain edge corner point features, and identify key corner points in the enhanced image of the house based on the edge corner point features;

[0134] A three-dimensional construction module 202 is used to perform structural analysis on the house enhanced image based on the key corner points to obtain house structural features, and to perform three-dimensional construction on the house surface of the target house based on the house structural features to obtain a three-dimensional house model;

[0135] The thickness value calculation module 203 is used to identify the house structure facilities in the three-dimensional house model, grid the house structure facilities to obtain structural grid units, and calculate the house surface thickness value corresponding to the target house based on the structural grid units;

[0136] The critical value calculation module 204 is used to identify the thickness abnormal point corresponding to the thickness value of the housing surface, determine the abnormal area corresponding to the thickness abnormal point, perform risk marking on the abnormal area to obtain the risk marked area, perform regional force analysis on the risk marked area to obtain regional force data, and calculate the force critical value corresponding to the regional force data;

[0137] The report generation module 205 is used to determine the structural safety level corresponding to the target house based on the stress critical value, monitor and evaluate the target house based on the structural safety level and preset safety standards, obtain evaluation parameters, and generate a safety monitoring report corresponding to the target house based on the evaluation parameters.

[0138] In detail, each module described in the security monitoring system 200 for realizing the house status based on computer vision in the embodiment of the present invention adopts the same technical means as the security monitoring method for realizing the house status based on computer vision described in the accompanying drawings when used, and can produce the same technical effects, which will not be repeated here.

[0139] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A method for realizing safety monitoring of housing status based on computer vision, characterized in that: The method comprises: Acquire a house panoramic image corresponding to a target house, perform image enhancement on the house panoramic image to obtain a house enhanced image, perform edge corner point detection on the house enhanced image to obtain edge corner point features, and identify key corner points in the house enhanced image based on the edge corner point features; Based on the key corner points, a structural analysis is performed on the enhanced image of the house to obtain structural features of the house, and based on the structural features of the house, a three-dimensional construction is performed on the surface of the target house to obtain a three-dimensional house model; Identify the house structure facilities in the three-dimensional house model, mesh the house structure facilities to obtain structural mesh units, and calculate the house surface thickness value corresponding to the target house based on the structural mesh units; Identify the abnormal thickness point corresponding to the thickness value of the housing surface, determine the abnormal area corresponding to the abnormal thickness point, perform risk marking on the abnormal area to obtain the risk marked area, perform regional force analysis on the risk marked area to obtain regional force data, and calculate the critical force value corresponding to the regional force data, wherein the calculation of the critical force value corresponding to the regional force data includes: The critical force value corresponding to the force data of the area is calculated using the following formula: in, Indicates the critical force value corresponding to the force data of the area, Represents the element index in the force data of the region, Indicates the number of elements in the force data of the region, Indicates the force data of the area The area of ​​an element, Indicates The material strength of each element, Indicates The force on an element, Indicates The length of the elements, Represents the index of the external load in the force data of the region, represents the amount of external load in the force data of the area, Indicates The size of the external load, Indicates The direction of external load action, Indicates The element and The matrix relationship between the external loads; Based on the stress critical value, the structural safety level corresponding to the target house is determined; based on the structural safety level and preset safety standards, the target house is monitored and evaluated to obtain evaluation parameters; based on the evaluation parameters, a safety monitoring report corresponding to the target house is generated.

2. The method for realizing safety monitoring of house status based on computer vision according to claim 1, characterized in that: The step of identifying key corner points in the house enhanced image based on the edge corner point features includes: Smoothing the edge corner point features to obtain smoothed corner point features; Calculating the gradient amplitude corresponding to the smoothed corner feature; Based on the gradient magnitude, marking the corner point candidate points in the house enhanced image; Querying the window response value corresponding to the corner point candidate point; Based on the window response value, key corner points in the house enhanced image are identified.

3. The computer vision-based safety monitoring method for housing status according to claim 1, characterized in that: The method of constructing the surface of the target house in three dimensions based on the house structure characteristics to obtain a three-dimensional house model includes: Performing point cloud collection on the surface of the target house to obtain point cloud data; Performing filtering on the point cloud data to obtain a filtered point cloud; Perform feature mapping on the filtered point cloud and the building structure features to obtain point cloud feature data; Based on the point cloud feature data, a three-dimensional house model corresponding to the house surface of the target house is constructed.

4. The method for realizing safety monitoring of housing status based on computer vision according to claim 1, characterized in that: The calculating, based on the structural grid unit, a house surface thickness value corresponding to the target house comprises: The house surface thickness value corresponding to the target house is calculated using the following formula: in, represents the house surface thickness value corresponding to the target house, represents the index of the structural grid cell, represents the total number of structural grid cells, Indicates The actual thickness value measured by each grid cell is Indicates The area of ​​a grid cell, Indicates The weight coefficient of each grid cell.

5. The computer vision-based safety monitoring method for housing status according to claim 1, characterized in that: The identifying of the abnormal thickness point corresponding to the building surface thickness value includes: Extracting thickness distribution points corresponding to the building surface thickness values; generating a thickness distribution map corresponding to the thickness distribution point; Performing statistical analysis on the thickness distribution diagram to obtain statistical analysis indicators; Based on the statistical analysis indicator, setting an abnormal threshold corresponding to the target house; Based on the abnormal threshold, the thickness abnormal point corresponding to the building surface thickness value is identified.

6. The computer vision-based safety monitoring method for housing status according to claim 1, characterized in that: The step of risk marking the abnormal area to obtain a risk marked area includes: Extracting the boundary of the abnormal area to obtain the area boundary points; Based on the boundary points of the area, the risk level corresponding to the abnormal area is divided; Based on the risk level, setting a risk label corresponding to the abnormal area; Based on the risk label, the abnormal area is risk marked to obtain a risk marked area.

7. The method for realizing safety monitoring of house status based on computer vision according to claim 1, characterized in that: The performing regional force analysis on the risk marked area to obtain regional force data includes: Identify the stress factors corresponding to the risk marked area; Based on the stress factors, construct a stress framework corresponding to the risk marked area; Based on the force-bearing frame, load simulation is performed on the risk-marked area to obtain load simulation parameters; Based on the load simulation parameters, regional force analysis is performed on the risk marked area to obtain regional force data.

8. The computer vision-based safety monitoring method for housing status according to claim 1, characterized in that: The determining, based on the stress critical value, a structural safety level corresponding to the target house includes: Determine a standard force range corresponding to the critical force value; Based on the standard stress range, identifying the overall stress structure corresponding to the target house; Analyze the load-bearing ratio corresponding to the overall force-bearing structure; Based on the load-bearing ratio, a structural safety level corresponding to the target house is determined.

9. A computer vision-based security monitoring system for housing status, characterized in that: The system is used to implement the computer vision-based security monitoring method for housing status according to any one of claims 1 to 8, the system comprising: A corner point recognition module is used to obtain a panoramic image of a house corresponding to a target house, perform image enhancement on the panoramic image of the house to obtain an enhanced image of the house, perform edge corner point detection on the enhanced image of the house to obtain edge corner point features, and identify key corner points in the enhanced image of the house based on the edge corner point features; A three-dimensional construction module is used to perform structural analysis on the house enhanced image based on the key corner points to obtain house structural features, and based on the house structural features, perform three-dimensional construction on the house surface of the target house to obtain a three-dimensional house model; A thickness value calculation module is used to identify the house structure facilities in the three-dimensional house model, grid the house structure facilities to obtain structural grid units, and calculate the house surface thickness value corresponding to the target house based on the structural grid units; The critical value calculation module is used to identify the thickness abnormal point corresponding to the thickness value of the housing surface, determine the abnormal area corresponding to the thickness abnormal point, perform risk marking on the abnormal area to obtain the risk marked area, perform regional force analysis on the risk marked area to obtain regional force data, and calculate the critical force value corresponding to the regional force data, wherein the calculation of the critical force value corresponding to the regional force data includes: The critical force value corresponding to the force data of the area is calculated using the following formula: in, Indicates the critical force value corresponding to the force data of the area, Represents the element index in the force data of the region, Indicates the number of elements in the force data of the region, Indicates the force data of the area The area of ​​an element, Indicates The material strength of each element, Indicates The force on an element, Indicates The length of the elements, Represents the index of the external load in the force data of the region, represents the amount of external load in the force data of the area, Indicates The size of the external load, Indicates The direction of external load action, Indicates The element and The matrix relationship between the external loads; A report generation module is used to determine the structural safety level corresponding to the target house based on the stress critical value, monitor and evaluate the target house based on the structural safety level and preset safety standards, obtain evaluation parameters, and generate a safety monitoring report corresponding to the target house based on the evaluation parameters.

Images