Earthquake disaster personnel risk assessment method, system and electronic device

Abstract

The application aims to provide a seismic disaster personnel risk assessment method, system and electronic equipment, and relates to the technical field of seismic disaster personnel risk determination. According to a remote sensing image of a to-be-measured region, the application determines the structure type, function type and population of a plurality of buildings in the to-be-measured region; and according to the probability of different damage degrees of buildings of different structure types under a current seismic intensity, the application determines the number of injured and the number of dead of each building under the current seismic intensity. The application can perform seismic disaster personnel risk assessment before an earthquake occurs, and the risk assessment result is used for formulating a seismic disaster relief plan and a disaster relief plan, thereby improving the rationality and efficiency of disaster relief work after an earthquake occurs.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of earthquake disaster personnel risk determination, and particularly relates to an earthquake disaster personnel risk assessment method, system and electronic equipment. BACKGROUND

[0002] Earthquake is one of the most destructive natural disasters in the world, which has caused extremely serious damage to human life and property. So far, there have been 9 major earthquakes in the world, causing more than 200,000 deaths.

[0003] Since earthquakes cannot be prevented, all possible and effective measures should be taken, such as improving earthquake casualty assessment, to reduce high earthquake disaster risk. The recent earthquakes that have occurred all over the world have proved the importance of the rapid and reliable information that disaster planners and managers have about possible losses, expected casualties and their locations; these information is crucial for making pre-disaster public policy and planning, emergency management and recovery planning. Although many studies have considered disaster loss assessment after an earthquake occurs, little attention has been paid to pre-earthquake risk assessment. Therefore, it is necessary to develop a reliable and effective method to estimate the exposure and the number of casualties. Earthquake risk assessment models and systems provide the necessary information for disaster risk management and play a fundamental role in the entire disaster risk management process.

[0004] In recent decades, researchers around the world have developed many earthquake risk assessment models and systems, mainly divided into global and regional systems, such as GDACS, WAPMERR and Open Quake project, etc. The HAZUS method and the corresponding software are the most advanced at present. However, HAZUS is a time-consuming system that requires a large amount of and high-precision input data, and the accuracy and completeness of the input information will affect the accuracy of the disaster loss estimation. Through the summary of the current existing earthquake risk assessment models, it can be found that: there is no accurate and reliable method to quickly estimate the number and location of casualties. The accuracy of the earthquake casualty assessment model depends on the accuracy of the input data. This highlights the importance of fine-scale data for accurate estimation of earthquake casualties. In many developed countries, exposure data for earthquake casualty assessment is available at the fine scale level, but in developing countries, such data is often only available at the administrative unit level. However, within the administrative unit, the hazard-affected bodies (such as personnel) in the earthquake disaster are usually assumed to be uniformly distributed, ignoring the strong spatial differences in earthquake casualties. In addition, the exposure of hazard-affected bodies is also related to time variation: affected by human activity patterns, population density and spatial distribution have strong time dependence. SUMMARY

[0005] The application aims to provide a method and system for evaluating the risk of people in earthquake disasters and an electronic device, which can evaluate the risk of people in earthquake disasters before the occurrence of an earthquake, and the risk evaluation result is used for the formulation of earthquake disaster relief drills and disaster relief plans, thereby improving the rationality and efficiency of disaster relief work after the occurrence of an earthquake.

[0006] To achieve the above-mentioned purpose, the application provides the following solutions.

[0007] A method for evaluating the risk of people in earthquake disasters, comprising the following steps.

[0008] Obtaining a remote sensing image of a to-be-tested region;

[0009] Extracting the outlines of buildings in the remote sensing image to determine a plurality of buildings in the to-be-tested region;

[0010] Determining the height of each building in the to-be-tested region by using ArcGIS software;

[0011] Determining the number of floors of each building in the to-be-tested region according to the height of the building;

[0012] Determining the structure type of each building in the to-be-tested region according to the number of floors of the building;

[0013] Determining the function type of each building;

[0014] Determining the population of each building in the to-be-tested region according to the function type of the building and a population attraction curve;

[0015] Determining any seismic intensity as a current seismic intensity;

[0016] Determining the probability of different damage degrees of buildings of different structure types under the current seismic intensity;

[0017] Determining the injury probability and death probability of buildings of different structure types under the current seismic intensity according to the injury probability corresponding to different damage degrees of the building;

[0018] Determining the number of injured people and the number of dead people of each building under the current seismic intensity according to the population, the structure type, the injury probability and the death probability of the same building.

[0019] Optionally, after the determination of the number of injured people and the number of dead people of each building under the current seismic intensity according to the population, the structure type, the injury probability and the death probability of the same building, the method further comprises the following steps.

[0020] Updating the current seismic intensity and returning to the step of determining the probability of different damage degrees of buildings of different structure types under the current seismic intensity until all seismic intensities are traversed to determine the number of injured people and the number of dead people of each building under different seismic intensities.

[0021] Optionally, after listing the number of injuries and fatalities for each building under different earthquake intensities, include:

[0022] Based on the number of injuries and deaths in each building under different earthquake intensities, determine the disaster relief plan corresponding to different earthquake intensities;

[0023] Conduct disaster relief drills according to the disaster relief plan, and adjust the disaster relief plan based on the results of the drills.

[0024] Optionally, determining the functional type of each building includes:

[0025] The area to be tested is divided into grids;

[0026] Determine the kernel density value for each grid cell for different types of interest points; the types of interest points correspond one-to-one with the function types.

[0027] Select any grid cell as the current grid cell;

[0028] Determine the function type of the current raster as the function type corresponding to the maximum kernel density value in the current raster.

[0029] The function type of each building is determined based on the function type of the multiple grids corresponding to each building.

[0030] Optionally, the formula for calculating the population of each building in the area to be measured is:

[0031] P i,j =A i ×F i ×D i ×k j ;

[0032] In the formula, P i,j A represents the number of people in building i with function type j; i F represents the building area. i Let D be the number of floors in building i. i k is the population density of the grid cell containing building i; j This is the population correction factor for buildings of function type j.

[0033] An earthquake disaster risk assessment system for personnel includes:

[0034] The remote sensing image acquisition module is used to acquire remote sensing images of the area to be measured.

[0035] The building extraction module is used to extract building outlines from the remote sensing image and identify multiple buildings in the area to be measured.

[0036] a height determination module configured to determine the height of each building in the to-be-tested region by using ArcGIS software;

[0037] a floor number determination module configured to determine the floor number of each building in the to-be-tested region according to the height of the building;

[0038] a structure type determination module configured to determine the structure type of each building in the to-be-tested region according to the floor number of the building;

[0039] a function type determination module configured to determine the function type of each building;

[0040] a population number determination module configured to determine the population number of each building in the to-be-tested region according to the function type of the building and a population attraction curve;

[0041] a current seismic intensity determination module configured to determine any seismic intensity as the current seismic intensity;

[0042] a damage degree probability determination module configured to determine the probability of different damage degrees of the building of different structure types under the current seismic intensity;

[0043] a death probability determination module configured to determine the injury probability and the death probability of the building of different structure types under the current seismic intensity according to the injury probability corresponding to different damage degrees of the building;

[0044] a death number determination module configured to determine the number of injured and dead people in each building under the current seismic intensity according to the population number, the structure type, the injury probability and the death probability of the same building.

[0045] An electronic device includes a memory and a processor, the memory is used to store a computer program, and the processor runs the computer program to make the electronic device execute the one kind of earthquake disaster personnel risk assessment method.

[0046] Optionally, the memory is a readable storage medium.

[0047] According to the specific embodiments provided by the present application, the following technical effects are provided:

[0048] The application aims to provide a method and system for evaluating the risk of people in earthquake disasters and an electronic device. The method comprises the following steps: determining the structure type, function type and population of a plurality of buildings in a to-be-tested region according to a remote sensing image of the to-be-tested region; and determining the number of injured and dead people of each building under a current seismic intensity according to the probability of different damage degrees of buildings of different structure types under the current seismic intensity. The method can evaluate the risk of people in earthquake disasters before the earthquake occurs. The risk evaluation result is used for earthquake disaster relief drills and the development of disaster relief plans, thereby improving the rationality and efficiency of disaster relief work after the earthquake occurs. BRIEF DESCRIPTION OF DRAWINGS

[0049] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments will be briefly introduced below. Obviously, the drawings in the following description only constitute some embodiments of the present application, and other drawings can be obtained by those skilled in the art without any creative labor based on these drawings.

[0050] Figure 1 The flow chart of the method for evaluating the risk of people in earthquake disasters in the embodiment 1 of the present application;

[0051] Figure 2 The spatial position diagram of the research region in the embodiment 2 of the present application;

[0052] Figure 3 The flow chart of the building information extraction method in the embodiment 2 of the present application;

[0053] Figure 4 The spatio-temporal distribution model of the population in the embodiment 2 of the present application;

[0054] Figure 5 The personnel risk evaluation model in the embodiment 2 of the present application;

[0055] Figure 6 The spatial relationship between the building and the SIPD (SIPD) in the embodiment 2 of the present application;

[0056] Figure 7 The S / B building remote sensing image in the embodiment 2 of the present application;

[0057] Figure 8 The B / C building remote sensing image in the embodiment 2 of the present application;

[0058] Figure 9 The R / C building remote sensing image in the embodiment 2 of the present application;

[0059] Figure 10 The POI reclassification diagram in the embodiment 2 of the present application;

[0060] Figure 11 Flow chart of spatial function quantitative identification algorithm in embodiment 2 of the present application;

[0061] Figure 12 Schematic diagram of population time attraction law curve in embodiment 2 of the present application;

[0062] Figure 13 Schematic diagram of building number identification result in embodiment 2 of the present application;

[0063] Figure 14 Schematic diagram of structure type identification result in embodiment 2 of the present application;

[0064] Figure 15 Male county residential POI core density map in embodiment 2 of the present application;

[0065] Figure 16 Male county office POI core density map in embodiment 2 of the present application;

[0066] Figure 17 Male county education POI core density map in embodiment 2 of the present application;

[0067] Figure 18 Male county medical POI core density map in embodiment 2 of the present application;

[0068] Figure 19 Male county commercial POI core density map in embodiment 2 of the present application;

[0069] Figure 20 Male county tourism POI core density map in embodiment 2 of the present application;

[0070] Figure 21 Schematic diagram of male county region spatial function area quantitative identification result in embodiment 2 of the present application;

[0071] Figure 22 Schematic diagram of building function type identification result in embodiment 2 of the present application;

[0072] Figure 23 Schematic diagram of building 0:00-06:59 population spatiotemporal distribution in embodiment 2 of the present application;

[0073] Figure 24 Schematic diagram of building 0:00-06:59 population spatiotemporal distribution in embodiment 2 of the present application. DETAILED DESCRIPTION

[0074] With reference to the accompanying drawings: clearly and completely describe the technical solutions in the embodiments of the present application, obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative labor fall within the scope of the present application.

[0075] The purpose of the present application is to provide a seismic disaster personnel risk assessment method, system and electronic equipment, which can assess the risk of personnel in a seismic disaster before the earthquake occurs. The risk assessment result is used for the formulation of earthquake disaster relief drills and disaster relief plans, thereby improving the rationality and efficiency of disaster relief work after the earthquake.

[0076] In order to make the above-mentioned purposes, characteristics and advantages of the present application more apparent and easy to understand, the present application will be further described in detail below with reference to the drawings and specific embodiments.

[0077] Embodiment 1

[0078] As shown in the figure, the present embodiment provides a seismic disaster personnel risk assessment method, comprising: Figure 1

[0079] Step 101: Obtain a remote sensing image of the region to be measured.

[0080] Step 102: Extract the building contour of the remote sensing image to determine a plurality of buildings in the region to be measured.

[0081] Step 103: Determine the height of each building in the region to be measured using ArcGIS software.

[0082] Step 104: Determine the number of floors of each building in the region to be measured according to the height of the building.

[0083] Step 105: Determine the structure type of each building in the region to be measured according to the number of floors of the building.

[0084] Step 106: Determine the functional type of each building.

[0085] Step 107: Determine the population of each building in the region to be measured according to the functional type of the building and the population attraction curve; the calculation formula of the population of each building in the region to be measured is:

[0086] P i,j =A i ×F i ×D i ×k j .

[0087] In the formula, P i,j ​represents the number of people in building i of function type j; A i represents the area of the building, F i represents the number of floors of building i, D i represents the population density of the grid where building i is located; k j represents the population correction coefficient of the building of function type j.

[0088] Step 108: Determine any seismic intensity as the current seismic intensity.

[0089] Step 109: Determine the probability of different damage levels of buildings of different structure types under the current seismic intensity.

[0090] Step 1010: According to the injury probability corresponding to the different damage levels of the building, determine the injury probability and death probability of buildings of different structure types under the current seismic intensity.

[0091] Step 1011: According to the population number, structure type, injury probability and death probability of the same building, determine the number of injured and dead people in each building under the current seismic intensity.

[0092] Step 1012: Update the current seismic intensity and return to step 109 until all seismic intensities are traversed to determine the number of injured and dead people in each building under different seismic intensities.

[0093] Step 1013: According to the number of injured and dead people in each building under different seismic intensities, determine the disaster relief plan corresponding to different seismic intensities.

[0094] Step 1014: According to the disaster relief plan, conduct disaster relief drills, and adjust the disaster relief plan according to the results of the disaster relief drills.

[0095] For example, step 106 includes:

[0096] Step 1061: Perform grid division processing on the area to be measured.

[0097] Step 1062: Determine the kernel density value of each grid for different types of interest points; the types of interest points correspond one-to-one with the function types.

[0098] Step 1063: Determine any grid as the current grid.

[0099] Step 1064: Determine the function type of the current grid as the function type corresponding to the maximum kernel density value in the current grid.

[0100] Step 1065: Determine the function type of each building according to the function types of the multiple grids corresponding to each building.

[0101] Embodiment 2

[0102] To better estimate the casualties caused by earthquakes, the embodiment proposes a building-based earthquake disaster personnel risk assessment method, which combines population exposure, building damage probability matrix, and semi-empirical method based on the relationship between building damage and casualties at a more appropriate spatio-temporal scale. Take Xiong'an New Area in Hebei Province as an example (as shown in Figure 2 ), the method is applied.

[0103] Research area data:

[0104] Research area

[0105] Xiong'an New Area (115.6-116.3°E, 38.7-39.2°N) is a national new area under the jurisdiction of Hebei Province, located in the hinterland of Beijing, Tianjin, and Baoding. As of November 2020, the permanent population of Xiong'an New Area was 1205440. This embodiment mainly takes Xiong County in Xiong'an New Area as an example to carry out earthquake disaster personnel risk assessment, and the spatial location of the study area is as shown in Figure 1 .

[0106] Table 1 Description of research data

[0107]

[0108] Research data:

[0109] High-resolution remote sensing images, building-related local knowledge (Br-LK) and road data are mainly used for building information extraction based on remote sensing images and Br-LK. Population distribution grid, time attraction rate curve and point of interest data are mainly used for population spatio-temporal distribution model based on building function type. Building damage probability matrix, building injury matrix and peak ground acceleration data are mainly used for personnel risk assessment model based on building structure type. The specific description of the data is shown in Table 1.

[0110] Research principle:

[0111] Earthquake disaster risk refers to the possible earthquake damage of an area in a certain period of time, which is the product of earthquake hazard, vulnerability of disaster-affected body and its spatial distribution (exposure). Earthquake disaster risk assessment evaluates the potential earthquake damage of disaster-affected body from the potential earthquake hazard, vulnerability of disaster-affected body and potential exposure of disaster-affected body unit in the area. Disaster risk assessment generally assesses the disaster risk of disaster-affected body by calculating the risk value and risk index of disaster loss. Disaster risk assessment methods are divided into deterministic method and probabilistic method. The embodiment considers the personnel casualty risk under a certain earthquake scenario, so the deterministic method is adopted to evaluate the personnel risk level of the study area by considering the personnel casualty under the earthquake scenario. Considering that building crushing is the main cause of personnel casualty, the embodiment focuses on the evaluation of the disaster risk of personnel in buildings. Therefore, the risk index is the personnel casualty rate under a certain earthquake scenario, and the risk value is the number of personnel casualties corresponding to different personnel casualty rates. The personnel risk assessment model of earthquake disaster proposed in the embodiment is as follows: D = P (S) x R (E, H) (1)

[0112] In the formula, D is the disaster risk value, i.e. the number of casualties that may be caused by different scenario earthquakes (related to intensity and occurrence time); P is the maximum disaster loss, S is the exposed population in the building, R is the disaster risk index, i.e. the personnel casualty rate in the building, E is the building damage condition, and H is the seismic intensity.

[0113] The calculation formula of the maximum disaster loss is as follows: P = R o (fun, t) x p x a x flo (2)

[0114] In the formula, fun is the building function type, t is the occurrence time, R o is the population in the room rate, p is the population density, a is the building floor area, and flo is the building floor number.

[0115] The calculation model of the disaster risk index is as follows: R = (1-e) x L i,j,k (3)

[0116] In the formula, e is the effective disaster reduction degree, i.e. the proportion of the number of people who can escape from the damaged building by themselves or through mutual aid after the earthquake occurs, and L i,j,k is the population casualty rate in the building, and the calculation formula is as follows:

[0117]

[0118] In the formula, j is the structure type of building i; k is the seismic intensity under a certain scenario; d s is the damage degree of the building; and DPM (d s |j, k) is the damage degree of the building with structure type j under the earthquake scenario with intensity ks the probability of a building with damage degree d s ) is the number of casualties that a building with damage degree d s may cause.

[0119] Method framework:

[0120] Building information extraction:

[0121] In earthquake disasters, building crushing is the main cause of casualties. Therefore, the structure of the building needs to be prioritized in the area where the crushed personnel are concentrated for earthquake emergency rescue work. How to accurately and quickly determine the distribution and damage degree of damaged buildings is of great significance for delineating emergency search and rescue areas and allocating disaster relief resources. Remote sensing, as a science and technology that obtains surface reflection information in a non-contact manner, can quickly collect disaster area ground images through unmanned aerial vehicles or satellites without being affected by the ground disaster. By analyzing the damage of buildings and other ground objects in the image, it can provide decision-making assistance for the determination of emergency search and rescue areas and the allocation of disaster relief resources. This embodiment uses high-resolution remote sensing images obtained from Google Earth, adopts an object-oriented classification method to extract building outline information, and extracts building height in combination with the building height extraction method based on SIPD proposed by Su et al. Through the local building knowledge base (Building-relevant Local Knowledge, Br-LK) of the study area, a mapping model of building height and floor number and a mapping model of building floor number and structure type are established. The building information extraction process is shown in Figure 3 .

[0122] Population spatiotemporal distribution model

[0123] Different regions will gradually form different functional units such as residential areas, office areas and commercial areas in the process of geographical environment change and social development, which are affected by human life rules and government management planning. Different spatial functional areas often show different population distribution rules and personnel in-room conditions at different time periods. For example, the daytime population density of office areas is usually higher than the nighttime population density, and the residential areas are the opposite. Buildings are the basic units of cities, and people's daily work and life are basically carried out in buildings. Therefore, it is of great significance to accurately describe the population distribution and the in-room condition of personnel at the time of the earthquake by determining the spatial functional zoning of the study area and identifying the building types in each functional area. This embodiment proposes a population spatio-temporal distribution model based on building functional types: the spatial function of the study area is quantitatively identified by point of interest (POI) data, and the building functional types are determined on this basis, and the population in-room rate in buildings at different time periods is determined by combining the population time attraction rate curve. On this basis, the number of people in buildings at different time scenarios is determined by combining building floor, area data and population density grid data, as shown in Figure 4 .

[0124] Personnel risk assessment model

[0125] Based on building information extraction and population spatio-temporal distribution model, personnel risk assessment is based on building structure type. Damage probability matrices (DPMS) and harm probability matrices (HPMs) are constructed to calculate casualty probability matrices (CPMs). The peak ground acceleration distribution of the study area reflects the ground motion under different return periods, which can determine the peak acceleration of strong and rare earthquakes (50-year exceedance probability 2%-3%, return period 1642-2475 years) and medium earthquakes (50-year exceedance probability 10%, return period 475 years) and small earthquakes (50-year exceedance probability 63%, return period 51 years). This embodiment mainly considers the scenario simulation of medium earthquakes to describe the seismic risk at different time scenarios. According to the population distribution in buildings at different time scenarios and CPMs, the disaster risk value is calculated, as shown in Figure 5 .

[0126] Building information extraction based on remote sensing image and Br-LK

[0127] Building contour extraction:

[0128] The remote sensing image information extraction method includes artificial visual interpretation method, supervised classification method, unsupervised classification method, decision tree classification method and object-oriented classification method. The artificial visual interpretation method classifies the ground object information through artificial interpretation and manual vectorization. This method can accurately extract the spatial information of the ground object and determine the ground object category; but the processing speed is slow, and it is suitable for small range ground object information extraction and high classification accuracy requirement. The supervised classification and unsupervised classification belong to the traditional classification method based on spectrum, and the two take the single pixel of the image as the unit, and the classification method focuses on the spectral information characteristics of the ground object and ignores the rich spatial geometric information contained in the image. It is proved by experiment that the supervised classification and unsupervised classification have poor processing effect on the high-resolution image with rich texture information. The decision tree classification method is a method of classifying ground information according to the spectral characteristics of the pixel, the spatial relationship of the ground object and other prior knowledge. Considering the difficulty of obtaining prior knowledge and the number of image spectral bands, the decision tree classification method is difficult to meet the needs of disaster emergency assessment in the embodiment. The object-oriented classification method considers the pixel and image object in the image analysis process, and combines the adjacent pixels in the identification element process to classify and identify the ground object. It is proved by the previous experiment that this method has good effect on the extraction of ground object texture information, and can effectively extract the building contour information, but due to the complex situation of the building, there are still some misclassification and missing classification phenomena.

[0129] Considering the experimental verification results, the spatial resolution and the number of bands of the remote sensing image, the object-oriented classification method based on samples is adopted in the embodiment, the edge detection operator and the merging algorithm are used to extract the ground object contour information including the building. In the image of the study area, samples are collected according to the spectral characteristics of the building, and the support vector machine (SVM) classifier is used to classify the ground object information. The correction of building extraction information includes three parts: (1) correction of building extraction result based on road. Considering that the study area belongs to the urban area, the buildings are relatively dense, by checking whether the extracted building intersects with the road, the purpose of removing part of the misclassification result can be achieved. Therefore, the road data and the building data are superimposed and analyzed, and the building elements intersecting with the road are deleted. (2) correction of building extraction result based on attribute information. Based on the attribute of the information extraction result, it is checked whether there is a building element missing in the non-building element. (3) correction of extraction result based on topological relationship. The topological relationship of the building extraction result is checked.

[0130] Building height extraction

[0131] The methods of estimating building height using high-resolution remote sensing images are divided into two kinds: one is to calculate the building height using stereo image pairs; the other is to estimate the building height using the building shadow information in a single high-resolution remote sensing image. Early studies show that the building height can be more accurately estimated using stereo image pairs. However, stereo pairs have the problem of difficult data acquisition. Currently, stereo pairs are mainly obtained through custom or purchase. When the research area is large, using this method will lead to high cost. Considering that the extraction of building height in this embodiment is to estimate the number of building floors, the accuracy requirement of the height evaluation result is not high, therefore, this embodiment estimates the height of the building by extracting the building shadow information in the single image.

[0132] There are mainly two methods for extracting building height based on building shadow information. One is to estimate the height of the building by measuring the entire length of the building shadow. This method has the simplest mathematical theoretical basis and is the earliest developed method. The other is to estimate the height of the building by measuring the building orientation and the shadow length perpendicular to the building orientation. However, the first method cannot solve the problem of incomplete or overlapping shadows due to various interferences of surrounding buildings or other ground objects. The second method improves the deficiency of the first method, but needs many complex measurement parameters. For a research area containing a large number of buildings, parameter identification needs to consume a lot of labor and time.

[0133] In view of the above problems, Su et al. proposed a building height extraction algorithm based on the imaging point distance (SIPD). SIPD is the distance between the imaging position of a point on the roof edge of a building and the imaging position of the shadow of the point. The imaging geometry of the building and its SIPD is shown in Figure 6 . In this figure, the gray cube represents the building, T represents a point on the roof edge of the building, B is the vertical projection of the T point on the ground, H is the height of the building, P is the position of the T point collected by the satellite image, and S is the projection position of the T point on the ground. The distance between P and S is the SIPD.

[0134] According to the basic geometric principles of Figure 6 and Three angle, it can be obtained that:

[0135] H = tan θ × L sb (5)

[0136] H = tan ω × L pb (6)

[0137]

[0138] In the formula, H is the height of the building, L spSIPD, ω and α are the satellite's elevation and azimuth angles, and θ and β are the sun's elevation and azimuth angles.

[0139] Based on the above formula, the calculation formula of building height can be obtained:

[0140]

[0141] In the above formula, ω and α can be obtained by querying satellite-related parameters, and θ and β can be determined by image acquisition time related parameters. Therefore, only L sp , the building height can be calculated.

[0142] When applied to a large number of buildings in a large area with rapid social and economic development, the SIPD method has strong applicability. The SIPD method does not consider the actual length of the building shadow (i.e. Figure 6 The SIPD method avoids the difficulty of identifying this point by considering that the B point often has high gray similarity with the surrounding pixels. The SIPD-based method theoretically only removes one necessary parameter. But avoids the complex process of tracking and determining the azimuth angle, normal line and direction of each building respectively, thereby saving a lot of time and labor. Under the impetus of rapid social and economic growth, it is crucial to save time and labor for large areas with a large number of buildings.

[0143] Building height and floor number mapping model

[0144] Since the structural type of the building needs to be deduced from the building floor number and area, it is necessary to accurately estimate the floor number of the building. Xiong'an New Area is a typical county in North China, and the buildings conform to the building style in the north. In addition, the urbanization rate of the study area is high, and the buildings in the city generally conform to modern building standards and specifications. In order to study the general correlation between the height and floor number of residential and public office buildings in this region, this embodiment uses Br-LK (including archives, yearbooks, building specifications and urban planning reports) to construct a mapping model of building height and floor number: buildings with an estimated height of about 3 to 4 meters have one floor, buildings with an estimated height of about 6 to 7 meters have two floors, and buildings with an estimated height of about 9 to 10 meters have three floors. According to this mapping relationship, combined with the building height extraction result, the floor number of all evaluated buildings in the study area is quickly determined.

[0145] Building floor number and structure type mapping model

[0146] To assess the relationship between the structural type and the number of floors of residential and public office buildings in the study area, the present embodiment used Br-LK (including archives, reports, books, yearbooks, statistical yearbooks, local chronicles, building codes, online data, and journals) to construct a mapping model of the number of floors and the structural type of buildings: (1) There are three main types of buildings in Xiong County: single-story brick buildings (S / B buildings), brick-concrete buildings (B / C buildings), and reinforced concrete buildings (R / C buildings). (2) The Chinese Building Seismic Design Codes TJ11-74, TJ11-78, GBJ11-89, GB50011-2001, GB50011-2010, etc. all clearly specify the maximum number of floors for various structural types of buildings according to certain seismic design requirements. For example, according to relevant building requirements, in areas with an earthquake fortification intensity of VIII (according to the Chinese Seismic Intensity Scale), the maximum height of B / C buildings is 6 floors. Buildings with more than six floors must be of the R / C type. Therefore, the maximum number of floors for B / C residential and public office buildings is determined to be six floors. (3) The purpose, age, spatial distribution, and clustering pattern of buildings can be used as references for determining the structural type of buildings. Figures 7-9 The remote sensing images of different types of buildings in the county town of Xiong County are shown, where S / B residential buildings are mainly distributed in the outskirts of the city. B / C buildings are rectangular and spaced about 20 to 25 meters apart. This type of building has two to six floors and is usually used for urban residential purposes. R / C buildings are mostly high-rise buildings (usually 10 floors or higher) located in or near the city center, most of which are used for residential purposes, and some are used for public service offices.

[0147] Population spatiotemporal distribution model based on building function type:

[0148] Building function type identification based on POI:

[0149] In terms of spatial function division methods, expert evaluation and expert survey and statistical methods are widely used, but these two methods have the problems of strong subjectivity, large workload, and low accuracy. As a type of geographic spatial big data with rich semantic information, POI contains information such as the spatial location and function type of the entity it corresponds to. The characteristics of fast update speed, low acquisition cost, and rich attribute content of POI data make it widely used in the quantitative identification of urban functional areas. Considering the actual situation of the study area, the present embodiment reclassified the POI data; used kernel density analysis to use the reclassified POI data for quantitative identification of spatial functional areas; and based on the determination of the spatial functional division of the study area, used spatial analysis methods to identify the function type of buildings.

[0150] POI data reclassification

[0151] The POI data used in this embodiment comes from the Gaode map development platform. According to the Gaode map POI classification code and city code table, the POI data is divided into 23 first-level categories (such as medical and health services), 267 second-level categories (such as specialized hospitals), and 869 third-level categories (such as ophthalmology hospitals). Because the classification method has many project categories and there is overlap in content, this embodiment excludes categories with low spatial function recognition, and selects 10 first-level categories suitable for spatial function zoning from the Gaode map POI data. According to the actual situation of the study area, it is reclassified into residential, office, education, medical, commercial and tourism categories, and the POI data reclassification is as shown in Figure 10

[0152] Kernel density analysis

[0153] Kernel density estimation (KDE) is a density analysis method that gradually transmits the intensity of the center point in the form of a curved surface. It reflects the spatial position difference of discrete points and also shows the characteristics of the influence of the center point on the surrounding area decreasing with distance. This method uses each sample point i as the center and h as the search radius. The kernel density value of all grid cells within the search range is calculated using the kernel function. The formula for kernel density analysis is as follows:

[0154]

[0155] In the formula, f is the kernel density estimation function; h is the search radius (also known as bandwidth); x, y are the coordinates of a grid cell within the search radius; x i , y i are the coordinates of sample point i; and n is the number of sample points within the search radius.

[0156] This embodiment uses the kernel density analysis tool of ArcToolbox to calculate the kernel density values of different types of POI. Because different types of POI have different data volumes, the kernel density values calculated by the kernel density analysis method for each type of POI data differ in order of magnitude. Therefore, the processing results obtained by kernel density analysis for different types of POI need to be further normalized. The formula is as follows:

[0157] Quantitative identification of spatial function area:

[0158] ​According to the first law of geography, there is a correlation between any two things, and the closer the distance between the two things, the greater the correlation. The kernel density analysis calculates the kernel density value of each grid cell for different types of POIs, which can be used as an indicator to determine the functional type of the grid cell. The quantitative identification method of spatial functional area is as follows: for each grid, compare the normalized kernel density values calculated by different types of POIs, and take the POI type corresponding to the maximum value as the functional area type of the grid; if the kernel density value of all types of POIs in a grid cell is 0, it is determined that the grid cell is a no-data cell. The algorithm flow of quantitative identification of spatial functional area is shown in Figure 11

[0159] The specific process of the spatial functional quantitative identification algorithm is as follows: (1) number the POI categories (1 to 6) to achieve the purpose of quantitative description. (2) Convert the kernel density value of category 1 POI to the first round of quantitative identification parameter value: if the kernel density value is 0, the quantitative identification parameter is equal to 0; otherwise, the quantitative identification parameter is equal to the kernel density value plus 1. (3) Compare the kernel density value of the i+1 type POI with the i round quantitative identification parameter value: if both are 0, the i+1 round output quantitative value is 0; if the decimal part of the i round quantitative identification parameter value is greater than or equal to the kernel density value of the i+1 type POI, the i+1 round quantitative identification parameter value is equal to the i round; otherwise, the i+1 round quantitative identification parameter value is equal to the sum of the number i+1 and the kernel density value of the i+1 type POI. (4) Repeat step 3 until all POI categories are compared, and output the quantitative identification parameter value. (5) Round the quantitative identification parameter, determine the POI category corresponding to the maximum kernel density value of each grid cell through the rounded parameter value, and output the quantitative identification result of spatial functional area according to the POI category.

[0160] Building type quantitative identification:

[0161] Buildings of different functional types provide corresponding places for various social activities of human beings and have different population attraction capabilities. Therefore, based on the spatial functional area identification result, this embodiment combines POI data and determines the building functional type based on spatial analysis method. Through the building type quantitative identification result, the functional type of damaged buildings is determined, which lays a foundation for subsequent population in-room situation analysis and refined assessment of personnel disaster degree.

[0162] ​The embodiment first determines the building function type by judging the function area where the building is located. For the building located in multiple function areas, the building center position is used to determine the attribution. After the preliminary identification of the function area where the building is located, the building function type is corrected according to the point number of different types of POIs contained in the building. If a building does not contain POI points, the function type of the building is consistent with the function area where the building is located. Otherwise, the function type of the building is the type of one or more POIs with the largest point number in the building. The building function type correction formula is as follows:

[0163]

[0164] In the formula, K is the building type; D is the function area type where the building is located; a[i] is the number of the ith type of POI in the building; and max(a) is the maximum value of the number of POIs of different types in the building.

[0165] Population spatiotemporal distribution:

[0166] When an earthquake occurs, the population distribution in the building is related to the casualties caused by the earthquake. Through the overlay analysis of the building and the population density grid, the population density in the building can be preliminarily determined. However, to determine the population distribution at the time of the earthquake, it is necessary to further study the personnel in the room at the time of the earthquake. Li Shujuan et al. combined statistical data and actual investigation data to give the population attraction curve of different types of buildings at different times of a day. In this embodiment, the population curve is resampled into 7 time periods (0:00-06:59, 07:00-08:59, 09:00-11:59, 12:00-13:59, 14:00-17:59, 18:00-21:59, 22:00-23:59) according to the actual situation of the study area. The values corresponding to different time periods are taken as the personnel in-room rate of the corresponding function type building. The personnel in-room rate is used to correct the indoor population density to determine the population number in the building at the time of the earthquake. The building population number calculation formula is as follows: i，j i i i j (12)

[0167] In the formula, the function type of building i is j, P i,j is the number of personnel in building i; A i is the building area, F i is the number of floors of the building, D i is the population density of the grid where the building is located; and k j is the population number correction coefficient of the jth type of building, i.e. the personnel in-room rate, k j ​​​​The value of the building structure type is as shown in Table 1. Figure 12

[0168] Personnel risk assessment model based on building structure type

[0169] Earthquake scenario simulation

[0170] According to the statistical analysis of the earthquake occurrence probability in North China, Northwest China and Southwest China, the intensity with a 63% probability of exceeding in 50 years is the most common intensity, which is taken as the first level intensity in the specification; the intensity with a 10% probability of exceeding in 50 years is the basic intensity (i.e. the design basic seismic peak acceleration in the Chinese seismic parameter zoning map), which is taken as the second level intensity in the specification; the intensity with a 2%-3% probability of exceeding in 50 years is the rare intensity, which is taken as the third level intensity in the specification. Based on the design basic seismic peak acceleration in the Chinese seismic parameter zoning map, this embodiment simulates the earthquake with a 10% probability of exceeding in 50 years.

[0171] Building vulnerability assessment

[0172] In view of the fact that this embodiment aims to analyze the building-induced personnel casualties under different seismic intensities, this embodiment uses the intensity as the seismic hazard parameter. The identification of the DPM (building damage probability) is divided into three steps: (1) the DPMs of different structural buildings under different intensities are obtained by using the DPMs based on the seismic intensity proposed by Yin Zhikun; (2) since the above DPMs do not include the intensity matrix of Ⅺ, the DPMs of the Ⅺ intensity area in the 2008 Wenchuan earthquake are used as a supplement; (3) in order to make the DPMs more consistent with the building characteristics of the study area, the matrix obtained in the first two steps is adjusted based on the relevant results of the earthquake loss estimation research in the central cities of northern China in recent years. The final DPM is shown in Table 2.

[0173] Personnel vulnerability assessment

[0174] This embodiment proposes a personnel casualty rate calculation model for buildings of different structural types:

[0175]

[0176] In the formula, the structural type of building i is j, the intensity of the simulated earthquake is k, d s is the damage degree of the building, DPM(d s |j, k) is the probability of the building with the structural type j being damaged to the degree d s under the scenario of the intensity k, DH(d s ) is the probability of the building being damaged to the degree d s ​The number of casualties that can be caused by the building, the value of DH is shown in Table 3.

[0177] Table 2 DPM of buildings with different structural types

[0178]

[0179]

[0180] Table 3 Building injury probability matrix

[0181]

[0182] Based on Table 2 and Table 3, the embodiment calculates the casualty probability matrix (CPM) as shown in Table 4.

[0183] Table 4 Casualty probability matrix

[0184]

[0185]

[0186] Research results

[0187] Building information extraction

[0188] The specific process of building contour extraction is as follows: (1) Google Earth Pro image preprocessing: using the geographic registration tool of ArcGIS software, according to the WGS-84 coordinates provided by Google Earth Pro software, the image is registered. (2) Object-oriented information extraction: realized by using the workflow tool Example Based Feature Extraction Workflow of ENVI software. In the object creation (Object creation) process, by adjusting the scale parameter of the edge operator and the fusion level of the merging algorithm, the same object is minimized under the condition of correct image segmentation; in the rule-based classification (Rule-based classification) process, the building, shadow, bare land, farmland, forest land and water body information in the image are classified, and the houses with different spectral characteristics are subdivided to ensure the accuracy of the building classification result. (3) Extract information correction: in ArcGIS software, the road network data collected by OpenStreetMap is analyzed by buffer zone according to different levels, and the buildings intersecting with the buffer zone are deleted; the computational geometry tool is used to calculate the area of the building, and the building elements with an area less than 100 square meters are deleted; the extracted results are topologically checked, and the overlapping elements are merged.

[0189] The building height is calculated by using ArcGIS software: on the basis of the building contour extraction result, the SIPD of the building feature is manually identified, the height of the building feature is calculated by using field calculator tool and formula (8). According to the mapping model of building height, floor number and structure type, the corresponding attribute of the building feature is identified. The building information extraction result is shown in FIG. 8. Figure 13 and Figure 14 The building function type is identified

[0190] In this embodiment, 4091 POI data are crawled from the Gaode map development platform. Since the data are stored in Excel table format, the batch conversion of data format is completed in the Python programming environment by using the Excel table conversion function (ExcelToTable_conversion) in the ArcPy package. In ArcMap, the table is converted into a point feature class. The reclassification operation of POI data is performed by using the merge tool of ArcToolbox. In this embodiment, a model is built in ArcGIS Model Builder to realize the reclassification operation of data.

[0191] The kernel density analysis of the reclassified POI data is realized by calling the ArcPy package in the Python programming environment to achieve the purpose of batch processing of data. Considering that the AOI (Area of Interest) area and influence range of different types of POI data are different, the search radius of the kernel density analysis is classified in this embodiment: for office, commercial and medical POI, the search radius is set to 500 meters, for residential and educational POI, the search radius is set to 1000 meters, and for tourism POI, the search radius is set to 1500 meters. In order to compare the kernel density analysis results of different types of POI, the normalization processing is performed. The kernel density map of each POI data after normalization is shown in FIG. 9. Figures 15-20

[0192] This embodiment proposes a quantitative identification algorithm for spatial functional area. In the Python programming environment, the conditional analysis, mathematical analysis and arithmetic operation functions in the ArcPy package are used to realize the algorithm. Through batch processing of the POI kernel density grid, the spatial functional partition of Xiongxian County is quantitatively identified, and the result is shown in FIG. 10. Figure 21

[0193] ​​On the basis of the quantitative identification of spatial function zones in Xiong County, the quantitative identification model is proposed to determine the building function types in Xiong County. The quantitative model is realized in the Python programming environment through the ArcPy package: (1) Convert the spatial function zone raster to vector data, and perform spatial connection with the building feature class. The matching mode is "HAVE_THEIR_CENTER_IN" (if the center of the building feature is located in a spatial function zone feature, the attribute of the spatial function zone feature is connected to the building feature class). The output building feature class contains the type field of the spatial function zone. (2) Perform spatial connection between the building feature class and the reclassified POI data in turn. Since this step is to count the number of POI points in the building feature, the connection operation is set to "JOIN_ONE_TO_MANY", and the matching mode is "CONTAINS" (if the building feature contains the POI feature, the POI feature is matched with the building feature class). (3) Through the field calculation tool, the building function type quantitative identification model proposed in this embodiment is realized. The building function type recognition result is shown in Figure 22 .

[0194] Population spatiotemporal distribution:

[0195] This embodiment proposes a population spatiotemporal distribution model based on building function types, and takes Xiong County as the research area and realizes it in ArcGIS software. The population distribution 100m grid is used to preliminarily determine the population distribution in the building through the spatial connection tool; according to the population time attraction rate curve, the population in-room rate in different time periods is determined, and the population distribution is adjusted. According to formula (12), the field calculator tool is used to calculate the number of people in the building in different time periods. The result of 0:00-06:59 is shown in Figure 23 .

[0196] Personnel risk assessment result

[0197] According to the Chinese Seismic Parameter Zoning Map (GB18306-2015), the peak ground acceleration of Xiong County is 0.1, and the corresponding seismic intensity is VII. Therefore, this embodiment simulates the VII intensity earthquake to evaluate the personnel risk value under different earthquake occurrence time. According to the personnel vulnerability assessment results (Table 4), when the intensity is VII, the population mortality rate in S / B structure buildings is 0.0028072, and the injury rate is 0.0116864; the population mortality rate in B / C structure buildings is 0.0001551, and the injury rate is 0.0007090; the population mortality rate in R / C structure buildings is 0, and the injury rate is 0.0000114. Combined with the spatio-temporal distribution structure of the population in the study area, the personnel disaster risk value under different earthquake scenarios is obtained in ArcGIS software using the field calculator tool. The personnel disaster risk value from 0:00 to 06:59 is shown in Table 5. Figure 24

[0198] This embodiment is based on the design basic peak ground acceleration specified in the Chinese Seismic Parameter Zoning Map, and simulates the 50-year return period of about 10% earthquake (i.e. the earthquake with a return period of 475 years). The personnel risk assessment results show that the maximum personnel disaster risk value of each building is 0.58, and the risk level is relatively low. Considering that the building seismic code and other seismic design specifications use the intensity value with a 50-year return period of 10%, therefore, if the urban construction of the study area follows the corresponding specifications, the personnel disaster risk of Xiong County to earthquake disaster is low.

[0199] Combined with the personnel risk assessment results and the identification results of building structure type and function type, this embodiment proposes the following research findings and suggestions: the personnel risk value of single-story brick structure buildings is large, therefore, the seismic fortification reconstruction of old buildings needs to be strengthened. The daytime population aggregation area is mainly concentrated in office buildings, educational buildings and medical buildings, therefore, personnel emergency evacuation drills need to be strengthened in the above-mentioned areas; the nighttime population aggregation area is mainly concentrated in residential buildings, for which residents can be encouraged to stockpile earthquake emergency supplies.

[0200] This embodiment proposes a building-based seismic disaster personnel risk assessment method, and takes Xiong County in Xiong'an New Area as an example to apply the research method. The research results of this embodiment are as follows:

[0201] (1) A building information extraction method based on remote sensing images and Br-LK is proposed. Using this method, the building contour, floor number and structure type and other information are extracted from remote sensing images and Br-LK in the study area.

[0202] (2) A population spatio-temporal distribution model based on building function type is proposed. Based on building information extraction, population grid data and POI data are used to obtain the population distribution in buildings at different time periods. ​

[0203] (3) A personnel risk assessment model based on building structure type is proposed. Through DPMs and HPMs, a personnel casualty probability matrix is constructed. According to the peak ground acceleration of the study area, the personnel risk value under different time scenarios is calculated.

[0204] The embodiment improves the spatial accuracy of the assessment results by conducting personnel risk assessment at the building scale; and improves the temporal accuracy of the assessment results by considering the influence of the time of occurrence on population distribution. The embodiment provides necessary information for disaster risk management and decision support for urban earthquake disaster reduction engineering construction.

[0205] Embodiment 3

[0206] In order to perform the method corresponding to the above-mentioned embodiment 1 to realize the corresponding functions and technical effects, a seismic disaster personnel risk assessment system is provided below, comprising:

[0207] A remote sensing image acquisition module is configured to acquire a remote sensing image of a to-be-measured region.

[0208] A building extraction module is configured to extract a building contour from the remote sensing image and determine a plurality of buildings in the to-be-measured region.

[0209] A height determination module is configured to determine the height of each building in the to-be-measured region by using ArcGIS software.

[0210] A floor number determination module is configured to determine the floor number of each building in the to-be-measured region according to the height of the building.

[0211] A structure type determination module is configured to determine the structure type of each building in the to-be-measured region according to the floor number of the building.

[0212] A function type determination module is configured to determine the function type of each building.

[0213] A population number determination module is configured to determine the population number of each building in the to-be-measured region according to the function type of the building and a population attraction curve.

[0214] A current seismic intensity determination module is configured to determine any seismic intensity as the current seismic intensity.

[0215] A damage degree probability determination module is configured to determine the probability of different damage degrees of buildings of different structure types under the current seismic intensity.

[0216] A death probability determination module is configured to determine the injury probability and death probability of buildings of different structure types under the current seismic intensity according to the injury probability corresponding to different damage degrees of the buildings.

[0217] The casualty number determination module is configured to determine the number of injured and the number of dead in each building under the current seismic intensity according to the population number, the structure type, the injury probability and the death probability of the same building.

[0218] Embodiment 4

[0219] The embodiment provides an electronic device, comprising a memory and a processor, the memory is used for storing a computer program, and the processor runs the computer program to make the electronic device implement the method for evaluating the risk of people in earthquake disasters in embodiment 1. The memory is a readable storage medium.

[0220] The embodiments in the specification are described in a progressive manner, and each embodiment focuses on the difference from other embodiments. The same or similar parts of each embodiment can be referred to each other. For the system disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple, and the related parts can be referred to the method part.

[0221] In the embodiment, the principle and implementation manner of the present application are described by using specific examples. The above embodiment is only used to help understand the method and core idea of the present application. Meanwhile, for the general skilled in the art, the specific implementation manner and application range can be changed according to the idea of the present application. In conclusion, the content of the specification should not be understood as the limitation of the present application.

Claims

1. A method of assessing the risk of personnel in a seismic hazard, characterized in that, The method comprises the following steps: acquiring a remote sensing image of a to-be-detected area; extracting a building contour from the remote sensing image to determine a plurality of buildings in the to-be-detected area; determining the height of each building in the to-be-detected area based on an imaging point distance building height extraction algorithm by using ArcGIS software; determining the number of floors of each building in the to-be-detected area based on the height of the building by using a building height and floor number mapping model; determining the structure type of each building in the to-be-detected area based on the number of floors of the building by using a building floor number and structure type mapping model; determining the function type of each building; determining the population of each building in the to-be-detected area based on the function type of the building and a population attraction curve; determining a current seismic intensity; determining the probability of different damage degrees of buildings of different structure types under the current seismic intensity; determining the injury probability and death probability of buildings of different structure types under the current seismic intensity based on the injury probability corresponding to the different damage degrees of the buildings; determining the number of injured people and the number of dead people of each building under the current seismic intensity based on the population, the structure type, the injury probability and the death probability of the same building; The method for determining the function type of each building comprises the following steps: performing a grid division on the to-be-detected area; determining the kernel density value of each grid for different types of POIs; the types of POIs correspond to the function types one by one; determining a current grid; determining the function type of the current grid as the function type corresponding to the maximum kernel density value in the current grid; determining the function type of each building based on the function types of the plurality of grids corresponding to each building; The method for determining the function type of the current grid comprises the following steps: for each grid, comparing the normalized kernel density values calculated by different types of POIs, and taking the POI type corresponding to the maximum value as the function type of the grid; if the kernel density values of all types of POIs in a certain grid unit are 0, it is determined that the grid unit is a no-data unit; The method for determining the function type of the current grid comprises the following steps: Step 1: numbering the POI categories; Step 2: converting the POI kernel density value of category 1 into the first round of quantitative recognition parameter value: if the kernel density value is 0, the quantitative recognition parameter is equal to 0; otherwise, the quantitative recognition parameter is equal to the kernel density value plus 1; Step 3: comparing the kernel density value of the i+1th POI with the i round of quantitative recognition parameter value: if both are 0, the i+1th round of quantitative value is 0; if the decimal part of the i round of quantitative recognition parameter value is greater than or equal to the kernel density value of the i+1th POI, the i+1th round of quantitative recognition parameter value is equal to the i round; otherwise, the i+1th round of quantitative recognition parameter value is equal to the sum of the number i+1 and the kernel density value of the i+1th POI; Step 4: repeating Step 3 until all POI categories are compared, and outputting the quantitative recognition parameter value; Step 5: rounding the quantitative identification parameters, determining the POI category corresponding to the maximum nuclear density value of each grid unit through the rounded parameter value, and outputting the quantitative identification result of the spatial function area according to the POI category; According to the function type of each building corresponding to multiple grids, the function type of each building is determined, including: The buildings in multiple function areas at the same time are determined to belong to by the center position of the building; After completing the preliminary identification of the function area where the building is located, the function type of the building is corrected according to the number of POIs of different types contained in the building: if a building does not contain POI points, its function type is consistent with the function area where it is located; Otherwise, one or more POI types with the most points in the building are taken as the function type of the building; The building function type correction formula is as follows: In the formula, K is the building type; D is the function area type where the building is located; a[i] is the number of the ith POI in the building; max(a) is the maximum value of the number of POIs of different types in the building.

2. The method of claim 1, wherein, After determining the number of injured and dead people of each building under the current seismic intensity according to the population number, structure type, injury probability and death probability of the same building, including: Updating the current seismic intensity and returning to the step "determining the probability of different damage degrees of buildings of different structure types under the current seismic intensity" until all seismic intensities are traversed to determine the number of injured and dead people of each building under different seismic intensities. 3.The method of claim 1, wherein, After determining the number of injured and dead people of each building under different seismic intensities, including: According to the number of injured and dead people of each building under different seismic intensities, determine the disaster relief plan corresponding to different seismic intensities; According to the disaster relief plan, carry out disaster relief drill, and adjust the disaster relief plan according to the disaster relief drill result.

4. The method of claim 1, wherein, The calculation formula of the population number of each building in the to-be-measured area is: P i,j = A i × F i × D i × k j ; where P i,j represents the number of people in building i with function type j; A i is the building area, F i is the number of floors of building i, D i is the population density of the grid in which building i is located; k j is the population number correction coefficient for buildings with function type j.

5. A seismic hazard personnel risk assessment system, characterized by, The earthquake disaster personnel risk assessment system applies the earthquake disaster personnel risk assessment method according to any one of claims 1-4, and the earthquake disaster personnel risk assessment system comprises: A remote sensing image acquisition module for acquiring a remote sensing image of a to-be-measured area; A building extraction module for extracting the building contour of the remote sensing image to determine a plurality of buildings in the to-be-measured area; A height determination module for determining the height of each building in the to-be-measured area by using ArcGIS software; A floor number determination module for determining the floor number of each building in the to-be-measured area according to the height of the building; A structure type determination module for determining the structure type of each building in the to-be-measured area according to the floor number of the building; A function type determination module for determining the function type of each building; A population number determination module for determining the population number of each building in the to-be-measured area according to the function type of the building and the population attraction curve; A current seismic intensity determination module for determining any seismic intensity as the current seismic intensity; A damage degree probability determination module for determining the probability of different damage degrees of buildings of different structure types under the current seismic intensity; The injury and death probability determination module is configured to determine the injury probability and the death probability of the buildings of different structure types under the current seismic intensity according to the injury probability corresponding to different damage degrees of the buildings. The injury and death number determination module is configured to determine the number of injured people and the number of dead people of each building under the current seismic intensity according to the population number, the structure type, the injury probability and the death probability of the same building.

6. An electronic device, comprising: The electronic device comprises a memory and a processor, wherein the memory is configured to store a computer program, and the processor is configured to run the computer program to enable the electronic device to perform the method for evaluating the personnel risk of earthquake disasters according to any one of claims 1 to 4.

7. The electronic device of claim 6, wherein, The memory is a readable storage medium.

Images