Method, system, electronic device and storage medium for analyzing hidden dangers of nine small places

Abstract

The application discloses a kind of nine small places dangerous hidden danger analysis method, system, electronic device and storage medium, including S1: collect the picture of the scene for nine small places, construct image feature database, and establish the multi-modal picture hidden danger identification model for hidden danger feature identification in picture;S2: construct multidimensional risk assessment index system, determine the grade coefficient of each dimension of hidden danger feature identified by multi-modal picture hidden danger identification model;S3: establish hidden danger weighted calculation model, and the integrated risk value is calculated by the weighted integration of the grade coefficient of each dimension;S4: establish dynamic grading early warning mechanism;S5: execution stage.Adopting the application can realize intelligent processing from hidden danger discovery to risk assessment, the efficiency of evaluation is improved, risk prediction accuracy is improved, especially suitable for nine small places dispersion, the characteristics of complex hidden danger type.

Description

Technical Field

[0001] This invention relates to the fields of computer vision and artificial intelligence technology, specifically to a method, system, electronic device, and storage medium for analyzing potential hazards in small locations. Background Technology

[0002] In recent years, with the rapid development of computer vision and artificial intelligence technologies, the security field has made significant progress in intelligent image analysis. Technologies such as lane occupancy recognition have been deployed and applied in large public places, and target detection algorithms have achieved high accuracy in general scenarios. Some platforms have already achieved automated archiving and basic classification functions for hazard images. However, the analysis of hazard images in the special scenario of "nine small places" (small shops, primary schools, small hospitals, etc.) still faces the following technical bottlenecks: 1. Insufficient adaptability to different scenarios Existing image analysis technologies are mostly designed for standardized industrial scenarios and are ill-suited to the complex environmental characteristics of small businesses. These businesses are sensitive to environmental interference, often exhibiting uneven lighting (e.g., shops at night) and severe obstruction (dense stacking of goods), leading to high false alarm rates in traditional image segmentation algorithms when identifying hazards such as haphazard electrical wiring and obstructed fire extinguishers. Furthermore, the lack of a dedicated feature database addresses the unique hazards of small businesses, such as those specific to "three-in-one" premises (combined accommodation / production / warehousing) and the illegal storage of flammable materials.

[0003] 2. Weak multimodal analysis capabilities The current technology system has not yet achieved deep integration of hazard images and multi-source data, the knowledge base is insufficient, the hazard judgment rules are still mainly based on human experience, and a computable knowledge graph containing standards such as the "Code for Fire Protection Design of Buildings" has not been constructed, resulting in a difference of more than 20% in the judgment results of similar hazard images in different systems.

[0004] 3. Delayed real-time response mechanism Existing systems mostly adopt a post-event analysis model, where images of potential hazards are obtained through manual inspections and then processed in batches by background algorithms. This results in a time lag of several hours to several days in hazard response. This cannot meet the real-time response needs of "nine small venues" for high-frequency, fragmented hazard incidents. Summary of the Invention

[0005] The technical objective of this invention is to address the above-mentioned shortcomings by providing a method, system, electronic device, and storage medium for analyzing hazards in small venues. It combines image recognition technology with a multi-dimensional risk calculation model and establishes a dynamic weighted system based on spatiotemporal characteristics to achieve intelligent processing from hazard discovery to risk assessment. Compared with traditional methods, it improves assessment efficiency and increases the accuracy of risk prediction, making it particularly suitable for the characteristics of small venues that are dispersed and have complex hazard types.

[0006] In a first aspect, the present invention provides a method for analyzing potential hazards in small venues, including the following steps: S1: Collect images of scenes in nine small venues, construct an image feature database, use transfer learning technology to train the multimodal model, and establish a multimodal image hazard identification model for identifying hazard features in images; S2: Construct a multi-dimensional risk assessment index system, determine the level coefficients in each dimension through the analytic hierarchy process, and determine the level coefficients of each dimension of the hazard characteristics identified by the multi-modal image hazard identification model based on the multi-dimensional risk assessment index system. S3: Establish a weighted calculation model for potential hazards, and calculate the comprehensive risk value by weighting and integrating the level coefficients of each dimension; S4: Establish a dynamic hierarchical early warning mechanism to provide hierarchical early warnings based on the comprehensive risk value; S5: During the execution phase, cameras installed at various high-risk locations in different venues capture on-site images and input them into the multimodal image hazard identification model to identify hazard features in the images. Based on the risk assessment index system, the level coefficients of the hazards in each dimension are obtained, and the comprehensive risk value is calculated. Based on the comprehensive risk value, a graded early warning is issued.

[0007] In step S1, images of scenes in nine small venues are collected to construct an image feature database, including: establishing a multi-dimensional feature map for the scenes in the nine small venues, and collecting and labeling real scene images based on the multi-dimensional feature map to construct an image feature database that at least includes scene images related to fire-fighting facilities, electrical wiring, escape routes, and the storage of hazardous materials.

[0008] In S2, a multi-dimensional risk assessment index system is established, and the level coefficients within each dimension are determined by the analytic hierarchy process (AHP). This includes at least four dimensions: hazard type coefficient, severity coefficient, regional distribution coefficient, and time accumulation coefficient. Based on the AHP, the hazard type coefficient includes at least four level coefficients: basic score for fire-related hazards, basic score for electrical hazards, basic score for structural hazards, and basic score for storage hazards. The severity coefficient includes at least three level coefficients: immediate danger, major risk, and general hazard. The regional distribution coefficient includes at least three level coefficients: key area, ordinary area, and peripheral area. The time accumulation coefficient includes at least two level coefficients: newly discovered hazards and overdue unrectified hazards.

[0009] In S3, the comprehensive risk value is calculated as R=Σ(K1_i×K2_i×K3_i×K4_i)×W_i, where W_i is the regional weight factor, i is the i-th hidden danger, K1 is the hidden danger type coefficient, K2 is the severity coefficient, K3 is the regional distribution coefficient, and K4 is the time accumulation coefficient.

[0010] Following S5, a time-series prediction model is introduced to predict future risk trends based on historical comprehensive risk value sequences.

[0011] Following S5, a dual-cycle optimization mechanism is set up, adjusting the regional weight factor every 72 hours based on newly collected data, and updating the basic score of each hazard type through expert evaluation every quarter.

[0012] In S4, a graded early warning is performed based on the comprehensive risk value, including: dividing the R value into 4 levels from high to low. The early warnings corresponding to the 4 levels from high to low are: triggering an automatic alarm and pushing it to the emergency command center, generating a rectification notice and reminding the person in charge via SMS, automatically recording and generating an inspection memo by the system, and including it in the regular monitoring scope.

[0013] Secondly, the present invention provides a hazard analysis system for small venues, comprising: Multimodal image hazard identification model building module: used to collect images of scenes in nine small venues, build an image feature database, use transfer learning technology to train the multimodal model, and build a multimodal image hazard identification model for identifying hazard features in images; Multidimensional risk assessment index system establishment module: used to construct a multidimensional risk assessment index system, determine the level coefficients in each dimension through the analytic hierarchy process, and determine the level coefficients of each dimension of the hazard characteristics identified by the multimodal image hazard identification model based on the multidimensional risk assessment index system. Hazard weighted calculation model establishment module: used to establish a hazard weighted calculation model, which integrates the level coefficients of each dimension to calculate the comprehensive risk value; Tiered early warning mechanism establishment module: used to establish a dynamic tiered early warning mechanism, used to provide tiered early warnings based on the comprehensive risk value; Execution module: During the execution phase, cameras set up at various high-risk scene locations in various venues capture on-site images and input them into the multimodal image hazard identification model to identify hazard features in the images. Based on the risk assessment index system, the level coefficient of the hazard in each dimension is obtained, and the comprehensive risk value is calculated. Based on the comprehensive risk value, a graded warning is issued.

[0014] Thirdly, the present invention provides an electronic device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-described method for analyzing hazards in nine small locations.

[0015] Fourthly, the present invention provides a computer-readable storage medium comprising a computer program, wherein when the computer program is executed by a processor, it implements the steps of the above-described method for analyzing hazards in nine small locations.

[0016] The present invention provides a method, system, electronic device, and storage medium for analyzing hazards in small venues, offering the following advantages: It significantly improves the efficiency and accuracy of identifying safety hazards in small venues through intelligent image analysis technology, solving the pain points of traditional manual inspections being time-consuming, labor-intensive, and prone to overlooking hazards; it automates the identification of typical hazards such as blocked fire exits, illegal electrical use, and flammable material storage through deep learning algorithms, improving the timeliness of hazard detection and response, and effectively preventing fires and other safety accidents; it can seamlessly integrate with existing safety supervision platforms, enabling cloud storage, intelligent classification, and visualization of hazard data, providing real-time dynamic decision support for regulatory departments. Compared to traditional methods, it can save a significant amount of manpower costs for inspections annually. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] The invention will be further described below with reference to the accompanying drawings.

[0019] Figure 1 This is a flowchart of the hazard analysis method for small venues according to Embodiment 1 of the present invention; Figure 2 This is a logical structure block diagram of the hazard analysis system for small venues according to Embodiment 2 of the present invention; Detailed Implementation

[0020] The present invention will be further described below with reference to the accompanying drawings and specific embodiments, so that those skilled in the art can better understand and implement the present invention. However, the embodiments are not intended to limit the present invention. In the absence of conflict, the embodiments of the present invention and the technical features in the embodiments can be combined with each other.

[0021] It should be understood that in the description of the embodiments of the present invention, terms such as "first" and "second" are used only for descriptive purposes and should not be construed as indicating or implying relative importance, nor as indicating or implying order. In the embodiments of the present invention, "multiple" refers to two or more.

[0022] In this embodiment of the invention, "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, B existing alone, or both A and B existing simultaneously. Furthermore, in this document, the character " / " generally indicates that the preceding and following related objects have an "or" relationship.

[0023] Definitions: "Nine Small Places" refers to a collective term for places with small-scale production and operation, dense personnel, and high fire safety risks. It is a specific concept in China's fire management field, aiming to improve the level of fire safety at the grassroots level through classified supervision. Examples include small shops, primary schools, small hospitals, and small warehouses. Example

[0024] like Figure 1 As shown in this embodiment, the method for analyzing hazards in small venues may include the following steps: S1: Collect images of scenes from nine small venues, construct an image feature database, and use transfer learning technology to train a multimodal model to establish a multimodal image hazard identification model for identifying hazard features in images. The multimodal image hazard identification model has higher recognition accuracy.

[0025] The construction of the image feature database includes establishing multi-dimensional feature maps for the scenes of nine small venues, collecting and labeling real scene images based on the multi-dimensional feature maps, and constructing an image feature database containing at least 12 types of scene images, such as fire-fighting facilities, electrical circuits, escape routes, and storage of hazardous materials.

[0026] By adapting the multimodal model to the data using transfer learning techniques, a multimodal hazard identification model was established specifically for recognizing safety hazard features in images. In constructing the image feature database, a multi-dimensional feature map was first built around the typical environments of nine types of small venues. Based on this map, real-world scene images were extensively collected and meticulously labeled, gradually constructing an image database covering no fewer than 12 key scenarios, including fire-fighting facilities, electrical wiring, escape routes, and storage of hazardous materials.

[0027] Image data of various real-world scenarios in small venues can be collected through various channels. The multimodal image hazard identification model can be fine-tuned and optimized based on specific scenarios or image distribution characteristics to continuously improve its recognition effect and robustness in practical applications.

[0028] S2: Construct a multidimensional risk assessment index system, determine the level coefficients in each dimension through the analytic hierarchy process, and determine the level coefficients of each dimension of the hazard characteristics identified by the multimodal image hazard identification model based on the multidimensional risk assessment index system.

[0029] The construction of a multi-dimensional risk assessment indicator system includes at least four dimensions: hazard type coefficient, severity coefficient, regional distribution coefficient, and time accumulation coefficient. Based on the analytic hierarchy process, the hazard type coefficient should include at least four levels: basic score for fire-related hazards, basic score for electrical hazards, basic score for structural hazards, and basic score for storage hazards. The severity coefficient should include at least three levels: immediate danger, major risk, and general hazard. The regional distribution coefficient should include at least three levels: key area, ordinary area, and peripheral area. The time accumulation coefficient should include at least two levels: newly discovered hazards and overdue unrectified hazards.

[0030] In this embodiment, a four-dimensional risk assessment index system is established, and the weight of each index in each dimension is determined by the analytic hierarchy process.

[0031] Dimension 1 - Hazard Type Coefficient (K1) includes four indicators: fire safety hazards (missing / ineffective fire extinguishers), electrical hazards (aging / overloaded wiring), structural hazards (blocked escape routes), and storage hazards (illegal storage of hazardous chemicals). Using the analytic hierarchy process (AHP), a base score of 15 points is determined for fire safety hazards, 12 points for electrical hazards, 10 points for structural hazards, and 20 points for storage hazards. The K1 score is determined based on the hazard type selection from the hazard features identified by the multimodal image hazard identification model.

[0032] Dimension 2 – Severity Coefficient (K2) comprises three indicators: Immediate Hazard (potential for direct disaster), Major Risk (potential for secondary disasters), and General Hazard (requiring rectification within a specified timeframe). Using the Analytic Hierarchy Process (AHP), the weights for Immediate Hazard, Major Risk, and General Hazard are determined to be 2.0, 1.5, and 1.0, respectively. The K2 value is selected based on the severity of the hazard identified by the multimodal image hazard identification model.

[0033] Dimension 3 - Regional Distribution Coefficient (K3) includes three indicators: key areas (escape routes / power distribution rooms), general areas (business premises), and peripheral areas (building perimeter). Using the analytic hierarchy process (AHP), the weights for key areas are determined to be 1.8, general areas (business premises) to be 1.2, and peripheral areas to be 0.8. The K3 value is selected based on the hazard location characteristics identified by the multimodal image hazard identification model.

[0034] Dimension 4 - Time Cumulative Coefficient (K4) includes two indicators: newly discovered hazards and overdue unrectified hazards. Using the analytic hierarchy process (AHP), the weight of newly discovered hazards is determined to be 1.0, and the weight of overdue unrectified hazards is (incremented by 1.2 for every 7 days overdue). The value of K4 is selected based on the hazard accumulation time among the hazard features identified by the multimodal image hazard identification model.

[0035] S3: Establish a weighted calculation model for potential hazards, and calculate the comprehensive risk value by weighting and integrating the level coefficients of each dimension.

[0036] The comprehensive risk value is calculated as R=Σ(K1_i×K2_i×K3_i×K4_i)×W_i, where W_i is the regional weight factor (0.5 for key areas, 0.3 for ordinary areas, and 0.2 for peripheral areas), i is the i-th hidden danger in the identified image (there may be multiple hidden dangers in one image), K1 is the hidden danger type coefficient, K2 is the severity coefficient, K3 is the regional distribution coefficient, and K4 is the time accumulation coefficient.

[0037] S4: Establish a dynamic hierarchical early warning mechanism to provide hierarchical early warnings based on comprehensive risk values.

[0038] In this embodiment, the R value can be divided into four levels from high to low. The corresponding early warning methods for the four levels from high to low are: triggering an automatic alarm and pushing it to the emergency command center, generating a rectification notice and sending an SMS reminder to the person in charge, automatically recording and generating an inspection memo by the system, and including it in the regular monitoring scope. The four levels from high to low can be divided into (R≥80), (50≤R<80), (30≤R<50), and (R<30).

[0039] A four-color tiered early warning system can be used for easy display and observation. Red alert (R≥80): Triggers an automatic alarm and pushes it to the emergency command center; Orange alert (50≤R<80): Generates a rectification notice and sends an SMS reminder to the person in charge; Yellow alert (30≤R<50): The system automatically records and generates an inspection memo; Blue status (R<30): Included in the regular monitoring scope.

[0040] S5: During the execution phase, cameras installed at various high-risk locations in different venues capture on-site footage and input it into a multimodal image hazard identification model to identify hazard features in the footage. Based on the risk assessment index system, the level coefficients of the hazards in each dimension are obtained, and the comprehensive risk value is calculated. Based on the comprehensive risk value, a graded warning is issued.

[0041] The camera captures real-time footage of the scene and inputs the captured images into a multimodal image hazard identification model at set intervals. In this embodiment, the input can be performed every 10 minutes. After intelligent image analysis and calculation, the model is evaluated according to the comprehensive risk classification standard for small venues, thereby obtaining a hazard score and early warning for the image, which facilitates timely rectification of potential hazards.

[0042] To predict the risk situation, a time-series forecasting model is introduced after S5. Based on the historical comprehensive risk value sequence, it predicts future risk trends. An LSTM time-series forecasting model is introduced, and a forecasting function is established using historical hazard data: P(t+1)=α×R(t)+β×ΔR(t)+γ×E(t), where: α=0.6 (current risk weight), β=0.3 (risk change rate weight), γ=0.1 (environmental factors: including real-time parameters such as pedestrian flow and weather), R(t) is the actual comprehensive risk value at time t, ΔR(t) is the risk change rate at time t, E(t) is the environmental factor at time t, and ΔR(t)= R(t)-R(t-1).

[0043] Following S5, a dual-loop optimization mechanism can be set up. For short-term optimization, the regional weight factors are adjusted every 72 hours based on newly collected data; for long-term optimization, the base scores for each hazard type are updated quarterly through expert evaluation. A self-learning algorithm for the regional weight factors is established: W_new = W_old + η × (P_actual - P_predicted), to achieve dynamic optimization of model parameters, where η = 0.05 is the learning rate, W_old is the old regional weight factor, P_actual is the actual comprehensive risk value, and P_predicted is the predicted risk value. Example

[0044] like Figure 2 As shown, the hazard analysis system for small venues provided in this embodiment is a system corresponding to the hazard analysis method for small venues in Embodiment 1, and includes: Multimodal image hazard identification model building module: This module is used to collect images of scenes in nine small venues, build an image feature database, and use transfer learning technology to train the multimodal model to establish a multimodal image hazard identification model for identifying hazard features in images.

[0045] Multidimensional Risk Assessment Index System Establishment Module: This module is used to construct a multidimensional risk assessment index system. It determines the level coefficients within each dimension using the analytic hierarchy process (AHP). Based on the multidimensional risk assessment index system, it determines the level coefficients of each dimension of the hazard characteristics identified by the multimodal image hazard identification model.

[0046] The hazard weighted calculation model establishment module is used to establish a hazard weighted calculation model, which integrates the level coefficients of each dimension to calculate the comprehensive risk value.

[0047] Tiered early warning mechanism establishment module: used to establish a dynamic tiered early warning mechanism, used to issue tiered early warnings based on the comprehensive risk value.

[0048] Execution module: During the execution phase, cameras set up at various high-risk scene locations in various venues capture on-site images and input them into the multimodal image hazard identification model to identify hazard features in the images. Based on the risk assessment index system, the level coefficient of the hazard in each dimension is obtained, and the comprehensive risk value is calculated. Based on the comprehensive risk value, a graded warning is issued. Example

[0049] An electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the method for analyzing hazards in nine small locations as described in Embodiment 1. Example

[0050] A computer-readable storage medium includes a computer program that, when executed by a processor, implements the steps of the method for analyzing hazards in nine small locations as described in Embodiment 1.

[0051] The method, system, electronic device, and storage medium for analyzing hazards in small venues according to the present invention have been described above by way of example with reference to the accompanying drawings. However, those skilled in the art should understand that various modifications can be made to the method, system, electronic device, and storage medium for analyzing hazards in small venues according to the present invention without departing from the scope of the invention. Therefore, the scope of protection of the present invention should be determined by the appended claims.

Claims

1. A method for analyzing potential hazards in small venues, characterized in that, Includes the following steps: S1: Collect images of scenes in nine small venues, construct an image feature database, use transfer learning technology to train the multimodal model, and establish a multimodal image hazard identification model for identifying hazard features in images; S2: Construct a multi-dimensional risk assessment index system, determine the level coefficients in each dimension through the analytic hierarchy process, and determine the level coefficients of each dimension of the hazard characteristics identified by the multi-modal image hazard identification model based on the multi-dimensional risk assessment index system. S3: Establish a weighted calculation model for potential hazards, and calculate the comprehensive risk value by weighting and integrating the level coefficients of each dimension; S4: Establish a dynamic hierarchical early warning mechanism to provide hierarchical early warnings based on the comprehensive risk value; S5: During the execution phase, cameras installed at various high-risk locations in different venues capture on-site images and input them into the multimodal image hazard identification model to identify hazard features in the images. Based on the risk assessment index system, the level coefficients of the hazards in each dimension are obtained, and the comprehensive risk value is calculated. Based on the comprehensive risk value, a graded early warning is issued.

2. The method for analyzing potential hazards in small venues according to claim 1, characterized in that, In step S1, images of scenes in nine small venues are collected to construct an image feature database, including: establishing a multi-dimensional feature map for the scenes in the nine small venues, and collecting and labeling real scene images based on the multi-dimensional feature map to construct an image feature database that at least includes scene images related to fire-fighting facilities, electrical wiring, escape routes, and the storage of hazardous materials.

3. The method for analyzing potential hazards in small venues according to claim 1, characterized in that, In S2, a multi-dimensional risk assessment index system is established, and the level coefficients within each dimension are determined by the analytic hierarchy process (AHP). This includes at least four dimensions: hazard type coefficient, severity coefficient, regional distribution coefficient, and time accumulation coefficient. Based on the AHP, the hazard type coefficient includes at least four level coefficients: basic score for fire-related hazards, basic score for electrical hazards, basic score for structural hazards, and basic score for storage hazards. The severity coefficient includes at least three level coefficients: immediate danger, major risk, and general hazard. The regional distribution coefficient includes at least three level coefficients: key area, ordinary area, and peripheral area. The time accumulation coefficient includes at least two level coefficients: newly discovered hazards and overdue unrectified hazards.

4. The method for analyzing potential hazards in small venues according to claim 3, characterized in that, In S3, the comprehensive risk value is calculated as R=Σ(K1_i×K2_i×K3_i×K4_i)×W_i, where W_i is the regional weight factor, i is the i-th hidden danger, K1 is the hidden danger type coefficient, K2 is the severity coefficient, K3 is the regional distribution coefficient, and K4 is the time accumulation coefficient.

5. The method for analyzing potential hazards in small venues according to claim 4, characterized in that, Following S5, a time-series prediction model is introduced to predict future risk trends based on historical comprehensive risk value sequences.

6. The method for analyzing potential hazards in small venues according to claim 4, characterized in that, Following S5, a dual-cycle optimization mechanism is set up, adjusting the regional weight factor every 72 hours based on newly collected data, and updating the basic score of each hazard type through expert evaluation every quarter.

7. The method for analyzing potential hazards in small venues according to claim 1, characterized in that, In S4, a graded early warning is performed based on the comprehensive risk value, including: dividing the R value into 4 levels from high to low. The early warnings corresponding to the 4 levels from high to low are: triggering an automatic alarm and pushing it to the emergency command center, generating a rectification notice and reminding the person in charge via SMS, automatically recording and generating an inspection memo by the system, and including it in the regular monitoring scope.

8. A hazard analysis system for small venues, characterized in that, include: Multimodal image hazard identification model building module: used to collect images of scenes in nine small venues, build an image feature database, use transfer learning technology to train the multimodal model, and build a multimodal image hazard identification model for identifying hazard features in images; Multidimensional risk assessment index system establishment module: used to construct a multidimensional risk assessment index system, determine the level coefficients in each dimension through the analytic hierarchy process, and determine the level coefficients of each dimension of the hazard characteristics identified by the multimodal image hazard identification model based on the multidimensional risk assessment index system. Hazard weighted calculation model establishment module: used to establish a hazard weighted calculation model, which integrates the level coefficients of each dimension to calculate the comprehensive risk value; Tiered early warning mechanism establishment module: used to establish a dynamic tiered early warning mechanism, used to provide tiered early warnings based on the comprehensive risk value; Execution module: During the execution phase, cameras set up at various high-risk scene locations in various venues capture on-site images and input them into the multimodal image hazard identification model to identify hazard features in the images. Based on the risk assessment index system, the level coefficient of the hazard in each dimension is obtained, and the comprehensive risk value is calculated. Based on the comprehensive risk value, a graded warning is issued.

9. An electronic device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, when the processor executes the computer program, it implements the steps of the method for analyzing hazards in small locations as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a computer program that, when executed by a processor, implements the steps of the method for analyzing hazards in small locations as described in any one of claims 1 to 7.

Images