Method and device for recognizing dangerous areas based on machine vision

Abstract

The application provides a dangerous area recognition method and device based on machine vision, and relates to the technical field of dangerous area recognition.The application obtains monitoring image data of a danger source in a historical monitoring area, forms a training image set through preprocessing, constructs a danger recognition model, and trains the danger recognition model, performs danger source recognition on real-time monitoring image data according to the trained danger recognition model, demarcates a basic danger area according to a basic safety distance set according to the recognition result, calculates a danger coefficient of each type of static area according to historical data of each type of static area, adds an early warning distance, calculates a dynamic danger coefficient according to historical data of each type of dynamic equipment, calculates a buffer distance in front of a moving direction, obtains historical danger accident data of a to-be-recognized area, extracts an equipment feature set, and identifies a future safety state of a sub-area through a Bayesian prediction model.

Description

Technical Field

[0001] This invention relates to the field of hazardous area identification technology, specifically to a hazardous area identification method and apparatus based on machine vision. Background Technology

[0002] With the rapid development of industrial automation and intelligent manufacturing, production processes in factories are becoming increasingly complex and high-speed. Various types of processing machinery are emerging, material handling equipment is moving more frequently, and the working environment is becoming increasingly complex, leading to a growing number of hazardous areas within factories. This increases the risk of safety accidents. Therefore, accurately identifying hazardous areas is a key factor in ensuring worker safety, efficient equipment operation, and the smooth progress of production processes.

[0003] Currently, hazardous area identification technology in factories mainly relies on traditional safety monitoring methods, such as manual inspections and fixed surveillance cameras, which often suffer from problems such as poor real-time performance, limited coverage, and the influence of human factors. Furthermore, with the development of automation and intelligent technologies, there is an urgent need to improve the ability to identify hazardous areas through advanced technological means.

[0004] In the prior art, CN 114973140 A discloses a method and system for detecting personnel intrusion into dangerous areas based on machine vision. This method acquires monitoring images of a target construction area and inputs them into a preset dangerous target detection model to determine if a dangerous target exists within the area. If a dangerous target is found, its coordinates are acquired, and the coordinates of the dangerous area are determined based on these coordinates. The coordinates of personnel in the monitoring image are then identified and compared with the coordinates of the dangerous area to determine if a person is present within the dangerous area. If a person is present, an alarm signal is sent to the user. However, this solution only addresses static areas. During production processes, dynamic mobile devices often cause more harm to people, and this method also lacks the ability to predict dangerous areas.

[0005] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0006] The purpose of this invention is to provide a method and apparatus for identifying dangerous areas based on machine vision, so as to solve the problems mentioned in the background art.

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] A method for identifying hazardous areas based on machine vision, comprising the following steps:

[0009] Step 1: Establish a spatial coordinate system, divide the area to be identified into sub-areas, install cameras in the areas to be identified, and acquire monitoring image data of hazards in historical monitoring areas. Hazard types include static areas, dynamic equipment, and personnel. Preprocess the images to form a training image set, build a hazard identification model, and train the hazard identification model.

[0010] Step 2: Acquire real-time monitoring images, identify hazard sources using a trained hazard identification model, obtain the coordinates and area of ​​the hazard source based on the identification results, delineate basic hazard zones by setting basic safety distances, and simultaneously issue alarms for dynamic equipment and personnel that intrude into the basic hazard zones.

[0011] Step 3: Based on the historical data of each type of static area, obtain the number of times danger occurred, the area size, and the duration of danger for each type of static area. Calculate the danger coefficient for each type and increase the warning distance based on the basic safety distance and the danger coefficient.

[0012] Step 4: Based on the historical data of each type of dynamic device, continuously identify the dynamic device to obtain its historical moving speed, moving distance, and area size, calculate the dynamic hazard coefficient, and calculate the buffer distance in front of the moving direction based on the basic safety distance and the dynamic hazard coefficient.

[0013] Step 5: Obtain historical hazardous accident data for the area to be identified, including the number of hazardous accidents and the coordinates of the center location of the hazardous accidents. Obtain the running time, dynamic equipment status, coordinate location, and coordinate location and working time of the equipment operators identified in the historical data to form an equipment feature set. By establishing a Bayesian prediction model, the future safety status of the sub-area is identified.

[0014] Furthermore, the static area includes oil stain area, water stain area, maintenance area, and heat source area;

[0015] The preprocessing method is as follows: hazard sources in the monitoring image are selected by bounding boxes, which include coordinate positions and area size. Then, label information is input into the bounding boxes, which includes hazard source type and hazard source status.

[0016] The status of the hazard source includes the status of dynamic equipment and the status of personnel. The status of dynamic equipment includes running, malfunctioning, and shutting down, while the status of personnel includes working and not working.

[0017] Furthermore, the hazard identification model is based on a CNN neural network, specifically including convolutional layers, activation layers, pooling layers, and an output layer;

[0018] Convolutional layer; extracts features from the bounding boxes and labels of the input image through convolution operations. The calculation formula is as follows: in, For the input matrix, For convolution kernel, The coordinates of the output matrix. The convolution kernel in the th row and number The values ​​of the column;

[0019] Activation layer: in, The coordinates of the output matrix of the convolutional layer;

[0020] Pooling layer: in, For the input matrix, This is for max-pooling output;

[0021] Output layer: in, This is the output of the pooling layer. For the results of image recognition, As weight, This is the bias parameter.

[0022] Furthermore, the specific method for delineating the basic hazardous area is as follows:

[0023] When a static area is identified, a basic hazardous area is delineated based on the coordinates and size of the static area, and then marked as a hazardous area.

[0024] When a dynamic device is identified, a basic danger zone is delineated and marked as a danger zone based on the dynamic device's coordinates, area size, and status. When the dynamic device is in operation or malfunction, a basic danger zone is delineated and marked as a danger zone.

[0025] Furthermore, the calculation steps for increasing the warning distance based on the basic safety distance and the risk factor are as follows:

[0026] The formula for calculating the risk factor of each type of static area is as follows: in, For each type of static area, the risk factor is... The number of times danger occurs. For the first The size of the next region. For the first The duration of the danger zone.

[0027] The formula for calculating the warning distance is: in, For warning distance, Based on the safe distance, Warning distance adjustment factor, , The risk factor for each type of static area.

[0028] Furthermore, the historical movement speed and movement distance are calculated using the following methods: in, For the first The distance traveled each time, For the first The speed of movement each time, For the first Before the second move Coordinate position at time, For the first Before the second move The coordinates of the moment.

[0029] Furthermore, the specific steps for calculating the buffer distance in the direction of movement based on the basic safety distance and the dynamic hazard factor are as follows:

[0030] The formula for calculating the dynamic risk factor is: in, For dynamic risk factors, For the first The speed of movement each time, For the first The distance traveled each time, This refers to the area size of the dynamic device.

[0031] The formula for calculating the buffer distance is: in, For buffer distance, This is the dynamic distance adjustment factor. .

[0032] Furthermore, the specific steps for establishing the Bayesian-based comprehensive recognition model are as follows:

[0033] The safety status of each sub-region is calculated based on the coordinates of hazardous events in historical data, i.e., the prior probability of whether a hazardous event will occur: in, For the safety status of the sub-region The prior probability, This represents the number of dangerous events that occurred in historical data. For indicator functions, The first in historical data The types of security states, , The security status of a sub-region in historical data. This indicates a dangerous event is likely to occur. This means that no dangerous incident has occurred. For smoothing coefficients;

[0034] The formula for the indicator function is: ;

[0035] Calculate the likelihood probability of the device feature set for each security state: in, For a given safe state The observed device feature set The probability, For a given safe state The following device features were observed. The probability, The number of equipment feature types;

[0036] Calculate the posterior probability of the safe state of a sub-region using Bayes' theorem: in, In a safe state The posterior probability, It is a normalized constant;

[0037] Establish a safety state decision model: in, The final identified security status.

[0038] The present invention also provides a machine vision-based hazardous area identification device, which is used to perform the above-described machine vision-based hazardous area identification method, including:

[0039] The image acquisition module is used to acquire real-time monitoring images of the area to be identified, establish a spatial coordinate system, and obtain the number of monitoring images of the types of hazards in the historical monitoring area. The types of hazards include static areas, dynamic equipment, and personnel. The image acquisition module is used to form a training image set through preprocessing.

[0040] The hazard identification module is used to construct a hazard identification model and train the hazard identification model. The trained hazard identification model is used to identify hazard sources in real-time monitoring images. Based on the identification results, the coordinate position and area size of the hazard source are obtained. A basic hazard zone is delineated by setting a basic safety distance. At the same time, an alarm is triggered for dynamic equipment and personnel that enter the basic hazard zone.

[0041] The safety early warning module is used to obtain the number of times danger occurs in each static area, the area size, and the duration of danger based on historical data of each type of static area, calculate the danger coefficient of each type, and increase the warning distance based on the basic safety distance and the danger coefficient.

[0042] The dynamic early warning module is used to obtain the historical moving speed, moving distance and area size of the dynamic device by continuously identifying the dynamic device based on historical data of each type of dynamic device, calculate the dynamic danger coefficient, and calculate the buffer distance in the moving direction based on the basic safety distance and the current moving speed.

[0043] The regional prediction module is used to acquire historical hazardous accident data of the area to be identified, including the number of hazardous accidents and the coordinates of the center location of the hazardous accidents. It also acquires the running time, operating status, coordinate location of the identified dynamic equipment and the coordinate location and working time of the equipment operators from the historical data to form an equipment feature set. By establishing a Bayesian prediction model, it identifies the areas where hazardous events will occur in the future.

[0044] Compared with the prior art, the beneficial effects of the present invention are as follows: The present invention acquires monitoring image data of dangerous sources in historical monitoring areas, forms a training image set through preprocessing, constructs a hazard identification model, trains the hazard identification model, identifies dangerous sources in real-time monitoring images based on the trained hazard identification model, delineates basic hazard areas by setting basic safety distances based on the identification results, calculates the hazard coefficient of each type of static area based on historical data of each type, increases the warning distance, calculates the dynamic hazard coefficient based on historical data of each type of dynamic equipment, calculates the buffer distance in the direction of movement, extracts equipment feature sets by acquiring historical hazard accident data of the area to be identified, and identifies the future safety status of the sub-area by establishing a Bayesian prediction model.

[0045] This invention calculates the danger coefficient and increases the warning distance for each type of static area based on historical data, providing earlier warnings and reducing the probability of accidents. It not only effectively identifies hazards in areas where dynamic equipment is located but also reasonably increases the buffer distance based on its movement trajectory, effectively reducing the probability of accidents. By establishing a Bayesian prediction model, it identifies the future safety status of sub-areas, effectively and accurately identifying potential hazardous areas. Attached Figure Description

[0046] Figure 1 This is a schematic diagram of the overall method flow of the present invention;

[0047] Figure 2 This is a schematic diagram of the overall device structure of the present invention. Detailed Implementation

[0048] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0049] It should be noted that, unless otherwise defined, the technical or scientific terms used in this invention should have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," and similar terms used in this invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0050] Example

[0051] Please see Figure 1 The present invention provides a technical solution:

[0052] A method for identifying hazardous areas based on machine vision, comprising the following steps:

[0053] Step 1: Establish a spatial coordinate system, divide the area to be identified into sub-areas, install cameras in the areas to be identified, and acquire monitoring image data of hazards in historical monitoring areas. Hazard types include static areas, dynamic equipment, and personnel. Preprocess the images to form a training image set, build a hazard identification model, and train the hazard identification model.

[0054] Within the factory area to be identified, a coordinate system is established using the area's center point or landmark building as the origin, with east-west as the horizontal axis and north-south as the vertical axis, and the center point or landmark equipment as the origin. This makes the coordinate system more intuitive and easier to understand. People are more sensitive and clear about the spatial location of the center point or landmark equipment, which helps workers and managers quickly understand the location of the coordinate points. Defining the coordinate axes using geographical orientation (east, west, south, north) aligns with people's daily orientation habits, reducing learning and adaptation costs. This coordinate system is closer to natural orientation perception, allowing workers and managers to identify directions more quickly and improving efficiency.

[0055] Categorizing hazards into static areas, dynamic equipment, and workers helps to more clearly identify and manage different safety risks. For example, static areas such as oily or water-stained areas not only easily cause slips and falls, increasing workplace injuries, but also interfere with the normal use of equipment. Maintenance areas often have scattered tools and equipment, leading to tripping or cut risks, and may also expose workers to electrical or chemical hazards. Heat sources can cause burns or fires, especially near flammable materials; therefore, hazard identification is necessary for these areas. Dynamic equipment, such as conveyor belts, robots, and transport vehicles, can cause collisions, pinching, and other injuries due to malfunctions, misoperation, or improper maintenance; operators must pay special attention to their interactions with the equipment. Workers face human factor risks, including operational errors, lack of safety training, and psychological stress; attention must be paid to personnel behavior and safety awareness to prevent accidents. By classifying hazards, safety risks can be more effectively identified and assessed, and by delineating corresponding hazardous areas, appropriate management and emergency measures can be effectively developed.

[0056] In this embodiment, the static area includes an oil stain area, a water stain area, a maintenance area, and a heat source area;

[0057] The preprocessing method is as follows: hazard sources in the monitoring image are selected by bounding boxes, which include coordinate positions and area size. Then, label information is input into the bounding boxes, which includes hazard source type and hazard source status.

[0058] The status of the hazard source includes the status of dynamic equipment and the status of personnel. The status of dynamic equipment includes running, malfunctioning, and shutting down, while the status of personnel includes working and not working.

[0059] CNN neural networks are commonly used algorithms in deep learning image classification and recognition. They can effectively capture local feature information of images through convolution operations and gradually reduce the dimensionality of images through pooling operations. Due to the design of convolution operations and pooling layers, CNNs can recognize changes in the position of objects in images, thereby enhancing the ability to detect dangerous areas in dynamic environments. CNNs exhibit stronger feature extraction capabilities and computational efficiency when processing image data, making them an ideal choice for dangerous area identification.

[0060] In this embodiment, the danger recognition model is based on a CNN neural network, specifically including a convolutional layer, an activation layer, a pooling layer, and an output layer;

[0061] Convolutional layer; extracts features from the bounding boxes and labels of the input image through convolution operations. The calculation formula is as follows: in, For the input matrix, For convolution kernel, The coordinates of the output matrix. The convolution kernel in the th row and number The values ​​of the column;

[0062] Activation layer: in, The coordinates of the output matrix of the convolutional layer;

[0063] Pooling layer: in, For the input matrix, This is for max-pooling output;

[0064] Output layer: in, This is the output of the pooling layer. For the results of image recognition, As weight, This is the bias parameter.

[0065] The training image set is divided into two parts, training and testing, in a 7:3 ratio. The image data in the training part is used as the input to the model, and the bounding box and label information in the images are used as the output of the model. After training, the model is tested and evaluated using the image set in the testing part to evaluate the model's performance on new data beyond the training data it has seen.

[0066] Step 2: Acquire real-time monitoring images, identify hazard sources using a trained hazard identification model, obtain the coordinates and area of ​​the hazard source based on the identification results, delineate basic hazard zones by setting a basic safety distance, and simultaneously issue alarms for dynamic equipment and personnel that intrude into the basic hazard zones.

[0067] When the equipment is running, there are moving parts that may pose a risk of physical injury such as pinching or impact to people who are nearby. During the troubleshooting process, maintenance personnel may need to come into close contact with the equipment. If the equipment is not completely powered off or the residual energy is not released, there will be safety hazards. At the same time, the fault may cause the control system to fail, thereby causing the equipment to act unexpectedly or stop monitoring dangerous conditions normally.

[0068] In this embodiment, the specific method for delineating the basic hazardous area is as follows:

[0069] When a static area is identified, a basic hazardous area is delineated and marked as a hazardous area based on the coordinates and size of the static area.

[0070] When a dynamic device is identified, a basic danger zone is delineated and marked as a danger zone based on the dynamic device's coordinates, area size, and status. When the dynamic device is in operation or malfunction, a basic danger zone is delineated and marked as a danger zone.

[0071] Step 3: Based on the historical data of each type of static area, obtain the number of times danger occurred, the area size, and the duration of danger in each type of static area. Calculate the danger coefficient for each type and increase the warning distance based on the basic safety distance and the danger coefficient.

[0072] In this embodiment, the calculation steps for increasing the warning distance based on the basic safety distance and the risk factor are as follows:

[0073] The formula for calculating the risk factor of each type of static area is as follows: in, For each type of static area, the risk factor is... The number of times danger occurs. For the first The size of the next region. For the first The duration of the danger zone.

[0074] The hazard coefficient of a static area reflects the degree of risk of a dangerous event occurring in each type of static area. The higher the hazard coefficient, the more likely that a dangerous event is to occur in that type of static area. By statistically analyzing the frequency of dangerous events in static areas, high-risk areas can be identified. Larger dangerous areas affect the movement of more people and equipment, and are more likely to affect normal work. The greater the hazard, the longer the static area exists, and the higher the potential risk.

[0075] The formula for calculating the increase in warning distance based on the basic safety distance and the risk factor is as follows: in, For warning distance, Based on the safe distance, Warning distance adjustment factor, , The risk factor for each type of static area.

[0076] Increasing the warning distance appropriately can provide earlier alerts to personnel and reduce the probability of accidents. By setting a warning distance adjustment coefficient, the warning distance can be reasonably adjusted according to actual needs and the type of static area, avoiding excessive distance, which would occupy too much space and cause frequent alarms, while also avoiding insufficient warning distance, which would prevent effective warning.

[0077] Step 4: Based on the historical data of each type of dynamic device, continuously identify the dynamic device to obtain its historical moving speed, moving distance, and area size. Calculate the dynamic hazard coefficient, and based on the basic safety distance and the dynamic hazard coefficient, calculate the buffer distance in the direction of movement.

[0078] The speed of moving equipment directly affects its relative position to surrounding moving and static equipment and personnel. Higher speeds mean shorter reaction times, thus increasing the risk of potential collisions. The distance a moving device travels within a specific time period can help predict its future position and determine whether it will enter a danger zone. Furthermore, the size of the moving equipment and the space it occupies determine its impact on surrounding moving and static equipment and personnel; larger equipment may pose a greater threat to the surrounding area when moving. Therefore, combining the moving speed, the distance traveled, and the area occupied by the moving equipment can form a more comprehensive dynamic hazard coefficient, helping to more accurately assess the risk of equipment in a specific environment.

[0079] In this embodiment, the historical movement speed and movement distance are calculated as follows: in, For the first The distance traveled each time, For the first The speed of movement each time, For the first Before the second move Coordinates at time, For the first Before the second move The coordinates of the moment.

[0080] In this embodiment, the specific steps for calculating the buffer distance before the direction of movement based on the basic safety distance and the dynamic hazard factor are as follows:

[0081] The formula for calculating the dynamic risk factor is: in, For dynamic risk factors, For the first The speed of movement each time, For the first The distance traveled each time, This refers to the area size of the dynamic device.

[0082] The formula for calculating the buffer distance is: in, For buffer distance, This is the dynamic distance adjustment factor. .

[0083] The dynamic hazard coefficient reflects the degree of danger of dynamic equipment. The higher the dynamic hazard coefficient, the more likely that type of dynamic equipment is to cause dangerous events. By reasonably setting the buffer distance, sufficient safety distance can be reserved when the equipment approaches the dangerous area or affects other equipment, protecting the safety of personnel and dynamic equipment. Reasonably increasing the buffer distance can effectively reduce the probability of accidents. By setting the dynamic distance adjustment coefficient, the buffer distance can be reasonably adjusted according to actual needs and the type of dynamic equipment, avoiding excessive distance, excessive space occupation and frequent alarms, while avoiding insufficient warning distance, which cannot effectively provide buffer warning.

[0084] Step 5: Obtain historical hazardous accident data for the area to be identified, including the number of hazardous accidents and the coordinates of the center location of the hazardous accidents. Obtain the running time, dynamic equipment status, coordinate location, and coordinate location and working time of the equipment operators identified in the historical data to form an equipment feature set. By establishing a Bayesian prediction model, the future safety status of the sub-area is identified.

[0085] Bayesian models excel at handling uncertainty and noise, which is particularly important in dynamic and complex industrial environments. They leverage prior knowledge, such as historical accident data, to improve predictions. This means that even with limited data, the model can make relatively accurate judgments based on existing knowledge, and Bayesian models can simultaneously consider the impact of multiple variables on the results, reflecting the complex relationships between equipment, operators, and the environment. This comprehensiveness helps to more accurately identify future hazardous areas.

[0086] Within the factory area, historical hazardous incident data and dynamic equipment information are integrated, and a Bayesian predictive model is used to identify future safety conditions. First, historical hazardous incident data provides prior knowledge about the frequency and location of accidents, including the number of accidents and their center coordinates. This information helps the model understand which areas pose a higher risk.

[0087] The data on the dynamic equipment's operating time, status, coordinate location, and the operator's working hours constitute the equipment's feature set. These features not only reflect the potential hazards that the equipment may cause during operation but also provide information about the spatial relationship between the equipment and the operator, enhancing the ability to identify hazard factors.

[0088] Using these features, the Bayesian model calculates the prior probability of an accident and estimates the likelihood of future potential hazards based on the current state of dynamic equipment and its historical operating data. The model generates a predicted safe state for each sub-region by combining the prior probability with new observations and updating the posterior probability using Bayes' theorem.

[0089] The advantage of this method lies in its ability to handle uncertainty and combine multiple influencing factors, such as equipment status and operator behavior, to achieve dynamic prediction of dangerous areas. Finally, by outputting the posterior probability of each sub-region, decision-makers can intuitively assess the safety risks of different areas, thereby achieving the identification of future dangerous areas.

[0090] In this embodiment, the specific steps for establishing the comprehensive recognition model based on Bayes are as follows:

[0091] The safety status of each sub-region is calculated based on the coordinates of hazardous events in historical data, i.e., the prior probability of whether a hazardous event will occur: in, For the safety status of the sub-region The prior probability, This represents the number of dangerous events that occurred in historical data. For indicator functions, The first in historical data The types of security states, , The security status of a sub-region in historical data. This indicates a dangerous event is likely to occur. This means that no dangerous incident has occurred. For smoothing coefficients;

[0092] The formula for the indicator function is: ;

[0093] Calculate the likelihood probability of the device feature set for each security state: in, For a given safe state The observed device feature set The probability, For a given safe state The following device features were observed. The probability, The number of equipment feature types;

[0094] Calculate the posterior probability of the safe state of a sub-region using Bayes' theorem: in, In a safe state The posterior probability, It is a normalized constant;

[0095] Establish a safety state decision model: in, The final identified security status.

[0096] Please see Figure 2The present invention also provides a machine vision-based hazardous area identification device, which is used to perform the above-described machine vision-based hazardous area identification method, including:

[0097] The image acquisition module is used to acquire real-time monitoring images of the area to be identified, establish a spatial coordinate system, and obtain the number of monitoring images of different hazard source types in the historical monitoring area. Hazard source types include static areas, dynamic equipment, and personnel. A training image set is formed through preprocessing.

[0098] The hazard identification module is used to construct a hazard identification model and train the hazard identification model. The trained hazard identification model is used to identify hazard sources in real-time monitoring images. Based on the identification results, the coordinate position and area size of the hazard source are obtained. A basic hazard zone is delineated by setting a basic safety distance. At the same time, an alarm is triggered for dynamic equipment and personnel that enter the basic hazard zone.

[0099] The safety early warning module is used to obtain the number of times danger occurs in each static area, the area size, and the duration of danger based on historical data of each type of static area, calculate the danger coefficient of each type, and increase the warning distance based on the basic safety distance and the danger coefficient.

[0100] The dynamic early warning module is used to obtain the historical moving speed, moving distance and area size of the dynamic device by continuously identifying the dynamic device based on historical data of each type of dynamic device, calculate the dynamic danger coefficient, and calculate the buffer distance in the moving direction based on the basic safety distance and the current moving speed.

[0101] The regional prediction module is used to acquire historical hazardous accident data of the area to be identified, including the number of hazardous accidents and the coordinates of the center location of the hazardous accidents. It also acquires the running time, operating status, coordinate location of the identified dynamic equipment and the coordinate location and working time of the equipment operators in the historical data to form an equipment feature set. By establishing a Bayesian prediction model, it identifies the areas where hazardous events will occur in the future.

[0102] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0103] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented in software, the above embodiments can be implemented, in whole or in part, as a computer program product. Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution.

[0104] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0105] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A method for identifying hazardous areas based on machine vision, characterized in that, The specific steps include: Step 1: Establish a spatial coordinate system, divide the area to be identified into sub-areas, install cameras in the areas to be identified, and acquire monitoring image data of hazards in historical monitoring areas. Hazard types include static areas, dynamic equipment, and personnel. Preprocess the images to form a training image set, build a hazard identification model, and train the hazard identification model. Step 2: Acquire real-time monitoring images, identify hazard sources using a trained hazard identification model, obtain the coordinates and area of ​​the hazard source based on the identification results, delineate basic hazard zones by setting basic safety distances, and simultaneously issue alarms for dynamic equipment and personnel that intrude into the basic hazard zones. Step 3: Based on the historical data of each type of static area, obtain the number of times danger occurred, the area size, and the duration of danger for each type of static area. Calculate the danger coefficient for each type and increase the warning distance based on the basic safety distance and the danger coefficient. Step 4: Based on the historical data of each type of dynamic device, continuously identify the dynamic device to obtain its historical moving speed, moving distance, and area size, calculate the dynamic hazard coefficient, and calculate the buffer distance in front of the moving direction based on the basic safety distance and the dynamic hazard coefficient. Step 5: Obtain historical hazardous accident data for the area to be identified, including the number of hazardous accidents and the coordinates of the center location of the hazardous accidents. Obtain the running time, dynamic equipment status, coordinate location, and coordinate location and working time of the equipment operators identified in the historical data to form an equipment feature set. By establishing a Bayesian prediction model, the future safety status of the sub-area is identified. The specific method for delineating the basic hazard zone is as follows: When a static area is identified, a basic hazardous area is delineated and marked as a hazardous area based on the coordinates and size of the static area. When a dynamic device is identified, a basic hazardous area is delineated and marked as a hazardous area based on the coordinates, size, and status of the dynamic device, and when the dynamic device is in operation or malfunction. The calculation steps for increasing the warning distance based on the basic safety distance and the risk factor are as follows: The formula for calculating the hazard factor for each type of static area is as follows: in, For each type of static area, the risk factor is... The number of times danger occurs. For the first The size of the next region. For the first The duration of the secondary danger zone; The formula for calculating the warning distance is: in, For warning distance, Based on the safe distance, Warning distance adjustment factor, , The risk factor for each type of static area.

2. The method for identifying hazardous areas based on machine vision according to claim 1, characterized in that: The static area includes oil stain area, water stain area, maintenance area, and heat source area; The preprocessing method is as follows: hazard sources in the monitoring image are selected by bounding boxes, which include coordinate positions and area size. Then, label information is input into the bounding boxes, which includes hazard source type and hazard source status. The status of the hazard source includes the status of dynamic equipment and the status of personnel. The status of dynamic equipment includes running, malfunctioning, and shutting down, while the status of personnel includes working and not working.

3. The method for identifying hazardous areas based on machine vision according to claim 1, characterized in that: The hazard identification model is based on a CNN neural network, specifically including convolutional layers, activation layers, pooling layers, and an output layer; Convolutional layer: Extracts features from the bounding boxes and labels of the input image through convolution operations. The calculation formula is as follows: in, For the input matrix, For convolution kernel, The coordinates of the output matrix. The convolution kernel in the th row and number The value of the column; Activation layer: in, The coordinates of the output matrix of the convolutional layer; Pooling layer: in, For the input matrix, This is for max-pooling output; Output layer: in, This is the output of the pooling layer. For the results of image recognition, As weight, This is the bias parameter.

4. The method for identifying hazardous areas based on machine vision according to claim 1, characterized in that: The historical movement speed and movement distance are calculated using the following methods: in, For the first The distance traveled each time, For the first The speed of movement each time, For the first Before the second move Coordinates at time, For the first Before the second move The coordinates of the moment.

5. The method for identifying hazardous areas based on machine vision according to claim 4, characterized in that: The specific steps for calculating the buffer distance before the direction of movement based on the basic safety distance and dynamic hazard factor are as follows: The formula for calculating the dynamic risk factor is: in, For dynamic risk factors, For the first The speed of movement each time, For the first The distance traveled each time, The area size of the dynamic device; The formula for calculating the buffer distance is: in, For buffer distance, This is the dynamic distance adjustment factor. .

6. The method for identifying hazardous areas based on machine vision according to claim 1, characterized in that: The specific steps for establishing a comprehensive recognition model based on Bayes are as follows: The safety status of each sub-region is calculated based on the coordinates of hazardous events in historical data, i.e., the prior probability of whether a hazardous event will occur: in, For the safety status of the sub-region The prior probability, This represents the number of dangerous events that occurred in historical data. For indicator functions, The first in historical data The types of security states, , The security status of a sub-region in historical data. This indicates a dangerous event is likely to occur. This means that no dangerous incident has occurred. For smoothing coefficients; The formula for the indicator function is: ； Calculate the likelihood probability of the device feature set for each security state: in, For a given safe state The observed device feature set The probability, For a given safe state The following device features were observed. The probability, The number of equipment feature types; Calculate the posterior probability of the safe state of a sub-region using Bayes' theorem: in, In a safe state The posterior probability, It is a normalized constant; Establish a safety state decision model: in, The final identified security status.

7. A machine vision-based hazardous area identification device, characterized in that: The machine vision-based hazardous area identification device is used to execute the machine vision-based hazardous area identification method according to any one of claims 1-6, including: The image acquisition module is used to acquire real-time monitoring images of the area to be identified, establish a spatial coordinate system, and obtain the number of monitoring images of the types of hazards in the historical monitoring area. The types of hazards include static areas, dynamic equipment, and personnel. The image acquisition module is used to form a training image set through preprocessing. The hazard identification module is used to construct a hazard identification model and train the hazard identification model. The trained hazard identification model is used to identify hazard sources in real-time monitoring images. Based on the identification results, the coordinate position and area size of the hazard source are obtained. A basic hazard zone is delineated by setting a basic safety distance. At the same time, an alarm is triggered for dynamic equipment and personnel that enter the basic hazard zone. The safety early warning module is used to obtain the number of times danger occurs in each static area, the area size, and the duration of danger based on historical data of each type of static area, calculate the danger coefficient of each type, and increase the warning distance based on the basic safety distance and the danger coefficient. The dynamic early warning module is used to obtain the historical moving speed, moving distance and area size of the dynamic device by continuously identifying the dynamic device based on historical data of each type of dynamic device, calculate the dynamic danger coefficient, and calculate the buffer distance in the moving direction based on the basic safety distance and the current moving speed. The regional prediction module is used to acquire historical hazardous accident data of the area to be identified, including the number of hazardous accidents and the coordinates of the center location of the hazardous accidents. It also acquires the running time, operating status, coordinate location of the identified dynamic equipment and the coordinate location and working time of the equipment operators from the historical data to form an equipment feature set. By establishing a Bayesian prediction model, it identifies the areas where hazardous events will occur in the future.

Images