Ship hanging ladder climbing danger identification method and system

Abstract

The invention relates to a ship hanging ladder climbing danger identification method and system, and the method comprises the steps: recognizing the identity information of a target object when a collection image is received, and obtaining the monitoring data related to the identity information in real time; extracting an action trend feature of the target object and a state feature and an environment feature of the target hanging ladder; inputting the action trend characteristics, the state characteristics, the environment characteristics and the monitoring data into a danger prediction model; when the prediction information is received, triggering a corresponding response mechanism according to the danger level; according to the method, comprehensive and real-time danger identification of the ship hanging ladder climbing process is realized by combining image identification based on the motion trend of the sailors, the hanging ladder and the environment, physiological parameter monitoring of the sailors and a danger prediction model, so that the accuracy and timeliness of danger identification are improved, a corresponding response mechanism can be automatically triggered according to the danger level, and the safety of the ship climbing process is improved. Therefore, the safety risk in the climbing process is effectively reduced, and the life safety of sailors is guaranteed.

Description

Technical Field

[0001] The present application relates to the technical field of hazard prediction, and in particular to a method and system for identifying hazards of climbing a ship's suspended ladder. Background Art

[0002] In marine transportation and port operations, ship ladders are the main channel for crew members to get on and off the ship, and their safe use is of vital importance. With the development of intelligent technology, the demand for safety monitoring during the climbing process of ship ladders is increasing, aiming to use technical means to identify and prevent potential dangerous situations in advance and ensure the personal safety of crew members.

[0003] In the existing technology, the safety management of ship ladders mainly relies on manual supervision and the implementation of safety regulations, but this method has many limitations. On the one hand, manual monitoring is easily affected by factors such as fatigue and distraction, and it is difficult to maintain high-efficiency alertness; on the other hand, it is difficult to fully cover the complex and changeable working environment and the individual status differences of crew members based on experience, which makes it difficult to discover safety hazards in time. In recent years, although some technologies have attempted to monitor the use of ladders by installing surveillance cameras, most of them can only achieve basic image recording functions, and lack the ability to deeply analyze crew behavior, ladder status and surrounding environment and predict dangers.

[0004] Specifically, the defects of the prior art are mainly reflected in the following aspects:

[0005] Lack of intelligent identification and analysis: Most monitoring systems can only perform simple image capture and are unable to identify the crew's movement trends and the specific status of the ladder and environment, making it difficult to predict potential dangers;

[0006] Lack of physiological status monitoring: The physical condition of the crew (such as heart rate, blood pressure and other physiological parameters) directly affects the safety of their climbing the suspended ladder, but existing technologies often ignore the real-time monitoring and integrated analysis of this important factor. Summary of the invention

[0007] In order to solve the above-mentioned defects, the present application provides a method and system for identifying climbing hazards of ship suspended ladders.

[0008] The above-mentioned invention objective of the present application is achieved through the following technical solutions:

[0009] A method for identifying climbing hazards of a ship suspended ladder comprises the following steps:

[0010] When receiving the captured image sent by the image acquisition terminal, identifying the identity information of the target object, and acquiring the monitoring data of the physiological parameter monitoring device associated with the identity information in real time;

[0011] Extracting the motion trend features of the target object in the captured image and the state features and environmental features of the target suspension ladder;

[0012] The extracted action trend features, state features, environmental features and monitoring data are input into the pre-trained risk prediction model for risk prediction;

[0013] When the prediction information output by the danger prediction model is received, the corresponding response mechanism is triggered according to its danger level.

[0014] By adopting the above technical solution, the present application combines image recognition technology, physiological parameter monitoring and danger prediction model, aiming to provide real-time danger identification and response for the ship's suspended ladder climbing process: receiving real-time images from the image acquisition terminal, and identifying the identity information of the target object (i.e., the crew member) who is climbing the suspended ladder, associating the identity information of the crew member, obtaining the monitoring data transmitted by the physiological parameter monitoring device worn by the crew member in real time, conducting in-depth analysis of the collected images, extracting the motion trend characteristics of the target object, such as climbing speed, body posture changes, etc., and identifying the state characteristics of the ladder and the surrounding environment characteristics at the same time, and extracting the extracted motion trend characteristics, ladder state characteristics, environmental characteristics and crew member's physiological parameter monitoring data. The data are input into a pre-trained danger prediction model. Based on a large amount of historical data and machine learning algorithms, the model can predict the possibility and level of danger in the current climbing situation. When the prediction information output by the danger prediction model is received, the corresponding response mechanism is immediately triggered according to the predicted danger level. The present application realizes comprehensive and real-time danger identification in the process of climbing a ship's ladder by combining image recognition based on crew movement trends, ladders and the environment, monitoring of crew physiological parameters and danger prediction models. It not only improves the accuracy and timeliness of danger identification, but also can automatically trigger corresponding response mechanisms according to the danger level, thereby effectively reducing the safety risks in the climbing process and ensuring the lives of crew members.

[0015] In a preferred example, the present application can be further configured as follows: the step of extracting the action trend characteristics of the target object in the captured image and the state characteristics and environmental characteristics of the target hanging ladder includes the steps of:

[0016] Identify the body contour features, joint movement features, and speed change features of the target object in the captured image;

[0017] Identify the shaking amplitude characteristics and surface state characteristics of the target suspension ladder in the collected image, and identify the obstacle characteristics and weather and lighting characteristics of the environment surrounding the target suspension ladder;

[0018] Extract motion trend features based on the target object's body contour features, joint movement features, speed change features, and the swing amplitude features of the target suspended ladder;

[0019] The state characteristics and environmental characteristics of the target suspension ladder are extracted based on the surface state characteristics, shaking amplitude characteristics, obstacle characteristics and weather and lighting characteristics of the target suspension ladder's surrounding environment.

[0020] By adopting the above technical solution, the key features in the collected image are extracted: the body contour features of the target object are identified, the joint movement features of the target object, such as the angle change of the joint, the movement trajectory, etc., and the speed change features of the target object, that is, the dynamic information such as acceleration and deceleration during the climbing process, etc., are captured and analyzed. The above features together constitute the movement trend characteristics of the target object, which provides an important basis for evaluating its stability and safety during the climbing process; the state of the target suspension ladder is analyzed in detail: the swing amplitude characteristics of the suspension ladder are detected, that is, the degree of swing of the suspension ladder under the influence of external factors such as wind and waves, and at the same time, attention is paid to the surface state characteristics of the suspension ladder, such as whether there are potential safety hazards such as wear, rust, cracks, etc., and the surrounding environment of the suspension ladder is comprehensively scanned to identify possible obstacle features, such as protruding hull structures, floating objects, etc., as well as weather and lighting characteristics, such as light intensity, whether it is raining, whether there is haze, etc. , all of which may affect the climbing safety of the crew and the stability of the ladder; based on the above-mentioned identified features, the motion trend features, the state characteristics of the ladder and the environmental characteristics are further extracted. The motion trend features integrate information such as the crew's body contour, joint movement and speed change, and can reflect the stability and coordination of the crew's movements during the climbing process, while the state characteristics and environmental characteristics of the ladder cover the ladder's own conditions and external environmental factors, providing comprehensive information for assessing potential risks during the climbing process; this application provides a comprehensive and accurate information basis for hazard identification during the climbing process of a ship's ladder by detailed identification and analysis of the crew's motion trend characteristics, the state characteristics of the ladder and the environmental characteristics, which not only improves the accuracy and efficiency of hazard identification, but also can evaluate the safety risks in the climbing process in real time according to the identified features, effectively improving the protection of the crew's life safety and the safe operation of the ship.

[0021] In a preferred example, the present application can be further configured as follows: the risk prediction model includes an integration layer, a coupling analysis layer, an adjustment layer, and an evaluation layer, and the step of inputting the extracted action trend features, state features, environmental features, and monitoring data into the pre-trained risk prediction model for risk prediction includes the steps of:

[0022] The integration layer dynamically integrates the input environmental features to obtain a comprehensive environmental feature vector;

[0023] The coupling analysis layer performs coupling effect analysis based on comprehensive environmental feature vectors, action trend features, state features, and monitoring data, and outputs analysis result information;

[0024] The adjustment layer adjusts the parameters and structure of the risk prediction model based on the analysis result information;

[0025] The assessment layer performs hazard level assessment based on the analysis result information, the adjustment results of the adjustment layer, and the preset hazard level classification standards, and outputs prediction information.

[0026] By adopting the above technical solution, the danger prediction model includes an integration layer, a coupling analysis layer, an adjustment layer and an evaluation layer. The integration layer receives the input environmental features and dynamically integrates them to fuse them into a comprehensive environmental feature vector. The coupling analysis layer receives the comprehensive environmental feature vector, as well as action trend features, state features and monitoring data, and performs a deep coupling effect analysis on the received features, and outputs analysis result information containing correlation information between these features. The adjustment layer dynamically adjusts the parameters and structure of the danger prediction model according to the analysis result information output by the coupling analysis layer. The evaluation layer performs a danger level assessment on the climbing situation of the current target object based on the received analysis result information of the coupling analysis layer, the adjustment result of the adjustment layer, and the pre-set danger level classification standard, and outputs prediction information. The present application constructs a danger prediction model including an integration layer, a coupling analysis layer, an adjustment layer and an evaluation layer to achieve comprehensive analysis and dynamic prediction of multiple features in the process of climbing a ship's suspended ladder, which not only improves the accuracy and timeliness of danger prediction, but also can dynamically adjust the model according to the current environment and crew status, thereby enhancing the adaptability and robustness of the model.

[0027] In a preferred example, the present application can be further configured as follows: the step of the adjustment layer adjusting the parameters and structure of the risk prediction model based on the analysis result information includes the following steps:

[0028] The adjustment layer evaluates the contribution of comprehensive environment feature vectors, action trend features, state features, and monitoring data based on the analysis results;

[0029] The adjustment layer assigns attention weights to the comprehensive environment feature vector, action trend feature, state feature, and monitoring data based on the contribution evaluation results;

[0030] The adjustment layer adjusts the parameters and structure of the danger prediction model based on the attention weights.

[0031] By adopting the above technical solution, the adjustment layer evaluates the contribution of the comprehensive environment feature vector, action trend feature, state feature and monitoring data according to the analysis result information output by the coupling analysis layer, so as to measure the importance and influence of each feature in predicting danger. Based on the contribution evaluation result, the adjustment layer assigns an attention weight to each feature, which reflects the relative importance of the feature in the model decision-making process. The adjustment layer dynamically adjusts the parameters (such as the weight and bias of the neural network) and structure (such as the number of network layers, the number of neurons in each layer, etc.) of the danger prediction model according to the assigned attention weight; this application introduces a dynamic adjustment strategy based on contribution evaluation and attention weight, so that the danger prediction model can more accurately capture the features that are crucial to danger prediction, and dynamically adjust the model parameters and structure to adapt to the changes in these features, which not only improves the prediction accuracy of the model, but also enhances the robustness and adaptability of the model, so that it can better cope with the complex and changeable ship ladder climbing environment.

[0032] In a preferred example, the present application can be further configured as follows: the evaluation layer performs a hazard level evaluation based on the analysis result information, the adjustment result of the adjustment layer, and the preset hazard level classification standard, and outputs the prediction information, including the steps of:

[0033] The assessment layer conducts initial hazard level assessment based on the analysis result information, the adjustment result of the adjustment layer and the preset hazard level classification standards;

[0034] The assessment layer obtains the stored historical dangerous event data and crew behavior pattern data, and performs similarity calculation on the initial danger level assessment results;

[0035] The evaluation layer performs the final hazard level assessment based on the similarity calculation results and outputs the prediction information.

[0036] By adopting the above technical solution, the assessment layer comprehensively considers the analysis result information, the adjustment results of the model parameters and structure according to the real-time situation by the adjustment layer, and the preset danger level classification standards, and conducts an initial danger level assessment. The assessment layer obtains the stored historical danger event data and crew behavior pattern data from the database. These data include dangerous events that have occurred in the past and their related characteristics, as well as the behavioral responses of the crew in similar situations. The initial danger level assessment results are compared with these historical data using similarity calculation technology to find the historical data or historical cases that are most similar to the current situation. Based on the similarity calculation results, the assessment layer corrects or confirms the initial danger level assessment results to obtain the final danger level assessment results.

[0037] The second object of the invention is achieved by the following technical solutions:

[0038] A ship ladder climbing hazard identification system, comprising:

[0039] The data acquisition module is used to identify the identity information of the target object when receiving the captured image sent by the image acquisition terminal, and to obtain the monitoring data of the physiological parameter monitoring device associated with the identity information in real time;

[0040] A feature extraction module is used to extract the motion trend features of the target object in the captured image and the state features and environmental features of the target suspension ladder;

[0041] An input module is used to input the extracted action trend features, state features, environmental features and monitoring data into a pre-trained risk prediction model for risk prediction;

[0042] The response module is used to trigger a corresponding response mechanism according to the danger level when receiving the prediction information output by the danger prediction model.

[0043] By adopting the above technical solution, the data acquisition module is used to identify the identity information of the target object when receiving the captured image sent by the image acquisition terminal, and to obtain the monitoring data of the physiological parameter monitoring device associated with the identity information in real time; the feature extraction module is used to extract the action trend characteristics of the target object in the captured image and the state characteristics and environmental characteristics of the target suspension ladder; the input module is used to input the extracted action trend characteristics, state characteristics, environmental characteristics and monitoring data into a pre-trained risk prediction model for risk prediction; the response module is used to trigger a corresponding response mechanism according to its risk level when receiving the prediction information output by the risk prediction model.

[0044] In a preferred example, the present application can be further configured as follows: the feature extraction module includes:

[0045] The first recognition submodule is used to recognize the body contour features, joint movement features and speed change features of the target object in the captured image;

[0046] The second recognition submodule is used to recognize the shaking amplitude characteristics and surface state characteristics of the target suspension ladder in the collected image, and recognize the obstacle characteristics and weather and lighting characteristics of the surrounding environment of the target suspension ladder;

[0047] The first extraction submodule is used to extract the motion trend feature based on the body contour feature, joint movement feature, speed change feature of the target object and the shaking amplitude feature of the target suspension ladder;

[0048] The second extraction submodule is used to extract state characteristics and environmental characteristics of the suspension ladder based on the surface state characteristics and shaking amplitude characteristics of the target suspension ladder and the obstacle characteristics and weather and illumination characteristics of the surrounding environment of the target suspension ladder.

[0049] By adopting the above technical solution, the first recognition submodule is used to recognize the body contour features, joint movement features and speed change features of the target object in the captured image; the second recognition submodule is used to recognize the shaking amplitude features and surface state features of the target suspension ladder in the captured image, and recognize the obstacle features and weather and lighting features of the environment surrounding the target suspension ladder;

[0050] The first extraction submodule is used to extract motion trend features based on the body contour features, joint movement features, speed change features of the target object and the sway amplitude features of the target suspension ladder; the second extraction submodule is used to extract the state features and environmental features of the suspension ladder based on the surface state features, sway amplitude features of the target suspension ladder and the obstacle features and weather and lighting features of the surrounding environment of the target suspension ladder.

[0051] In summary, the present application includes at least one of the following beneficial technical effects:

[0052] 1. This application realizes comprehensive and real-time hazard identification of the ship's suspended ladder climbing process by combining image recognition based on crew action trends, ladders and the environment, crew physiological parameter monitoring and hazard prediction models. It not only improves the accuracy and timeliness of hazard identification, but also can automatically trigger corresponding response mechanisms according to the hazard level, thereby effectively reducing the safety risks in the climbing process and ensuring the life safety of the crew;

[0053] 2. This application provides a comprehensive and accurate information basis for the identification of dangers during the climbing process of the ship's suspended ladder by identifying and analyzing the crew's action trend characteristics, the status characteristics of the suspended ladder, and the environmental characteristics in detail. It not only improves the accuracy and efficiency of hazard identification, but also can evaluate the safety risks in the climbing process in real time according to the identified characteristics, effectively improving the protection of the crew's life safety and the safe operation of the ship;

[0054] 3. This application constructs a danger prediction model including an integration layer, a coupling analysis layer, an adjustment layer and an evaluation layer to achieve a comprehensive analysis and dynamic prediction of various characteristics during the climbing process of a ship's suspended ladder, which not only improves the accuracy and timeliness of danger prediction, but also dynamically adjusts the model according to the current environment and crew status, thereby enhancing the adaptability and robustness of the model;

[0055] 4. This application introduces a dynamic adjustment strategy based on contribution evaluation and attention weight, so that the hazard prediction model can more accurately capture the features that are critical to hazard prediction, and dynamically adjust the model parameters and structure to adapt to changes in these features. This not only improves the prediction accuracy of the model, but also enhances the robustness and adaptability of the model, enabling it to better cope with the complex and changeable ship ladder climbing environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1This is a flow chart of an embodiment of a method for identifying climbing hazards of a ship ladder in the present application;

[0057] Figure 2 This is a flowchart of implementing step S20 in an embodiment of a method for identifying the hazard of climbing a ship ladder in the present application;

[0058] Figure 3 This is a flowchart for implementing step S30 in an embodiment of a method for identifying climbing hazards of a ship ladder in the present application;

[0059] Figure 4 This is a flowchart for implementing step S33 in an embodiment of a method for identifying the hazard of climbing a ship ladder in the present application;

[0060] Figure 5 This is a flowchart for implementing step S34 in an embodiment of a method for identifying climbing hazards of a ship ladder in the present application. DETAILED DESCRIPTION

[0061] The following is combined with Figure 1-5 This application is described in further detail.

[0062] In one embodiment, if Figure 1 As shown, the present application discloses a method for identifying the hazard of climbing a ship ladder, which specifically includes the following steps:

[0063] S10: When receiving the captured image sent by the image acquisition terminal, identifying the identity information of the target object, and acquiring the monitoring data of the physiological parameter monitoring device associated with the identity information in real time;

[0064] In this embodiment, the image acquisition terminal is a device for capturing and transmitting real-time video images, including a camera, an infrared thermal imager, etc.; the target object is a person climbing the escalator, specifically a crew member on the ship; the identity information is the pre-stored basic personal identity information of each crew member; the physiological parameter monitoring device is a device for monitoring human physiological indicators (such as heart rate, blood pressure, etc.); the monitoring data is the human physiological indicator data monitored by the physiological parameter monitoring device;

[0065] Specifically, the real-time image is received from the image acquisition terminal, and the identity information of the target object (i.e., the crew member) who is climbing the ladder is identified, the identity information of the crew member is associated, and the monitoring data transmitted by the physiological parameter monitoring device worn by the crew member is obtained in real time;

[0066] S20: extracting the motion trend features of the target object in the captured image and the state features and environmental features of the target suspension ladder;

[0067] In this embodiment, the motion trend feature is the dynamic properties of the target object in the continuous captured images, such as the motion mode, direction, speed, acceleration, etc.; the target ladder is a specific ladder structure in the captured images, and its state features include the inclination, stability, whether it is damaged or worn, whether someone is using it, etc.; the environmental features are other environmental factors in the captured images except the target object and the target ladder, including weather conditions (sunny, rainy, snowy, etc.), lighting conditions (bright, dim), the layout and type of surrounding objects, etc.;

[0068] Specifically, the collected images are analyzed in depth to extract the target object's motion trend characteristics, such as climbing speed, body posture changes, etc., while identifying the state characteristics of the suspended ladder and the surrounding environment characteristics;

[0069] S30: Inputting the extracted action trend features, state features, environmental features and monitoring data into the pre-trained risk prediction model to perform risk prediction;

[0070] In this embodiment, the danger prediction model is a model built based on historical data and machine learning algorithms to predict potential danger situations and their levels;

[0071] Specifically, the extracted motion trend characteristics, ladder status characteristics, environmental characteristics and crew member physiological parameter monitoring data are input into a pre-trained danger prediction model. This model is based on a large amount of historical data and machine learning algorithms, and can predict the possibility and danger level of danger in the current climbing situation.

[0072] S40: When receiving the prediction information output by the risk prediction model, triggering a corresponding response mechanism according to its risk level;

[0073] In this embodiment, the prediction information is the electrical signal information containing the dangerous situation and its level output by the dangerous prediction model after the dangerous situation prediction; the dangerous level is a pre-set level mechanism for quantitatively grading the dangerous situation; the response mechanism is the response measures or action plans taken based on the dangerous level when the predicted information is identified to contain the dangerous situation;

[0074] Specifically, when the prediction information output by the danger prediction model is received, the corresponding response mechanism is immediately triggered according to the predicted danger level.

[0075] In one embodiment, if Figure 2 As shown, step S20 includes the steps of:

[0076] S21: Identify the body contour features, joint movement features, and speed change features of the target object in the captured image;

[0077] S22: identifying the shaking amplitude characteristics and surface state characteristics of the target suspension ladder in the collected image, and identifying the obstacle characteristics and weather and illumination characteristics of the surrounding environment of the target suspension ladder;

[0078] S23: extracting motion trend features based on the body contour features, joint movement features, speed change features of the target object and the sway amplitude features of the target suspension ladder;

[0079] S24: extracting state features and environmental features of the target suspension ladder based on surface state features, sway amplitude features, obstacle features, and weather and illumination features of the surrounding environment of the target suspension ladder;

[0080] In this embodiment, the body contour feature is the feature of the external shape and contour of the target object's body in the captured image, including the overall size, shape, proportion of the body and any significant body curves or protrusions; the joint movement feature is the movement of the target object's joints, including the angle change, movement speed, movement range, etc. of the joints; the speed change feature is the change in speed when the target object moves in the image, including the increase or decrease in speed, whether it is uniform, etc.; the sway amplitude feature is the degree of sway of the target ladder under the influence of factors such as wind force and user activities, which is quantified by the swing amplitude or frequency of the ladder; the surface state feature is the condition of the surface of the target ladder, including whether there is damage, wear, corrosion, stains, etc.; the obstacle feature is the obstacles in the surrounding environment of the target ladder, which may affect the safety of the ladder. The weather and lighting features are the weather conditions and lighting conditions when the image is collected, such as sunny, cloudy, rainy, snowy, and the intensity and direction of the light. The action trend features are extracted based on the target object's body contour features, joint movement features, speed change features, and the swing amplitude features of the target ladder, and are used to describe the characteristics of the target object's possible future actions or behavior trends. The state features of the ladder are extracted based on the surface state features and swing amplitude features of the target ladder, and are used to describe the current state of the ladder (such as safety, stability, etc.). The environmental features are extracted based on the obstacle features and weather and lighting features of the surrounding environment of the target ladder, and are used to describe the characteristics of the environmental conditions (such as space restrictions, line of sight obstruction, weather influence, etc.) in which the ladder is located.

[0081] Furthermore, the infrared thermal imager can capture the temperature distribution image of the suspension ladder surface, and by analyzing the changes in temperature distribution, identify whether the suspension ladder surface is slippery, damaged or other safety hazards; for example, a slippery surface may cause uneven temperature distribution, while a damaged area may appear as abnormal hot and cold spots;

[0082] Specifically, the key features in the captured image are extracted: the body contour features of the target object are identified, the joint motion features of the target object, such as the angle changes of the joints, the motion trajectory, etc., and the speed change features of the target object, that is, the dynamic information such as acceleration and deceleration during the climbing process. The above features together constitute the motion trend characteristics of the target object, which provides an important basis for evaluating its stability and safety during the climbing process; the state of the target ladder is analyzed in detail: the swing amplitude characteristics of the ladder are detected, that is, the degree of swing of the ladder under the influence of external factors such as wind and waves, and at the same time, attention is paid to the surface state characteristics of the ladder, such as whether there are potential safety hazards such as wear, rust, cracks, etc., and the ladder is inspected. A comprehensive scan of the surrounding environment is performed to identify possible obstacle features, such as protruding hull structures, floating objects, etc., as well as weather and lighting features, such as light intensity, whether it is raining, whether there is haze, etc. These factors may affect the climbing safety of the crew and the stability of the ladder; based on the above-mentioned identified features, the action trend features, the state characteristics of the ladder and the environmental characteristics are further extracted. The action trend features integrate information such as the crew's body contour, joint movement and speed change, and can reflect the crew's movement stability and coordination during the climbing process. The state characteristics and environmental characteristics of the ladder cover the ladder's own conditions and external environmental factors, providing comprehensive information for assessing potential risks during the climbing process.

[0083] In one embodiment, the risk prediction model includes an integration layer, a coupling analysis layer, an adjustment layer, and an evaluation layer. Figure 3 As shown, step S30 includes the steps of:

[0084] S31: The integration layer dynamically integrates the input environmental features to obtain a comprehensive environmental feature vector;

[0085] S32: The coupling analysis layer performs coupling effect analysis based on the comprehensive environment feature vector, action trend feature, state feature and monitoring data, and outputs analysis result information;

[0086] S33: the adjustment layer adjusts the parameters and structure of the risk prediction model based on the analysis result information;

[0087] S34: the assessment layer performs a hazard level assessment based on the analysis result information, the adjustment result of the adjustment layer, and the preset hazard level classification standard, and outputs prediction information;

[0088] In this embodiment, the integration layer is the first layer in the danger prediction model, which is used to dynamically integrate the input environmental features (such as weather, lighting, obstacles, etc.); the comprehensive environmental feature vector is the result output by the integration layer, which is specifically a vector containing all input environmental feature information. The vector integrates various environmental features in a specific way (such as weighted average, feature fusion, etc.) to facilitate further analysis by subsequent layers; the coupling analysis layer is the second layer in the danger prediction model, which performs coupling effect analysis based on the comprehensive environmental feature vector, action trend features, state features and monitoring data to analyze the interactions and associations between these features, as well as their impact on danger prediction; the analysis result information is the result output by the coupling analysis layer, which contains the analysis results on the coupling effects between features, including the correlation, impact degree, trend, etc. between features, for subsequent layers The adjustment layer is the third layer in the hazard prediction model, which adjusts the parameters and structure of the model based on the analysis result information output by the coupling analysis layer; the evaluation layer is the last layer in the hazard prediction model, which performs hazard level evaluation based on the analysis result information of the coupling analysis layer, the adjustment result of the adjustment layer and the preset hazard level classification standard, so as to output a clear hazard level prediction information so as to take corresponding response measures; the hazard level classification standard is the standard used by the evaluation layer for hazard level evaluation, which is set according to specific application scenarios and requirements, and usually includes multiple hazard levels (such as low risk, medium risk, high risk, etc.), as well as the degree of hazard and possible consequences corresponding to each level; the prediction information is the final result output by the evaluation layer, which contains predictions about the current situation or the hazard situation and its corresponding hazard level in the future;

[0089] The coupling analysis layer conducts multi-dimensional coupling effect analysis based on comprehensive environmental feature vectors, motion trend features, state features, and monitoring data (such as wind speed, wave height, crew heart rate, etc.). The analysis process includes but is not limited to:

[0090] Trend risk assessment: Combine the crew's physical condition and climbing movements, that is, combine monitoring data and movement trend characteristics to assess their risk level in the current environment. For example, uncoordinated joint movements or abnormal speed changes may indicate the risk of crew members becoming exhausted or losing balance.

[0091] Environmental adaptability analysis: Consider the impact of weather, light and ladder status on climbing safety, that is, combine the comprehensive environmental feature vector and state characteristics. For example, strong wind, heavy rain or slippery ladders may significantly increase the risk of climbing.

[0092] Obstacle impact assessment: Analyze the potential interference of surrounding obstacles on the climbing path and how they interact with the crew's climbing actions, that is, combine the comprehensive environmental feature vector, action trend characteristics and state characteristics to analyze whether there is an increased risk of collision or tripping;

[0093] A comprehensive analysis combining the above risk assessment, environmental adaptability analysis and obstacle impact assessment;

[0094] Specifically, the danger prediction model includes an integration layer, a coupling analysis layer, an adjustment layer and an evaluation layer. The integration layer receives input environmental features and dynamically integrates them to fuse them into a comprehensive environmental feature vector. The coupling analysis layer receives the comprehensive environmental feature vector, as well as action trend features, state features and monitoring data, and performs a deep coupling effect analysis on the received features, and outputs analysis result information containing correlation information between these features. The adjustment layer dynamically adjusts the parameters and structure of the danger prediction model according to the analysis result information output by the coupling analysis layer. The evaluation layer performs a danger level assessment on the climbing situation of the current target object based on the analysis result information received from the coupling analysis layer, the adjustment results of the adjustment layer, and the pre-set danger level classification standard, and outputs prediction information.

[0095] In one embodiment, if Figure 4 As shown, step S33 includes the steps of:

[0096] S331: The adjustment layer evaluates the contribution of the comprehensive environment feature vector, the action trend feature, the state feature, and the monitoring data based on the analysis results;

[0097] S332: The adjustment layer assigns attention weights to the comprehensive environment feature vector, action trend feature, state feature, and monitoring data based on the evaluation result of the contribution;

[0098] S333: The adjustment layer adjusts the parameters and structure of the danger prediction model based on the attention weight;

[0099] In this embodiment, the contribution is the importance or influence of each feature (comprehensive environmental feature vector, action trend feature, state feature and monitoring data) in the risk prediction model on the model prediction result; the attention weight is the weight value assigned according to the contribution of the feature, which is used to weight different features in the model; adjusting the parameters and structure of the risk prediction model specifically involves adjusting the parameters and structure of the risk prediction model according to the assigned attention weight, including modifying the model's weight, bias and other parameters, and possibly changing the model's network structure, number of layers, number of neurons, etc.;

[0100] Specifically, the adjustment layer evaluates the contribution of the comprehensive environmental feature vector, action trend features, state features and monitoring data according to the analysis result information output by the coupling analysis layer to measure the importance and influence of each feature in predicting danger. Based on the contribution evaluation results, the adjustment layer assigns an attention weight to each feature. This weight reflects the relative importance of the feature in the model decision-making process. The adjustment layer dynamically adjusts the parameters and structure of the danger prediction model according to the assigned attention weight.

[0101] In one embodiment, if Figure 5 As shown, step S34 includes the steps of:

[0102] S341: the assessment layer performs an initial hazard level assessment based on the analysis result information, the adjustment result of the adjustment layer, and the preset hazard level classification standard;

[0103] S342: the assessment layer obtains the stored historical dangerous event data and crew behavior pattern data, and performs similarity calculation on the initial danger level assessment results;

[0104] S343: The assessment layer performs a final risk level assessment based on the similarity calculation result and outputs prediction information;

[0105] In this embodiment, the historical dangerous event data is a data set that records dangerous events that occurred in the past and their related characteristics; the crew behavior pattern data is data that reflects the crew's behavioral responses in a specific situation; similarity calculation is a method of measuring the similarity between two data objects;

[0106] Furthermore, the crew behavior pattern data includes the crew’s skill level and experience to assess their risk during the climb;

[0107] Specifically, the assessment layer comprehensively considers the analysis result information (including multi-dimensional data such as environmental characteristics, movement trends, and ship status), the adjustment results of the model parameters and structure according to the real-time situation by the adjustment layer, and the pre-set danger level classification standards (such as low risk, medium risk, high risk, etc.), and conducts an initial danger level assessment. The assessment layer obtains the stored historical danger event data and crew behavior pattern data from the database. These data include dangerous events that have occurred in the past and their related characteristics, as well as the behavioral responses of crew members in similar situations. The initial danger level assessment results are compared with these historical data using similarity calculation techniques (such as cosine similarity, Euclidean distance, etc.) to find the historical data or historical cases that are most similar to the current situation. Based on the similarity calculation results, the assessment layer corrects or confirms the initial danger level assessment results to obtain the final danger level assessment results.

[0108] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0109] In one embodiment, a ship ladder climbing hazard identification system is provided, and the ship ladder climbing hazard identification system corresponds one-to-one to a ship ladder climbing hazard identification method in the above embodiment. A ship ladder climbing hazard identification system includes:

[0110] The data acquisition module is used to identify the identity information of the target object when receiving the captured image sent by the image acquisition terminal, and to obtain the monitoring data of the physiological parameter monitoring device associated with the identity information in real time;

[0111] A feature extraction module is used to extract the motion trend features of the target object in the captured image and the state features and environmental features of the target suspension ladder;

[0112] An input module is used to input the extracted action trend features, state features, environmental features and monitoring data into a pre-trained risk prediction model for risk prediction;

[0113] A response module is used to trigger a corresponding response mechanism according to the danger level when receiving the prediction information output by the danger prediction model;

[0114] Optionally, the feature extraction module includes:

[0115] The first recognition submodule is used to recognize the body contour features, joint movement features and speed change features of the target object in the captured image;

[0116] The second recognition submodule is used to recognize the shaking amplitude characteristics and surface state characteristics of the target suspension ladder in the collected image, and recognize the obstacle characteristics and weather and lighting characteristics of the surrounding environment of the target suspension ladder;

[0117] The first extraction submodule is used to extract the motion trend feature based on the body contour feature, joint movement feature, speed change feature of the target object and the shaking amplitude feature of the target suspension ladder;

[0118] The second extraction submodule is used to extract the state characteristics and environmental characteristics of the suspension ladder based on the surface state characteristics and shaking amplitude characteristics of the target suspension ladder and the obstacle characteristics and weather and illumination characteristics of the surrounding environment of the target suspension ladder;

[0119] Optionally, also include:

[0120] The integration layer module is used to dynamically integrate the input environmental features to obtain a comprehensive environmental feature vector;

[0121] The coupling analysis layer module is used to analyze the coupling effect based on the comprehensive environmental feature vector, action trend feature, state feature and monitoring data, and output the analysis result information;

[0122] An adjustment layer module, used to adjust the parameters and structure of the risk prediction model based on the analysis result information;

[0123] The evaluation layer module is used to evaluate the danger level based on the analysis result information, the adjustment result of the adjustment layer and the preset danger level classification standard, and output the prediction information;

[0124] Optional, adjustment layer modules include:

[0125] The contribution evaluation submodule is used to evaluate the contribution of the comprehensive environment feature vector, action trend feature, state feature and monitoring data based on the analysis results;

[0126] The weight allocation submodule is used to allocate attention weights to the comprehensive environment feature vector, action trend feature, state feature and monitoring data based on the contribution evaluation results;

[0127] An adjustment submodule, used to adjust the parameters and structure of the danger prediction model based on the attention weight;

[0128] Optionally, the evaluation layer modules include:

[0129] The initial assessment submodule is used to perform initial hazard level assessment based on the analysis result information, the adjustment result of the adjustment layer and the preset hazard level classification standard;

[0130] The similarity calculation submodule is used to obtain the stored historical dangerous event data and crew behavior pattern data, and perform similarity calculation on the initial danger level assessment results;

[0131] The final evaluation submodule is used to perform the final danger level evaluation based on the similarity calculation results and output the prediction information.

[0132] For the specific definition of a ship ladder climbing hazard identification system, please refer to the definition of a ship ladder climbing hazard identification method mentioned above, which will not be repeated here. Each module in the above-mentioned ship ladder climbing hazard identification system can be implemented in whole or in part through software, hardware and their combination. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0133] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A method for identifying climbing hazards of a ship's suspended ladder, characterized in that: Includes steps: When receiving the captured image sent by the image acquisition terminal, identifying the identity information of the target object, and acquiring the monitoring data of the physiological parameter monitoring device associated with the identity information in real time; Extracting the motion trend features of the target object in the captured image and the state features and environmental features of the target suspension ladder; The extracted action trend features, state features, environmental features and monitoring data are input into the pre-trained risk prediction model for risk prediction; When the prediction information output by the danger prediction model is received, the corresponding response mechanism is triggered according to its danger level.

2. A method for identifying ship ladder climbing hazards according to claim 1, characterized in that: The step of extracting the action trend characteristics of the target object in the captured image and the state characteristics and environmental characteristics of the target hanging ladder comprises the steps of: Identify the body contour features, joint movement features, and speed change features of the target object in the captured image; Identify the shaking amplitude characteristics and surface state characteristics of the target suspension ladder in the collected image, and identify the obstacle characteristics and weather and lighting characteristics of the environment surrounding the target suspension ladder; Extract motion trend features based on the target object's body contour features, joint movement features, speed change features, and the swing amplitude features of the target suspended ladder; The state characteristics and environmental characteristics of the target suspension ladder are extracted based on the surface state characteristics, shaking amplitude characteristics, obstacle characteristics and weather and lighting characteristics of the target suspension ladder's surrounding environment.

3. A method for identifying climbing hazards of a ship suspended ladder according to claim 1, characterized in that: The risk prediction model includes an integration layer, a coupling analysis layer, an adjustment layer and an evaluation layer. The step of inputting the extracted action trend features, state features, environmental features and monitoring data into the pre-trained risk prediction model for risk prediction includes the following steps: The integration layer dynamically integrates the input environmental features to obtain a comprehensive environmental feature vector; The coupling analysis layer performs coupling effect analysis based on comprehensive environmental feature vectors, action trend features, state features, and monitoring data, and outputs analysis result information; The adjustment layer adjusts the parameters and structure of the risk prediction model based on the analysis result information; The assessment layer performs hazard level assessment based on the analysis result information, the adjustment results of the adjustment layer, and the preset hazard level classification standards, and outputs prediction information.

4. A method for identifying ship ladder climbing hazards according to claim 3, characterized in that: The step of adjusting the parameters and structure of the risk prediction model based on the analysis result information by the adjustment layer includes the following steps: The adjustment layer evaluates the contribution of comprehensive environment feature vectors, action trend features, state features, and monitoring data based on the analysis results; The adjustment layer assigns attention weights to the comprehensive environment feature vector, action trend feature, state feature, and monitoring data based on the contribution evaluation results; The adjustment layer adjusts the parameters and structure of the danger prediction model based on the attention weights.

5. A method for identifying climbing hazards of a ship suspended ladder according to claim 3, characterized in that: The step of the assessment layer performing a hazard level assessment based on the analysis result information, the adjustment result of the adjustment layer and the preset hazard level classification standard and outputting prediction information comprises the steps of: The assessment layer conducts initial hazard level assessment based on the analysis result information, the adjustment result of the adjustment layer and the preset hazard level classification standards; The assessment layer obtains the stored historical dangerous event data and crew behavior pattern data, and performs similarity calculation on the initial danger level assessment results; The evaluation layer performs the final hazard level assessment based on the similarity calculation results and outputs the prediction information.

6. A ship ladder climbing hazard identification system, characterized in that: include: The data acquisition module is used to identify the identity information of the target object when receiving the captured image sent by the image acquisition terminal, and to obtain the monitoring data of the physiological parameter monitoring device associated with the identity information in real time; A feature extraction module is used to extract the motion trend features of the target object in the captured image and the state features and environmental features of the target suspension ladder; An input module is used to input the extracted action trend features, state features, environmental features and monitoring data into a pre-trained risk prediction model for risk prediction; The response module is used to trigger a corresponding response mechanism according to the danger level when receiving the prediction information output by the danger prediction model.

7. A ship ladder climbing hazard identification system according to claim 6, characterized in that: The feature extraction module comprises: The first recognition submodule is used to recognize the body contour features, joint movement features and speed change features of the target object in the captured image; The second recognition submodule is used to recognize the shaking amplitude characteristics and surface state characteristics of the target suspension ladder in the collected image, and recognize the obstacle characteristics and weather and lighting characteristics of the surrounding environment of the target suspension ladder; The first extraction submodule is used to extract the motion trend feature based on the body contour feature, joint movement feature, speed change feature of the target object and the shaking amplitude feature of the target suspension ladder; The second extraction submodule is used to extract state characteristics and environmental characteristics of the suspension ladder based on the surface state characteristics and shaking amplitude characteristics of the target suspension ladder and the obstacle characteristics and weather and illumination characteristics of the surrounding environment of the target suspension ladder.

Images