An artificial intelligence-based safety monitoring method and system for fishing vessel operators

Abstract

The present invention discloses a method and system for safety monitoring of fishing vessel operators based on artificial intelligence. The method includes basic data collection, safety area monitoring, fishing vessel environment monitoring, personnel behavior monitoring, and comprehensive safety monitoring. The present invention relates to the technical field of fishing vessel safety monitoring, and specifically refers to a method and system for safety monitoring of fishing vessel operators based on artificial intelligence. This solution adopts an integrated safety analysis scheme that combines a safety area monitoring sub-model, a fishing vessel environment monitoring sub-model, and a personnel behavior monitoring sub-model for subnet integrated weighted training; uses an improved vision detection model combined with a multi-level feature processing method for safety area monitoring, and enhances the small target recognition ability and overall calculation efficiency by improving the model structure; uses a convolutional long short-term memory neural network combined with the optimization of fishing vessel trajectory feature description for fishing vessel environment monitoring; uses a deep attention network combined with fuzzy reasoning and pre-trained vision extraction for personnel behavior monitoring.

Description

Technical Field

[0001] The present invention relates to the technical field of fishing vessel safety monitoring, and specifically refers to a method and system for monitoring the safety of fishing vessel operators based on artificial intelligence. Background Art

[0002] The method for monitoring the safety of fishing vessel operators based on artificial intelligence refers to using AI technology to monitor and analyze the activities of fishing vessel operators in real time, so as to identify and prevent potential safety risks, thereby ensuring the life safety of fishing vessel operators and improving the overall safety of fishing operations.

[0003] However, in the existing safety monitoring of fishing vessel operators, there are technical problems that traditional intelligent technology solutions often only perform single image analysis or single numerical analysis, which leads to insufficient and less referenceable intelligent data support for the safety monitoring of fishing vessel operators; in the existing safety area monitoring methods, there are situations where the existing standard image detection models have low computing efficiency and poor recognition of small targets in dealing with complex fishing vessel safety area monitoring tasks, which is likely to lead to poor safety area monitoring effects and thus affect the safety risks of fishing vessel operators; in the existing fishing vessel environment monitoring methods, there is a technical problem that traditional environmental detection only conducts a general intelligent analysis of the external environment and ignores the potential safety problems that may exist during the fishing vessel's navigation process; in the existing personnel behavior monitoring methods, there is a technical problem that personnel behavior monitoring involves both intuitive visual features and other potential features that cannot be directly observed through camera data. Summary of the Invention

[0004] In view of the above situation, to overcome the defects of the prior art, the present invention provides a method and system for predicting drug-target interactions based on artificial intelligence. In the existing safety monitoring of fishing boat operators, traditional intelligent technology solutions often only perform single image analysis or single numerical analysis, which leads to the technical problems that the intelligent data support for the safety monitoring of fishing boat operators is inevitably insufficiently comprehensive and has low reference value. This solution creatively adopts an integrated safety analysis scheme that combines a safety area monitoring sub-model, a fishing boat environment monitoring sub-model, and a personnel behavior monitoring sub-model for subnet integrated weighted training. By combining the three aspects of the fishing boat safety area, the fishing boat driving trajectory environment, and personnel behavior analysis, the comprehensiveness of the safety monitoring of fishing boat operators is improved. At the same time, through weighted processing, the intelligence and reliability of the safety monitoring of fishing boat operators are improved. In addition, the design of the sub-item sub-models is also beneficial to the independent operation and maintenance of sub-tasks, improving the overall stability and scalability of the system. In the existing safety area monitoring methods, there are technical problems that the existing standard image detection models have low computing efficiency and poor recognition of small targets in dealing with complex fishing boat safety area monitoring tasks, which is likely to lead to poor safety area monitoring effects and thus affect the safety risks of fishing boat operators. This solution creatively adopts a method of combining an improved vision detection model with multi-level feature processing for safety area monitoring. By improving the model structure, the small target recognition ability and overall computing efficiency are enhanced, and the overall usability of safety area monitoring is improved through multi-level feature fusion. In the existing fishing boat environment monitoring methods, there are technical problems that traditional environmental detection only conducts general intelligent analysis on the external environment and ignores the possible safety problems during the fishing boat navigation process itself. This solution creatively adopts a convolutional long short-term memory neural network combined with the optimization of fishing boat trajectory feature description for fishing boat environment monitoring. By designing four types of fishing boat navigation trajectory environment features and combining convolutional long short-term networks and clustering classification networks, complex fishing boat environment monitoring is realized, providing strong support for subsequent comprehensive monitoring and personnel behavior monitoring. In the existing personnel behavior monitoring methods, there are technical problems that personnel behavior monitoring involves both intuitive visual features and other potential features that cannot be directly observed through camera data. This solution creatively adopts a deep attention network combining fuzzy reasoning and pre-trained vision extraction for personnel behavior monitoring. Combining the results of fishing boat environment monitoring and the model improvement of safety area monitoring, the capture ability of the potential logic of personnel behavior analysis is improved together, and the overall practicality of the safety monitoring of fishing boat operators is improved.

[0005] The technical solution adopted by the present invention is as follows: A method for safety monitoring of fishing boat operators based on artificial intelligence provided by the present invention includes the following steps:

[0006] Step S1: Basic data collection;

[0007] Step S2: Safety area monitoring;

[0008] Step S3: Fishing boat environment monitoring;

[0009] Step S4: Personnel behavior monitoring;

[0010] Step S5: Comprehensive safety monitoring.

[0011] Furthermore, in Step S1, the basic data collection is used to collect the original dataset required for the safety monitoring of fishing boat operators. Specifically, the original data for the safety monitoring of fishing boats is collected through a camera device and sensors, and the original data for the safety monitoring of fishing boat operators is obtained;

[0012] The camera device is specifically placed in the fishing boat cockpit, on the fishing boat deck, and at the stern of the fishing boat for safety area monitoring and personnel behavior monitoring;

[0013] The sensors specifically include a temperature sensor, a humidity sensor, a gas sensor, an acceleration sensor, and a distance identification communication sensor, which are used to monitor environmental conditions and the hull state to assist in fishing boat environment monitoring;

[0014] The original data for the safety monitoring of fishing boat operators specifically includes camera original data and sensing original data;

[0015] The camera original data specifically includes area division data and personnel behavior data;

[0016] The sensing original data specifically includes temperature data, humidity data, gas data, fishing boat motion state data, and distance identification data.

[0017] Furthermore, in Step S2, the safety area monitoring is used to screen and pre-classify the safe operation areas of fishing boats. Specifically, based on the camera original data in the original data for the safety monitoring of fishing boat operators, a multi-level feature fusion improved visual monitoring model combined with small target optimization is used for safety area monitoring, and safety area monitoring reference data and a safety area monitoring sub-model are obtained;

[0018] The multi-level feature fusion improved visual monitoring model combined with small target optimization specifically includes an improved visual monitoring model, a spatial depth convolution conversion sub-block, and a multi-level feature fusion sub-block;

[0019] The steps of using the multi-level feature fusion improved visual monitoring model combined with small target optimization for safety area monitoring to obtain safety area monitoring reference data and a safety area monitoring sub-model include:

[0020] Step S21: Basic data optimization, specifically, data enhancement is performed on the original camera data through basic image enhancement and filtering denoising to obtain optimized safety area monitoring data;

[0021] The basic image enhancement specifically refers to data enhancement operations of random rotation, scaling, and flipping;

[0022] The filtering denoising specifically refers to Gaussian filtering operations;

[0023] Step S22: Improve the visual monitoring model. Specifically, a pre-trained standard YOLOv8 model is selected as the basic visual monitoring model, and an improved visual monitoring model is obtained by improving the hierarchical structure of the basic visual monitoring model;

[0024] The improvement of the hierarchical structure specifically refers to deleting the seventh and eighth downsampling layers of the backbone network of the standard YOLOv8 model to optimize the calculation efficiency, and adding a region monitoring optimization feature processing sub-block to the first convolutional layer of the backbone network of the standard YOLOv8 model to improve the recognition ability of the improved visual monitoring model for smaller targets in safety area monitoring;

[0025] The region monitoring optimization feature processing sub-block specifically includes a first two-dimensional convolutional layer, a second two-dimensional convolutional layer, a third two-dimensional convolutional layer, and a branch processing layer. The first two-dimensional convolutional layer and the second two-dimensional convolutional layer perform integrated feature extraction, and the obtained results are successively subjected to matrix multiplication, scale adjustment, and classification operations to obtain first region feature information. The third two-dimensional convolutional layer performs feature extraction, and the original feature information extracted by the third two-dimensional convolutional layer and the first region feature information are subjected to a second matrix multiplication. The predicted features obtained from the second matrix multiplication are randomly branched to construct a first feature branch and a second feature branch. The feature data in the first feature branch is optimized using the bounding box prediction loss, and the feature data in the second feature score is optimized using the classification loss to obtain a first optimized feature and a second optimized feature. The first optimized feature and the second optimized feature are weighted and fused to obtain region monitoring optimization feature data, and the region monitoring optimization feature data is used for feature downsampling of the improved visual monitoring model;

[0026] Step S23: Construct a spatial depth convolution conversion sub-block. Specifically, a spatial depth convolution conversion sub-block is used to replace the traditional pooling downsampling method for feature downsampling to obtain spatial depth feature conversion data;

[0027] Step S24: Construct a multi-level feature fusion sub-block. Specifically, decompose the feature map data in the spatial depth feature conversion data into three feature maps, and successively perform 1×1 convolution and two 3×3 convolution operations to obtain refined feature map data. Then, perform feature connection on the refined feature map data to obtain safety area monitoring feature data. Based on the safety area monitoring feature data, construct a fully connected layer and a classifier to perform the classification of safety area monitoring;

[0028] Step S25: Train the safety area monitoring sub-model. Specifically, through the data basic optimization, the improved visual monitoring model, the spatial depth convolution conversion sub-block, and the multi-level feature fusion sub-block, train the safety area monitoring sub-model to obtain the safety area monitoring sub-model Model SA ;

[0029] Step S26: Conduct safety area monitoring. Specifically, use the safety area monitoring sub-model Model SA , and based on the camera raw data in the original data of fishing vessel operator safety monitoring, conduct safety area monitoring to obtain safety area monitoring reference data;

[0030] The safety area monitoring reference data specifically includes the safety area monitoring category and the safety valuation of the safety area monitoring reference data;

[0031] The safety area monitoring category specifically includes high risk, medium risk, and low risk;

[0032] The value range of the safety valuation of the safety area monitoring reference data is 0 - 100%.

[0033] Furthermore, in step S3, the fishing vessel environment monitoring is used to evaluate the safety of the fishing vessel's navigation trajectory. Specifically, based on the sensing raw data in the original data of fishing vessel operator safety monitoring, use a convolutional long short-term memory neural network combined with the optimization of fishing vessel trajectory feature description to conduct fishing vessel environment monitoring to obtain fishing vessel environment monitoring reference data and a fishing vessel environment monitoring sub-model;

[0034] The clustering classification model combined with the optimization of fishing vessel trajectory feature description and convolutional long short-term memory coding specifically includes a trajectory parameter optimization sub-block, a trajectory feature optimization sub-block, a convolutional long short-term memory subnet, a clustering subnet, and a classification subnet;

[0035] The steps of using a long short-term memory neural network combined with the optimization of fishing vessel trajectory feature description to conduct fishing vessel environment monitoring to obtain fishing vessel environment monitoring reference data and a fishing vessel environment monitoring sub-model include:

[0036] Step S31: Optimize the trajectory parameters. Specifically, based on the sensing raw data in the original safety monitoring data of the fishing boat crew, perform basic data cleaning and construct the fishing boat trajectory to obtain optimized fishing boat trajectory data;

[0037] The basic data cleaning specifically refers to setting the time interval threshold of the fishing boat trajectory and cleaning abnormal data;

[0038] The setting of the time interval threshold of the fishing boat trajectory is used to maintain the continuity and non - overlap of the fishing boat trajectory data;

[0039] The cleaning of abnormal data specifically obtains optimized fishing boat trajectory data through fishing boat trajectory construction and abnormal data screening;

[0040] The construction of the fishing boat trajectory specifically regards the fishing boat as a moving - trajectory particle point, and converts the longitude and latitude coordinates of the moving - trajectory particle point into northeast - direction coordinates to construct the fishing boat trajectory;

[0041] The abnormal data screening specifically refers to calculating the motion parameters of the trajectory points according to the time stamp and deleting the speed - abnormal data points obtained by the calculation;

[0042] Step S32: Optimize the trajectory features. Specifically, based on the optimized fishing boat trajectory data, through artificial feature engineering design, construct a trajectory - feature - optimized data set, which specifically includes the fishing boat displacement - ratio feature, navigation stability feature, trajectory profile feature, and minimum moment of inertia feature;

[0043] Step S33: Convolutional long - short - term memory encoding. Specifically, construct a single - layer convolutional long - short - term memory neural subnet, perform feature encoding and feature decoding to obtain convolutional long - short - term feature decoding data;

[0044] Step S34: Construct a clustering subnet. Specifically, construct a standard deep clustering neural network based on K - means clustering as the clustering subnet, and perform clustering analysis based on the convolutional long - short - term feature decoding data to obtain clustering feature data;

[0045] Step S35: Construct a classification subnet. Specifically, construct a standard random forest model as the classifier, and based on the clustering feature data, perform fishing boat environment monitoring classification to obtain fishing boat trajectory safety classification data;

[0046] Step S36: Train the fishing boat environment monitoring sub - model. Specifically, through the above - mentioned trajectory parameter optimization, trajectory feature optimization, long - short - term memory encoding, clustering subnet, and classification subnet, train the fishing boat environment monitoring sub - model to obtain the fishing boat environment monitoring sub - model Model ED ;

[0047] Step S37: Fishing boat environment monitoring. Specifically, use the fishing boat environment monitoring sub - model ModelED , perform fishing vessel environment monitoring based on the sensing raw data in the raw data of the safety monitoring of fishing vessel operators, and obtain the reference data for fishing vessel environment monitoring;

[0048] The reference data for fishing vessel environment monitoring specifically includes the categories of fishing vessel environment monitoring and the safety valuation of the reference data for fishing vessel environment monitoring;

[0049] The categories of fishing vessel environment monitoring specifically include high risk, medium risk, and low risk;

[0050] The value range of the safety valuation of the reference data for fishing vessel environment monitoring is 0 - 100%.

[0051] Furthermore, in step S4, the personnel behavior monitoring is used to analyze the behavior of fishing vessel operators in non - safe areas. Specifically, based on the video raw data and the reference data for fishing vessel environment monitoring in the raw data of the safety monitoring of fishing vessel operators, a deep attention network combining fuzzy inference and pre - trained vision extraction is used for personnel behavior monitoring to obtain the reference data for personnel behavior monitoring and the sub - model for personnel behavior monitoring;

[0052] The deep attention network combining fuzzy inference and pre - trained vision extraction specifically includes a fuzzy layer, a visual feature extraction layer, a squeeze - and - excitation feature extraction layer, and a classification output layer;

[0053] The steps of using the deep attention network combining fuzzy inference and pre - trained vision extraction for personnel behavior monitoring to obtain the reference data for personnel behavior monitoring and the sub - model for personnel behavior monitoring include:

[0054] Step S41: Construct a fuzzy layer. Specifically, construct a Gaussian fuzzy membership function, calculate the degree of fuzziness, and suppress irrelevant features through the application of an aggregation operator or operation to obtain the output - dimension data of the fuzzy layer, which is used to optimize the recognition ability of uncertain data;

[0055] Step S42: Construct a visual feature extraction layer. Specifically, use the improved visual monitoring model in step S22 as the pre - trained visual feature extraction layer to extract the feature data for personnel behavior analysis;

[0056] Step S43: Construct a squeeze - and - excitation feature extraction layer. Specifically, perform average pooling, squeeze operation, and excitation operation on the feature data for personnel behavior analysis in sequence, and apply shortcut connection and non - linear transformation for feature data enhancement to obtain the optimized feature data for personnel behavior analysis;

[0057] Step S44: Construct a classification output layer. Specifically, construct a standard softmax classifier as the classification output layer, and classify the personnel behavior monitoring data based on the optimized feature data for personnel behavior analysis;

[0058] Step S45: Training the personnel behavior monitoring sub-model. Specifically, through the fuzzy layer, the visual feature extraction layer, the squeeze-and-excitation feature extraction layer, and the classification output layer, train the personnel behavior monitoring sub-model to obtain the personnel behavior monitoring sub-model Model BD ;

[0059] Step S46: Personnel behavior monitoring. Specifically, use the personnel behavior monitoring sub-model Model BD , and based on the camera raw data and the fishing boat environment monitoring reference data in the raw data of the fishing boat operator safety monitoring, conduct personnel behavior monitoring to obtain the personnel behavior monitoring reference data;

[0060] The personnel behavior monitoring reference data specifically includes the personnel behavior monitoring category and the safety valuation of the personnel behavior monitoring reference data;

[0061] The personnel behavior monitoring category specifically includes high risk, medium risk, and low risk;

[0062] The value range of the safety valuation of the personnel behavior monitoring reference data is 0 - 100%.

[0063] Furthermore, in step S5, the comprehensive safety monitoring is used to conduct weighted comprehensive safety monitoring by integrating three fishing boat operation safety attributes. Specifically, combine the safety area monitoring sub-model, the fishing boat environment monitoring sub-model, and the personnel behavior monitoring sub-model to conduct subnet integration weighted training to obtain the comprehensive safety monitoring model Model L , and by using the comprehensive safety monitoring model Model L , based on the raw data of the fishing boat operator safety monitoring, conduct real-time safety monitoring of the fishing boat operator to obtain the comprehensive safety situation valuation, and based on the comprehensive safety situation valuation, conduct feedback and alarm on the safety situation of the fishing boat operator;

[0064] The calculation formula for the subnet integration weighted training is:

[0065] ;

[0066] In the formula, Y L is the comprehensive safety situation valuation, used to represent the comprehensive safety monitoring output after subnet integration weighting, is the safety area monitoring weight, Y sa is the safety valuation in the safety area monitoring reference data, is the fishing boat environment monitoring weight, Y ed is the safety valuation in the fishing boat environment monitoring reference data, is the personnel behavior monitoring weight, Y bdIt is the safety valuation in the reference data of personnel behavior monitoring. Among them, .

[0067] An artificial intelligence-based safety monitoring system for fishing boat operators provided by the present invention includes a basic data collection module, a safety area monitoring module, a fishing boat environment monitoring module, a personnel behavior monitoring module, and a comprehensive safety monitoring module;

[0068] The basic data collection module is used for basic data collection. Through basic data collection, the original data of the safety monitoring of fishing boat operators is obtained, and the original data of the safety monitoring of fishing boat operators is sent to the safety area monitoring module, the fishing boat environment monitoring module, and the personnel behavior monitoring module;

[0069] The safety area monitoring module is used for safety area monitoring. Through safety area monitoring, the reference data of safety area monitoring and the sub-model of safety area monitoring are obtained, and the sub-model of safety area monitoring is sent to the comprehensive safety monitoring module;

[0070] The fishing boat environment monitoring module is used for fishing boat environment monitoring. Through fishing boat environment monitoring, the reference data of fishing boat environment monitoring and the sub-model of fishing boat environment monitoring are obtained, the reference data of fishing boat environment monitoring is sent to the personnel behavior monitoring module, and the sub-model of fishing boat environment monitoring is sent to the comprehensive safety monitoring module;

[0071] The personnel behavior monitoring module is used for personnel behavior monitoring. Through personnel behavior monitoring, the reference data of personnel behavior monitoring and the sub-model of personnel behavior monitoring are obtained, and the sub-model of personnel behavior monitoring is sent to the comprehensive safety monitoring module;

[0072] The comprehensive safety monitoring module is used for comprehensive safety monitoring. Through comprehensive safety monitoring, the comprehensive safety situation valuation is obtained, and based on the comprehensive safety situation valuation, the safety situation feedback and alarm of fishing boat operators are carried out.

[0073] The beneficial effects achieved by the present invention by adopting the above scheme are as follows:

[0074] (1) In the existing safety monitoring of fishing boat operators, traditional intelligent technology solutions often only perform single image analysis or single numerical analysis, which leads to the technical problems that the intelligent data support for the safety monitoring of fishing boat operators is inevitably insufficiently comprehensive and has low reference value. This solution creatively adopts an integrated safety analysis solution that combines a safety area monitoring sub-model, a fishing boat environment monitoring sub-model, and a personnel behavior monitoring sub-model for subnet integrated weighted training. By combining the three aspects of the fishing boat safety area, the fishing boat driving trajectory environment, and personnel behavior analysis, the comprehensiveness of the safety monitoring of fishing boat operators is improved. At the same time, through weighted processing, the intelligence and reliability of the safety monitoring of fishing boat operators are enhanced. In addition, the design of the sub-item sub-models is also conducive to the independent operation and maintenance of sub-tasks, improving the overall stability and scalability of the system;

[0075] (2) In the existing safety area monitoring methods, the existing standard image detection models have the problems of low calculation efficiency and poor recognition of small targets in dealing with complex fishing boat safety area monitoring tasks, which is likely to lead to poor safety area monitoring effects and affect the safety risks of fishing boat operators. This solution creatively adopts a method of combining an improved vision detection model with multi-level feature processing for safety area monitoring. By improving the model structure, the small target recognition ability and overall calculation efficiency are enhanced, and the overall usability of safety area monitoring is improved through multi-level feature fusion;

[0076] (3) In the existing fishing boat environment monitoring methods, traditional environment detection only conducts general intelligent analysis on the external environment and ignores the potential safety problems that may exist during the fishing boat navigation process. This solution creatively adopts a convolutional long short-term memory neural network combined with the optimization of fishing boat trajectory feature description for fishing boat environment monitoring. By designing four fishing boat navigation trajectory environment features and combining convolutional long short-term networks and clustering classification networks, complex fishing boat environment monitoring is achieved, providing strong support for subsequent comprehensive monitoring and personnel behavior monitoring;

[0077] (4) In the existing personnel behavior monitoring methods, personnel behavior monitoring involves both intuitive visual features and other potential features that cannot be directly observed through camera data. This solution creatively adopts a deep attention network combining fuzzy reasoning and pre-trained vision extraction for personnel behavior monitoring. By combining the results of fishing boat environment monitoring and the model improvement of safety area monitoring, the capture ability of the potential logic of personnel behavior analysis is improved, and the overall practicality of the safety monitoring of fishing boat operators is enhanced. Description of the Drawings

[0078] Figure 1Schematic flowchart of a method for monitoring the safety of fishing boat operators based on artificial intelligence provided by the present invention;

[0079] Figure 2 Schematic diagram of a system for monitoring the safety of fishing boat operators based on artificial intelligence provided by the present invention;

[0080] Figure 3 Schematic flowchart of the safety area monitoring in step S2;

[0081] Figure 4 Schematic flowchart of the fishing boat environment monitoring in step S3;

[0082] Figure 5 Schematic flowchart of the personnel behavior monitoring in step S4.

[0083] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. They are used together with the embodiments of the present invention to explain the present invention, but do not constitute a limitation to the present invention. Detailed implementation manners

[0084] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts belong to the scope of protection of the present invention.

[0085] In the description of the present invention, it should be understood that the terms "upper", "lower", "front", "rear", "left", "right", "top", "bottom", "inner", "outer", etc. indicating the orientation or positional relationship are based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus cannot be understood as a limitation to the present invention.

[0086] Example 1, refer to Figure 1 , A method for monitoring the safety of fishing boat operators based on artificial intelligence provided by the present invention, the method includes the following steps:

[0087] Step S1: Basic data collection;

[0088] Step S2: Safety area monitoring;

[0089] Step S3: Fishing boat environment monitoring;

[0090] Step S4: Personnel behavior monitoring;

[0091] Step S5: Comprehensive safety monitoring.

[0092] Example 2. Refer to Figure 1 、 Figure 2 and Figure 3 In step S1, the basic data collection is used to collect the original data set required for the safety monitoring of fishing boat operators. Specifically, the safety monitoring data of the fishing boat is collected through a camera device and sensors to obtain the original data of the safety monitoring of fishing boat operators;

[0093] The camera device is specifically placed in the fishing boat cockpit, fishing boat deck and fishing boat stern for safety area monitoring and personnel behavior monitoring;

[0094] The sensors specifically include a temperature sensor, a humidity sensor, a gas sensor, an acceleration sensor and a distance identification communication sensor, which are used to monitor environmental conditions and hull status to assist fishing boat environmental monitoring;

[0095] The original data of the safety monitoring of fishing boat operators specifically includes camera original data and sensing original data;

[0096] The camera original data specifically includes area division data and personnel behavior data;

[0097] The sensing original data specifically includes temperature data, humidity data, gas data, fishing boat motion state data and distance identification data.

[0098] By performing the above operations, in the existing supply chain management method, there are technical problems that supply chain data generally comes from different special processing systems, which leads to greater data complexity integration and more difficult decision-making in supply chain management. This solution creatively adopts a process design of demand forecasting, inventory management optimization and operation optimization in sequence, which can better process complex integrated supply chain data and execute various specific management tasks, provide a basis for subsequent inventory management and operation management by predicting market demand, improve the quality of supply chain management through inventory management and operation management, and at the same time, the sub-item function design also helps the operation and maintenance of the system.

[0099] Example 3. Refer to Figure 1 、 Figure 2 and Figure 3 Based on the above example, in step S2, the safety area monitoring is used to screen and pre-classify the safe operation areas of the fishing boat. Specifically, based on the camera original data in the original data of the safety monitoring of fishing boat operators, a multi-level feature fusion improved visual monitoring model combined with small target optimization is used for safety area monitoring to obtain safety area monitoring reference data and a safety area monitoring sub-model;

[0100] The improved visual monitoring model with multi-level feature fusion combined with small target optimization specifically includes an improved visual monitoring model, a spatial depth convolution conversion sub-block, and a multi-level feature fusion sub-block;

[0101] The steps of using the improved visual monitoring model with multi-level feature fusion combined with small target optimization for safety area monitoring to obtain safety area monitoring reference data and a safety area monitoring sub-model include:

[0102] Step S21: Basic data optimization, specifically, data enhancement is performed on the original camera data through basic image enhancement and filtering denoising to obtain optimized safety area monitoring data;

[0103] The basic image enhancement specifically refers to random rotation, scaling, and flipping data enhancement operations;

[0104] The filtering denoising specifically refers to Gaussian filtering operations;

[0105] Step S22: Improve the visual monitoring model. Specifically, a pre-trained standard YOLOv8 model is selected as the basic visual monitoring model, and the hierarchical structure of the basic visual monitoring model is improved to obtain an improved visual monitoring model;

[0106] The improvement of the hierarchical structure specifically means deleting the seventh and eighth downsampling layers of the backbone network of the standard YOLOv8 model to optimize the calculation efficiency, and adding a region monitoring optimization feature processing sub-block to the first convolutional layer of the backbone network of the standard YOLOv8 model to improve the recognition ability of the improved visual monitoring model for smaller targets in safety area monitoring;

[0107] The region monitoring optimization feature processing sub-block specifically includes a first two-dimensional convolutional layer, a second two-dimensional convolutional layer, a third two-dimensional convolutional layer, and a branch processing layer. The first two-dimensional convolutional layer and the second two-dimensional convolutional layer perform integrated feature extraction, and the obtained results are successively subjected to matrix multiplication, scale adjustment, and classification operations to obtain first region feature information. The third two-dimensional convolutional layer performs feature extraction, and the original feature information extracted by the third two-dimensional convolutional layer and the first region feature information are subjected to a second matrix multiplication. The predicted features obtained from the second matrix multiplication are randomly branched to construct a first feature branch and a second feature branch. The feature data in the first feature branch is optimized using the bounding box prediction loss, and the feature data in the second feature score is optimized using the classification loss to obtain a first optimized feature and a second optimized feature. The first optimized feature and the second optimized feature are weighted and fused to obtain region monitoring optimization feature data, and the region monitoring optimization feature data is used for feature downsampling of the improved visual monitoring model;

[0108] Step S23: Construct a spatial depth convolution transformation sub-block. Specifically, replace the traditional pooling downsampling method with a spatial depth convolution transformation sub-block to perform feature downsampling and obtain spatial depth feature transformation data. The calculation formula is as follows:

[0109] ;

[0110] In the formula, Y i,j,k is the feature output corresponding to the position (i, j) of the feature map and the k-th feature channel. By performing a pointwise convolution operation on the feature output Y i,j,k , spatial depth feature transformation data is obtained. i is the horizontal pixel index of the feature map, j is the vertical pixel index of the feature map, k is the feature channel index, K H is the height of the depth convolution kernel, m is the height index of the depth convolution kernel, K W is the width of the depth convolution kernel, n is the width index of the depth convolution kernel, W m,n,k is the convolution weight, X i+m,j+n+k is the spatial depth convolution input feature, which is used to represent the region monitoring optimization feature data;

[0111] Step S24: Construct a multi-level feature fusion sub-block. Specifically, decompose the feature map data in the spatial depth feature transformation data into three feature maps, and successively perform 1×1 convolution and two 3×3 convolution operations to obtain refined feature map data. Then, perform feature connection on the refined feature map data to obtain safety region monitoring feature data. And based on the safety region monitoring feature data, construct a fully connected layer and a classifier to perform the classification of safety region monitoring;

[0112] Step S25: Train the safety region monitoring sub-model. Specifically, through the data basic optimization, the improved visual monitoring model, the spatial depth convolution transformation sub-block, and the multi-level feature fusion sub-block, train the safety region monitoring sub-model to obtain the safety region monitoring sub-model Model SA ;

[0113] Step S26: Perform safety region monitoring. Specifically, use the safety region monitoring sub-model Model SA , and based on the camera raw data in the original data of fishing vessel operator safety monitoring, perform safety region monitoring to obtain safety region monitoring reference data;

[0114] The safety region monitoring reference data specifically includes the safety region monitoring category and the safety valuation of the safety region monitoring reference data;

[0115] The safety region monitoring category specifically includes high risk, medium risk, and low risk;

[0116] The value range of the security valuation of the security area monitoring reference data is 0-100%.

[0117] By performing the above operations, in the existing security area monitoring methods, there is a technical problem that in the existing standard image detection model for dealing with complex fishing vessel security area monitoring tasks, the calculation efficiency is low and the recognition of small targets is poor, which is likely to lead to poor security area monitoring effects and thus affect the safety risks of fishing vessel operators. This solution creatively uses an improved vision detection model combined with a multi-level feature processing method for security area monitoring. By improving the model structure, the small target recognition ability and overall calculation efficiency are enhanced, and the overall usability of security area monitoring is improved through multi-level feature fusion.

[0118] Example 4, refer to Figure 1 、 Figure 2 and Figure 4 Based on the above example, in step S3, the fishing vessel environment monitoring is used to evaluate the safety of the fishing vessel's navigation trajectory. Specifically, according to the sensing raw data in the raw data of fishing vessel operator safety monitoring, a convolutional long short-term memory neural network combined with the optimization of fishing vessel trajectory feature description is used for fishing vessel environment monitoring to obtain fishing vessel environment monitoring reference data and a fishing vessel environment monitoring sub-model;

[0119] The clustering classification model combined with the optimization of fishing vessel trajectory feature description and convolutional long short-term memory coding specifically includes a trajectory parameter optimization sub-block, a trajectory feature optimization sub-block, a convolutional long short-term memory subnet, a clustering subnet, and a classification subnet;

[0120] The steps of using a long short-term memory neural network combined with the optimization of fishing vessel trajectory feature description for fishing vessel environment monitoring to obtain fishing vessel environment monitoring reference data and a fishing vessel environment monitoring sub-model include:

[0121] Step S31: Trajectory parameter optimization, specifically, according to the sensing raw data in the raw data of fishing vessel operator safety monitoring, basic data cleaning and fishing vessel trajectory construction are performed to obtain optimized fishing vessel trajectory data;

[0122] The basic data cleaning specifically refers to the setting of the fishing vessel trajectory time interval threshold and the cleaning of abnormal data;

[0123] The setting of the fishing vessel trajectory time interval threshold is used to maintain the continuity and non-overlap of fishing vessel trajectory data;

[0124] The cleaning of abnormal data specifically obtains optimized fishing vessel trajectory data through fishing vessel trajectory construction and abnormal data screening;

[0125] The construction of the fishing boat trajectory specifically involves regarding the fishing boat as a moving trajectory particle, converting the longitude and latitude coordinates of the moving trajectory particle into northeast direction coordinates, and constructing the fishing boat trajectory;

[0126] The screening of abnormal data specifically refers to calculating the motion parameters of the trajectory points according to the time stamp and deleting the speed abnormal data points obtained by the calculation;

[0127] Step S32: Trajectory feature optimization, specifically, based on the optimized fishing boat trajectory data, through artificial feature engineering design, constructing a trajectory feature optimization data set, specifically including fishing boat displacement ratio feature, navigation stability feature, trajectory profile feature, and minimum moment of inertia feature;

[0128] The calculation formula of the fishing boat displacement ratio feature is:

[0129] ;

[0130] In the formula, D R is the fishing boat displacement ratio feature, T disp is the vector length from the starting trajectory point to the ending trajectory point, and T dist is the cumulative vector length of all trajectory segments;

[0131] The calculation formula of the navigation stability feature is:

[0132] ;

[0133] In the formula, C R is the navigation stability feature, T is the total number of trajectory points, t is the trajectory point index, cog t is the trajectory segment direction of the t-th trajectory point, and I t is the trajectory point duration;

[0134] The calculation formula of the trajectory profile feature is:

[0135] ;

[0136] In the formula, O R is the trajectory profile feature, N O is the number of grids covered by the trajectory segment on the fishing boat grid map, and N ng is the number of times of repeatedly entering the grid;

[0137] The calculation formula of the minimum moment of inertia feature is:

[0138] ;

[0139] In the formula, M R is the minimum moment of inertia feature, T is the total number of trajectory points, t is the trajectory point index, V is the total number of rotational inertia axes, v is the rotational inertia axis index, dtv is the distance from the t-th trajectory point to the v-th axis of rotational inertia;

[0140] The calculation formula for constructing the trajectory feature optimization data set is:

[0141] R R ={D R ;C R ;O R ;M R};

[0142] In the formula, R R is the trajectory feature optimization data set, D R is the fishing boat displacement ratio feature, C R is the sailing stability feature, O R is the trajectory profile feature, M R is the minimum moment of inertia feature;

[0143] Step S33: Convolutional long short-term memory encoding, specifically constructing a single-layer convolutional long short-term memory neural subnet, performing feature encoding and feature decoding, and obtaining convolutional long short-term feature decoding data;

[0144] Step S34: Construct a clustering subnet, specifically constructing a standard deep clustering neural network based on K-means clustering as the clustering subnet, and performing clustering analysis based on the convolutional long short-term feature decoding data to obtain clustering feature data;

[0145] Step S35: Construct a classification subnet, specifically constructing a standard random forest model as the classifier, and performing fishing boat environment monitoring classification based on the clustering feature data to obtain fishing boat trajectory safety classification data;

[0146] Step S36: Training of the fishing boat environment monitoring sub-model, specifically training the fishing boat environment monitoring sub-model through the trajectory parameter optimization, the trajectory feature optimization, the long short-term memory encoding, the clustering subnet, and the classification subnet to obtain the fishing boat environment monitoring sub-model Model ED ;

[0147] Step S37: Fishing boat environment monitoring, specifically using the fishing boat environment monitoring sub-model Model ED , and performing fishing boat environment monitoring based on the sensing raw data in the original fishing boat operator safety monitoring data to obtain fishing boat environment monitoring reference data;

[0148] The fishing boat environment monitoring reference data specifically includes the fishing boat environment monitoring category and the safety valuation of the fishing boat environment monitoring reference data;

[0149] The fishing boat environment monitoring category specifically includes high risk, medium risk, and low risk;

[0150] The safety estimate of the fishing vessel environmental monitoring reference data ranges from 0 to 100%.

[0151] By executing the above operations, in order to address the technical problem that in the existing fishing vessel environment monitoring methods, traditional environmental detection only conducts a general intelligent analysis of the external environment, while ignoring the possible safety issues in the fishing vessel navigation process itself, this scheme creatively adopts a convolutional long short-term memory neural network optimized with the fishing vessel trajectory feature description to monitor the fishing vessel environment. By designing four types of fishing vessel navigation trajectory environmental characteristics, combining the convolutional long short-term network and the clustering classification network, complex fishing vessel environment monitoring is realized, which provides strong support for subsequent comprehensive monitoring and personnel behavior monitoring.

[0152] Example 5, see Figure 1 , Figure 2 and Figure 5 This embodiment is based on the above embodiment. In step S4, the personnel behavior monitoring is used to analyze the behavior of fishing boat operators in non-safe areas. Specifically, based on the original camera data and the fishing boat environment monitoring reference data in the original data of the fishing boat operator safety monitoring, a deep attention network combining fuzzy reasoning and pre-trained visual extraction is used to perform personnel behavior monitoring to obtain personnel behavior monitoring reference data and a personnel behavior monitoring sub-model;

[0153] The deep attention network combining fuzzy reasoning and pre-trained visual extraction specifically includes a fuzzy layer, a visual feature extraction layer, a squeeze-excitation feature extraction layer, and a classification output layer;

[0154] The step of using a deep attention network combining fuzzy reasoning and pre-trained visual extraction to monitor personnel behavior and obtain personnel behavior monitoring reference data and a personnel behavior monitoring sub-model includes:

[0155] Step S41: constructing a fuzzy layer, specifically constructing a Gaussian fuzzy membership function, performing fuzziness calculation, and suppressing irrelevant features by applying an aggregation operator or operation to obtain fuzzy layer output dimension data for optimizing the recognition capability of uncertainty data;

[0156] The Gaussian blur membership function is constructed and the calculation formula for fuzziness calculation is:

[0157] ;

[0158] In the formula, is the output value of the Gaussian fuzzy membership function, which is used to apply the aggregation operator or operation and obtain the fuzzy layer output dimension data, where X is the fuzzy layer input data, which is used to represent the original camera data and the fishing vessel environment monitoring reference data after artificial feature processing and connection operation. is the mean of the Gaussian function, is the standard deviation of the Gaussian function, and exp(·) is the natural exponential function, is the index of the input data, is the index of the Gaussian fuzzy set;

[0159] Step S42: Construct a visual feature extraction layer. Specifically, use the improved visual monitoring model in Step S22 as a pre-trained visual feature extraction layer to extract the feature data for personnel behavior analysis;

[0160] Step S43: Construct a squeeze-and-excitation feature extraction layer. Specifically, perform average pooling, squeezing operation, and excitation operation on the feature data for personnel behavior analysis in sequence, and apply shortcut connection and non-linear transformation to enhance the feature data, obtaining the optimized feature data for personnel behavior analysis;

[0161] Step S44: Construct a classification output layer. Specifically, construct a standard softmax classifier as the classification output layer, and classify the monitoring data of personnel behavior based on the optimized feature data for personnel behavior analysis;

[0162] Step S45: Train the sub-model for personnel behavior monitoring. Specifically, train the sub-model for personnel behavior monitoring through the fuzzy layer, the visual feature extraction layer, the squeeze-and-excitation feature extraction layer, and the classification output layer to obtain the sub-model for personnel behavior monitoring, Model BD ;

[0163] Step S46: Monitor personnel behavior. Specifically, use the sub-model for personnel behavior monitoring, Model BD , and based on the raw camera data and the reference data for fishing vessel environment monitoring in the raw data for safety monitoring of fishing vessel operators, monitor personnel behavior to obtain the reference data for personnel behavior monitoring;

[0164] The reference data for personnel behavior monitoring specifically includes the category of personnel behavior monitoring and the safety evaluation value of the reference data for personnel behavior monitoring;

[0165] The category of personnel behavior monitoring specifically includes high risk, medium risk, and low risk;

[0166] The value range of the safety evaluation value of the reference data for personnel behavior monitoring is 0 - 100%.

[0167] By performing the above operations, in view of the technical problem that in the existing personnel behavior monitoring methods, personnel behavior monitoring involves both intuitive visual features and other potential features that cannot be directly observed through camera data, this solution creatively uses a deep attention network that combines fuzzy inference and pre-trained vision extraction for personnel behavior monitoring, combines the results of fishing boat environment monitoring and the model improvement of safety area monitoring, and together enhances the capture ability of the potential logic of personnel behavior analysis, thus enhancing the overall practicality of the safety monitoring of fishing boat operators.

[0168] Example 6, refer to Figure 1 and Figure 2 Based on the above example, in step S5, the comprehensive safety monitoring is used to perform weighted comprehensive safety monitoring for three fishing boat operation safety attributes. Specifically, it combines the safety area monitoring sub-model, the fishing boat environment monitoring sub-model, and the personnel behavior monitoring sub-model to perform subnet integration weighted training to obtain the comprehensive safety monitoring model Model L and by using the comprehensive safety monitoring model Model L , based on the original data of the safety monitoring of fishing boat operators, perform real-time safety monitoring of fishing boat operators to obtain the comprehensive safety situation estimate, and based on the comprehensive safety situation estimate, perform feedback and alarm on the safety situation of fishing boat operators;

[0169] The calculation formula for performing subnet integration weighted training is:

[0170] ;

[0171] In the formula, Y L is the comprehensive safety situation estimate, which is used to represent the output of the comprehensive safety monitoring after subnet integration weighting, is the safety area monitoring weight, Y sa is the safety estimate in the safety area monitoring reference data, is the fishing boat environment monitoring weight, Y ed is the safety estimate in the fishing boat environment monitoring reference data, is the personnel behavior monitoring weight, Y bd is the safety estimate in the personnel behavior monitoring reference data, where .

[0172] Table 1 is the reference description data table of the comprehensive safety situation estimate;

[0173] As shown in the table, the monitoring content column lists the specific task content for comprehensive safety monitoring. The valuation range column lists the specific value range of the safety valuation corresponding to the monitoring content. The remarks column lists the specific description of the safety valuation corresponding to the monitoring content. The alarm level column lists the alarm level for feedback on the safety situation and alarm of the safety valuation corresponding to the monitoring content. When the alarm level value is high, a safety alarm is issued. When the alarm level value is medium, a safety situation feedback is provided.

[0174] Table 1 Data Sheet for Reference Explanation of Comprehensive Safety Situation Valuation

[0175]

[0176] Example Seven. Refer to Figure 1 and Figure 2 This embodiment is based on the above-mentioned embodiment. An artificial intelligence-based safety monitoring system for fishing boat operators provided by the present invention includes a basic data collection module, a safety area monitoring module, a fishing boat environment monitoring module, a personnel behavior monitoring module, and a comprehensive safety monitoring module.

[0177] The basic data collection module is used for basic data collection. Through basic data collection, the original data for safety monitoring of fishing boat operators is obtained, and the original data for safety monitoring of fishing boat operators is sent to the safety area monitoring module, the fishing boat environment monitoring module, and the personnel behavior monitoring module.

[0178] The safety area monitoring module is used for safety area monitoring. Through safety area monitoring, safety area monitoring reference data and a safety area monitoring sub-model are obtained, and the safety area monitoring sub-model is sent to the comprehensive safety monitoring module.

[0179] The fishing boat environment monitoring module is used for fishing boat environment monitoring. Through fishing boat environment monitoring, fishing boat environment monitoring reference data and a fishing boat environment monitoring sub-model are obtained, and the fishing boat environment monitoring reference data is sent to the personnel behavior monitoring module, and the fishing boat environment monitoring sub-model is sent to the comprehensive safety monitoring module.

[0180] The personnel behavior monitoring module is used for personnel behavior monitoring. Through personnel behavior monitoring, personnel behavior monitoring reference data and a personnel behavior monitoring sub-model are obtained, and the personnel behavior monitoring sub-model is sent to the comprehensive safety monitoring module.

[0181] The comprehensive safety monitoring module is used for comprehensive safety monitoring. Through comprehensive safety monitoring, a comprehensive safety situation valuation is obtained, and based on the comprehensive safety situation valuation, feedback on the safety situation of fishing boat operators and alarm are carried out.

[0182] By performing the above operations, in the existing safety monitoring of fishing boat operators, there is a problem that traditional intelligent technology solutions often only perform single image analysis or single numerical analysis. This leads to the technical problems that the intelligent data support for the safety monitoring of fishing boat operators is inevitably insufficiently comprehensive and has low reference value. This solution creatively adopts an integrated safety analysis scheme that combines a safety area monitoring sub-model, a fishing boat environment monitoring sub-model, and a personnel behavior monitoring sub-model for subnet integrated weighted training. By combining the three aspects of the fishing boat safety area, the fishing boat driving trajectory environment, and personnel behavior analysis, the comprehensiveness of the safety monitoring of fishing boat operators is improved. At the same time, through weighted processing, the intelligence and reliability of the safety monitoring of fishing boat operators are improved. In addition, the design of the sub-item sub-models is also beneficial to the independent operation and maintenance of sub-tasks, improving the overall stability and scalability of the system.

[0183] It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or elements inherent to such process, method, article or device.

[0184] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention.

[0185] The above describes the present invention and its implementation manners. This description is not restrictive. What is shown in the drawings is only one of the implementation manners of the present invention, and the actual structure is not limited thereto. Generally speaking, if those of ordinary skill in the art are inspired by it and design similar structural manners and embodiments without creative efforts without departing from the purpose of the present invention, they shall fall within the protection scope of the present invention.

Claims

1. A safety monitoring method for fishing vessel operators based on artificial intelligence, characterized in that: The method comprises the following steps: Step S1: basic data collection, which is used to collect the original data set required for the safety monitoring of fishing vessel operators, specifically by collecting fishing vessel safety monitoring data through a camera device and a sensor, and obtaining the original data for the safety monitoring of fishing vessel operators; the original data for the safety monitoring of fishing vessel operators specifically includes the original camera data and the original sensor data; Step S2: Safety area monitoring, which is used to screen and pre-classify the safety areas for fishing boat operations, and improve the visual monitoring model by combining the multi-level feature fusion with small target optimization. The multi-level feature fusion improved visual monitoring model combined with small target optimization is used to monitor the safety area, and obtain the safety area monitoring reference data and the safety area monitoring sub-model; The multi-level feature fusion improved visual monitoring model combined with small target optimization specifically includes an improved visual monitoring model, a spatial depth convolution conversion sub-block and a multi-level feature fusion sub-block; The step of using a multi-level feature fusion improved visual monitoring model combined with small target optimization to monitor the safe area and obtain safe area monitoring reference data and a safe area monitoring sub-model includes: Step S21: basic data optimization, specifically, performing data enhancement on the original camera data through basic image enhancement and filtering denoising to obtain optimized safe area monitoring data; Step S22: improving the visual monitoring model, specifically selecting a pre-trained standard YOLOv8 model as a basic visual monitoring model, and improving the hierarchical structure of the basic visual monitoring model to obtain an improved visual monitoring model; The hierarchical structure improvement specifically refers to deleting the seventh and eighth downsampling layers of the standard YOLOv8 model backbone network to optimize computational efficiency, and adding a regional monitoring optimization feature processing sub-block to the first convolutional layer of the standard YOLOv8 model backbone network to enhance the improved visual monitoring model's ability to recognize smaller targets in safe area monitoring; Step S23: constructing a spatial depth convolution conversion sub-block, specifically using the spatial depth convolution conversion sub-block to replace the traditional pooling downsampling method, performing feature downsampling, and obtaining spatial depth feature conversion data; Step S24: construct a multi-level feature fusion sub-block, specifically, decomposing the feature map data in the spatial depth feature conversion data into three feature maps, and sequentially performing 1×1 convolution and two 3×3 convolution operations to obtain refined feature map data, and performing feature connection on the refined feature map data to obtain safe area monitoring feature data, and classifying safe area monitoring by constructing a fully connected layer and a classifier based on the safe area monitoring feature data; Step S25: training the safety area monitoring sub-model, specifically, training the safety area monitoring sub-model through the data basic optimization, the improved visual monitoring model, the spatial depth convolution conversion sub-block and the multi-level feature fusion sub-block to obtain the safety area monitoring sub-model Model SA ; Step S26: Safe area monitoring, specifically using the safe area monitoring sub-model Model SA , based on the original camera data in the original data of safety monitoring of fishing vessel operators, safety area monitoring is performed to obtain safety area monitoring reference data; The safety area monitoring reference data specifically includes the safety area monitoring category and the safety valuation of the safety area monitoring reference data; Step S3: fishing vessel environment monitoring, which is used to conduct safety assessment on the navigation track of the fishing vessel, specifically, based on the sensor raw data in the fishing vessel operator safety monitoring raw data, a convolutional long short-term memory neural network optimized in combination with the fishing vessel trajectory feature description is used to conduct fishing vessel environment monitoring, and obtain fishing vessel environment monitoring reference data and fishing vessel environment monitoring sub-model; The clustering classification model combining the optimization of fishing vessel trajectory feature description and convolutional long short-term memory encoding specifically includes a trajectory parameter optimization sub-block, a trajectory feature optimization sub-block, a convolutional long short-term memory subnet, a clustering subnet and a classification subnet; The step of using the long short-term memory neural network optimized in combination with the fishing vessel trajectory feature description to monitor the fishing vessel environment and obtain the fishing vessel environment monitoring reference data and the fishing vessel environment monitoring sub-model includes: Step S31: trajectory parameter optimization, specifically, performing basic data cleaning and fishing boat trajectory construction based on the sensor raw data in the fishing boat operator safety monitoring raw data to obtain optimized fishing boat trajectory data; Step S32: trajectory feature optimization, specifically, constructing a trajectory feature optimization data set based on the optimized fishing boat trajectory data through artificial feature engineering design, specifically including fishing boat displacement ratio feature, navigation stability feature, trajectory profile feature and minimum moment of inertia feature; The calculation formula of the displacement ratio characteristic of the fishing boat is: Where D R is the displacement ratio characteristic of the fishing boat, T disp is the vector length from the starting point to the ending point, T dist is the cumulative vector length of all trajectory segments; The calculation formula of the navigation stability characteristic is: In the formula, C R is the navigation stability characteristic, T is the total number of track points, t is the track point index, cog t is the trajectory segment direction of the t-th trajectory point, I t is the trajectory point duration; The calculation formula of the trajectory profile feature is: In the formula, O R is the trajectory profile feature, N O is the number of grids covered by the track segment on the fishing boat grid map, N ng is the number of repeated entries into the grid; The calculation formula of the minimum moment of inertia characteristic is: Where M R is the minimum moment of inertia feature, T is the total number of trajectory points, t is the trajectory point index, V is the total amount of the rotational inertia axis, v is the rotational inertia axis index, d tv is the distance from the tth trajectory point to the vth axis of inertia; The calculation formula for constructing the trajectory feature optimization data set is: R R ={D R ;C R ;O R ;M R }; In the formula, R R is the trajectory feature optimization dataset, D R is the displacement ratio characteristic of the fishing boat, C R is the navigation stability characteristic, O R is the trajectory profile feature, M R is the minimum moment of inertia characteristic; Step S33: convolutional long short-term memory encoding, specifically constructing a single-layer convolutional long short-term memory neural subnet, performing feature encoding and feature decoding, and obtaining convolutional long short-term feature decoding data; Step S34: constructing a clustering subnet, specifically constructing a standard deep clustering neural network based on K-means clustering as a clustering subnet, and performing cluster analysis based on the convolutional long-term and short-term feature decoding data to obtain clustering feature data; Step S35: constructing a classification subnet, specifically constructing a standard random forest model as a classifier, and classifying the fishing vessel environment monitoring based on the clustering feature data to obtain classification data on the safety of the fishing vessel trajectory; Step S36: Fishing vessel environment monitoring sub-model training, specifically, through the trajectory parameter optimization, the trajectory feature optimization, the long short-term memory encoding, the clustering sub-network and the classification sub-network, the fishing vessel environment monitoring sub-model training is performed to obtain the fishing vessel environment monitoring sub-model Model ED ; Step S37: Fishing vessel environment monitoring, specifically using the fishing vessel environment monitoring sub-model Model ED , based on the sensor raw data in the fishing vessel operator safety monitoring raw data, the fishing vessel environment monitoring is performed to obtain the fishing vessel environment monitoring reference data; The fishing vessel environmental monitoring reference data specifically includes the fishing vessel environmental monitoring category and the safety valuation of the fishing vessel environmental monitoring reference data Step S4: Personnel behavior monitoring, which is used to analyze the behavior of fishing boat operators in non-safe areas. Specifically, based on the original camera data and the fishing boat environment monitoring reference data in the original data of fishing boat operator safety monitoring, a deep attention network combining fuzzy reasoning and pre-trained visual extraction is used to perform personnel behavior monitoring, and obtain personnel behavior monitoring reference data and a personnel behavior monitoring sub-model; The deep attention network combining fuzzy reasoning and pre-trained visual extraction specifically includes a fuzzy layer, a visual feature extraction layer, a squeeze-excitation feature extraction layer, and a classification output layer; The step of using a deep attention network combining fuzzy reasoning and pre-trained visual extraction to monitor personnel behavior and obtain personnel behavior monitoring reference data and a personnel behavior monitoring sub-model includes: Step S41: constructing a fuzzy layer, specifically constructing a Gaussian fuzzy membership function, performing fuzziness calculation, and suppressing irrelevant features by applying an aggregation operator or operation to obtain fuzzy layer output dimension data for optimizing the recognition capability of uncertainty data; Step S42: constructing a visual feature extraction layer, specifically using the improved visual monitoring model in step S22 as a pre-trained visual feature extraction layer to extract personnel behavior analysis feature data; Step S43: constructing a squeeze-excitation feature extraction layer, specifically, performing average pooling, squeeze operations and excitation operations on the personnel behavior analysis feature data in sequence, and applying shortcut connections and nonlinear transformations to enhance the feature data to obtain optimized personnel behavior analysis features; Step S44: constructing a classification output layer, specifically constructing a standard softmax classifier as a classification output layer, and classifying the personnel behavior monitoring data according to the optimized personnel behavior analysis features; Step S45: training the sub-model for monitoring human behavior, specifically, training the sub-model for monitoring human behavior through the fuzzy layer, the visual feature extraction layer, the squeeze excitation feature extraction layer and the classification output layer, to obtain the sub-model for monitoring human behavior Model BD ; Step S46: Personnel behavior monitoring, specifically using the personnel behavior monitoring sub-model Model BD , based on the original camera data and the fishing vessel environment monitoring reference data in the original data of the fishing vessel operator safety monitoring, personnel behavior monitoring is performed to obtain personnel behavior monitoring reference data; The personnel behavior monitoring reference data specifically includes the personnel behavior monitoring category and the safety valuation of the personnel behavior monitoring reference data; Step S5: Comprehensive safety monitoring: Combine the safety area monitoring sub-model, the fishing vessel environment monitoring sub-model and the personnel behavior monitoring sub-model to perform sub-network integrated weighted training to obtain a comprehensive safety monitoring model Model L , and by using the comprehensive safety monitoring model Model L Based on the original data of safety monitoring of fishing vessel operators, real-time safety monitoring of fishing vessel operators is carried out to obtain a comprehensive safety situation estimation, and safety situation feedback and alarm of fishing vessel operators are carried out based on the comprehensive safety situation estimation.

2. The artificial intelligence-based safety monitoring method for fishing vessel operators according to claim 1 is characterized in that: In step S5, the comprehensive safety monitoring is used to conduct weighted comprehensive safety monitoring by integrating the three fishing vessel operation safety attributes. Specifically, the subnet integration weighted training is performed by combining the safety area monitoring submodel, the fishing vessel environment monitoring submodel and the personnel behavior monitoring submodel to obtain a comprehensive safety monitoring model Model L , and by using the comprehensive safety monitoring model Model L Based on the original data of safety monitoring of fishing vessel operators, real-time safety monitoring of fishing vessel operators is carried out to obtain a comprehensive safety situation estimation, and safety situation feedback and alarm of fishing vessel operators are carried out based on the comprehensive safety situation estimation.

3. The artificial intelligence-based safety monitoring method for fishing vessel operators according to claim 2 is characterized in that: In step S5, the calculation formula for performing subnet integrated weighted training is: THE L =αY sa +βY ed +γY bd 4 Where Y L is the comprehensive security situation estimation, which is used to represent the comprehensive security monitoring output after subnet integration and weighting, α is the security area monitoring weight, and Y sa is the safety estimate in the safety area monitoring reference data, β is the fishing vessel environmental monitoring weight, Y ed is the safety estimate in the reference data of fishing vessel environment monitoring, γ is the weight of personnel behavior monitoring, and Y bd It is the safety estimate in the personnel behavior monitoring reference data.

4. An artificial intelligence-based fishing vessel operator safety monitoring system, used to implement an artificial intelligence-based fishing vessel operator safety monitoring method as described in any one of claims 1 to 3, characterized in that: It includes basic data collection module, safety area monitoring module, fishing vessel environment monitoring module, personnel behavior monitoring module and comprehensive safety monitoring module; The basic data acquisition module is used for basic data acquisition, and obtains the original data of safety monitoring of fishing vessel operators through basic data acquisition, and sends the original data of safety monitoring of fishing vessel operators to the safety area monitoring module, the fishing vessel environment monitoring module and the personnel behavior monitoring module; The safety area monitoring module is used for safety area monitoring, obtains safety area monitoring reference data and safety area monitoring sub-model through safety area monitoring, and sends the safety area monitoring sub-model to the comprehensive safety monitoring module; The fishing vessel environment monitoring module is used for fishing vessel environment monitoring. Through fishing vessel environment monitoring, fishing vessel environment monitoring reference data and fishing vessel environment monitoring sub-model are obtained, and the fishing vessel environment monitoring reference data is sent to the personnel behavior monitoring module, and the fishing vessel environment monitoring sub-model is sent to the comprehensive safety monitoring module; The personnel behavior monitoring module is used for personnel behavior monitoring, obtains personnel behavior monitoring reference data and personnel behavior monitoring sub-model through personnel behavior monitoring, and sends the personnel behavior monitoring sub-model to the comprehensive safety monitoring module; The comprehensive safety monitoring module is used for comprehensive safety monitoring. Through comprehensive safety monitoring, a comprehensive safety situation estimation is obtained, and safety situation feedback and alarms are provided to fishing vessel operators based on the comprehensive safety situation estimation.

Images