An underwater target detection and recognition simulation system and method

Abstract

The present invention discloses an underwater target detection and recognition simulation system and method. The system takes the detection information collaborative processing module and the target detection result classification and recognition module as the core, combines the simulation components in the high-resolution model library, completes the simulation of the process of underwater target collaborative detection and recognition, and realizes a simulation system for underwater target collaborative detection and recognition.

Description

Technical Field

[0001] The present invention relates to the technical field of underwater detection, and particularly to an underwater target detection and recognition simulation system and method. Background Art

[0002] With the development of information-based naval warfare, underwater perception and collaborative detection technologies are becoming increasingly important. However, the marine environment is changeable, and at the same time, the underwater space has properties and particularities that are very different from other spaces. Obtaining sufficient multi-source detection information from the real environment requires a high cost.

[0003] With the development of computer technology, simulation analysis and application verification technologies that can help researchers, operators, etc. carry out "advanced practice" have been widely used in various countries around the world. However, the current simulation system, as a general system, includes multiple physical domains such as "land, sea, air, space, electricity, and network", has poor adaptability to the underwater detection field, low model resolution, and the system nodes do not have the ability of collaborative detection and recognition. Summary of the Invention

[0004] The technical problem to be solved by the present invention is that the current simulation system, as a general system, includes multiple physical domains such as "land, sea, air, space, electricity, and network", has poor adaptability to the underwater detection field, low model resolution, and the system nodes do not have the ability of collaborative detection and recognition. Therefore, the present invention provides an underwater target detection and recognition simulation system and method. On the basis of the existing simulation system, considering the actual processing process of underwater target distributed detection and recognition, by establishing a high-resolution model library for underwater collaborative detection and recognition, a detection information collaborative processing module, and an underwater target detection result classification and recognition module, the simulation simulation for underwater target collaborative detection and recognition can be realized, providing an effective solution for the simulation of underwater target distributed collaborative detection processing.

[0005] The present invention is realized by the following technical solutions:

[0006] An underwater target detection and recognition simulation system, comprising:

[0007] A model establishment and application module, configured to call corresponding simulation components from a high-resolution model library according to simulation requirements to generate a target detection model; after obtaining the target detection model, input the data to be simulated in the simulation requirements into the target detection model for processing, and output detection information;

[0008] A detection information collaborative processing module, configured to perform fusion processing on the detection information output by the target detection model, generate a target detection result, and send it to a simulation detection platform;

[0009] A target detection result classification and recognition module, configured to perform recognition processing and classification on the target detection result generated by the detection information collaborative processing module, obtain target recognition information, and send it to the simulation detection platform;

[0010] The simulation detection platform is used to send the received detection information and target recognition information to the data recording module for recording.

[0011] Further, the model establishment and application module includes:

[0012] The simulation component acquisition unit is used to call the corresponding simulation components from the high-resolution model library according to the simulation components required in the simulation requirements;

[0013] The simulation component assembly unit is used to assemble and associate the called simulation components according to the assembly relationship and association relationship in the simulation requirements to obtain the target detection model.

[0014] Further, the high-resolution model library includes a detection node model, a detection target model, a battlefield interference model, a sensor array model, and a network model;

[0015] Among them, the detection node model includes a sonar simulation component, a buoy and submersible buoy simulation component, an unmanned underwater vehicle simulation component, and a surface ship simulation component;

[0016] The detection target model includes a ship simulation component, a frogman simulation component, a submarine simulation component, and an unmanned underwater vehicle simulation component;

[0017] The battlefield interference model includes a Doppler effect model, a self-noise interference model, an underwater acoustic propagation model, a reverberation model, and an ocean ambient noise model;

[0018] The sensor array model includes a sensor line array simulation component, a sensor volume array simulation component, and a sensor cylindrical array simulation component;

[0019] The network model includes a time slot control simulation model, a network management simulation model, and a link transmission simulation model.

[0020] Further, the detection information includes the position information, speed information, and status information of the detection target; the detection information collaborative processing module includes:

[0021] The detection information spatial alignment unit is used to perform spatial alignment on the acquired position information, speed information, and status information by using a data alignment function, and filter the spatially aligned data through the Kalman filter algorithm to obtain the filtered data;

[0022] The detection information comprehensive processing unit is used to perform track preprocessing, track association processing, and track fusion processing on the filtered data to obtain the target fusion track;

[0023] An underwater target threat ranking unit, which is used to construct an underwater target threat ranking model based on the target fusion track, calculate the underwater target threat degree, and rank the underwater target threat degree to form a threat ranking result as the target detection result.

[0024] Further, the target detection result classification and recognition module includes:

[0025] A data preprocessing unit, which is used to call a preprocessing function, extract the original signal from the target detection result generated by the detection information collaborative processing module, and perform spatial alignment preprocessing on the extracted original signal to obtain preprocessed data;

[0026] A feature extraction unit, which is used to call a feature extraction function to extract features from the preprocessed data in the time domain, frequency domain, and time-frequency joint domain to obtain multi-dimensional features;

[0027] A feature selection unit, which is used to perform feature selection on the extracted multi-dimensional features according to the feature selection conditions in the simulation requirements to obtain effective features;

[0028] An effective feature processing unit, which is used to perform correlation analysis on the selected effective features to obtain correlation features, and call a target recognition algorithm to classify the correlation features to obtain a target classification result.

[0029] Further, the feature extraction unit includes:

[0030] A time-domain feature extraction unit, which is used to perform mean square value, peak value, and variance processing on the time-domain data of the preprocessed data to obtain time-domain features;

[0031] A frequency-domain feature extraction unit, which is used to extract features from the preprocessed data by using a Fourier transform function to obtain frequency-domain features;

[0032] A time-frequency feature extraction unit, which is used to extract features from the preprocessed data by using a short-time Fourier transform function to obtain time-frequency features.

[0033] Further, the underwater target detection and recognition simulation system further includes a comprehensive situation display module;

[0034] The comprehensive situation display module is used to display the detection information, target detection result, and target recognition information of the target detection model.

[0035] Further, the underwater target detection and recognition simulation system further includes a data storage module;

[0036] The data storage module is used to store the data in the model establishment and application module, detection information collaborative processing module, target detection result classification and recognition module, simulation detection platform, and data recording module.

[0037] Furthermore, the modules in the underwater target detection and recognition simulation system are scheduled and data is transmitted through an interface method.

[0038] A detection method based on the above underwater target detection and recognition simulation system includes:

[0039] Obtain underwater detection information according to the detection node model, and perform fusion processing on the obtained underwater detection information to obtain a fusion trajectory;

[0040] Construct an underwater target threat ranking model based on the fusion track, calculate the underwater target threat degree, and rank the underwater target threat degree to form a threat ranking result as the target detection result;

[0041] Call the preprocessing function to extract the original signal from the target detection result generated by the detection information collaborative processing module, and perform spatial alignment preprocessing on the extracted original signal to obtain preprocessed data;

[0042] Through the feature extraction function, extract features from the preprocessed data in the time domain, frequency domain, and time-frequency joint domain to obtain multi-dimensional features, and perform feature selection on the extracted multi-dimensional features according to the feature selection conditions in the simulation requirements to obtain effective features;

[0043] Perform correlation analysis on the obtained effective features to obtain correlation features, and call the target recognition algorithm to classify the correlation features to obtain the target classification result.

[0044] The present invention provides an underwater target detection and recognition simulation system and method. By designing a simulation platform for underwater target detection and recognition, it provides effective demonstration and evaluation means and a demonstration and analysis research platform for the structural composition, mode demonstration, collaborative effectiveness analysis, and key technology development analysis of the underwater network system, promoting the development of underwater equipment intelligence. In addition to having the support functions of a general simulation system, this system has established a high-resolution model library in the underwater field, a detection information collaborative processing module, a target detection result classification and recognition module, and a data recording module, solving the problems of poor adaptability and low resolution between the general simulation system and the underwater field, and the lack of collaborative detection and recognition capabilities of system nodes, providing an effective solution for the simulation of distributed collaborative detection and processing of underwater targets. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] The drawings described herein are used to provide a further understanding of the embodiments of the present invention, form a part of this application, and do not limit the embodiments of the present invention. In the drawings:

[0046] Figure 1 It is the schematic diagram of an underwater target detection and recognition simulation system of the present invention.

[0047] Figure 2 Schematic diagram of a high-resolution model library in an embodiment of the present invention.

[0048] Figure 3 Flowchart of a detection method for an underwater target detection and recognition simulation system of the present invention. Detailed implementation manners

[0049] To make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below in conjunction with embodiments and drawings. The illustrative embodiments of the present invention and their descriptions are only used to explain the present invention and do not limit the present invention.

[0050] Embodiment 1

[0051] As Figure 1 shown, the present invention provides an underwater target detection and recognition simulation system, including:

[0052] A model establishment and application module, configured to call corresponding simulation components from a high-resolution model library according to simulation requirements to generate a target detection model; after obtaining the target detection model, input the data to be simulated in the simulation requirements into the target detection model for processing, and output detection information.

[0053] A detection information collaborative processing module, configured to perform fusion processing on the detection information output by the target detection model, generate a target detection result, and send it to the simulation detection platform.

[0054] A target detection result classification and recognition module, configured to perform recognition processing and classification on the target detection result generated by the detection information collaborative processing module, obtain target recognition information, and send it to the simulation detection platform.

[0055] A simulation detection platform, configured to send the received detection information and target recognition information to a data recording module for recording.

[0056] In this embodiment, the model establishment and application module and the detection information collaborative processing module are set not to directly interact with the data recording module to reduce the coupling of the simulation system and effectively prevent data loss when one of the modules has a problem. If the information interaction is directly carried out between each module, it will lead to a high coupling between the modules. When adding a new functional module later, the logic of information interaction between the modules will also be relatively complex. When upgrading the system function or iterating the version update, the system will also be bloated and difficult to maintain. In addition, it will also cause data loss of the entire system if one of the modules has a problem.

[0057] Furthermore, as Figure 2As shown in the figure, the high-resolution model library in this embodiment includes a detection node model, a detection target model, a battlefield interference model, a sensor array model, and a network model;

[0058] Among them, the detection node model includes a sonar simulation component, a buoy and submersible buoy simulation component, an unmanned underwater vehicle simulation component, and a surface ship simulation component; the detection target model includes a ship simulation component, a frogman simulation component, a submarine simulation component, and an unmanned underwater vehicle simulation component; the battlefield interference model includes a Doppler effect model, a self-noise interference model, an underwater acoustic propagation model, a reverberation model, and a marine environmental noise model; the sensor array model includes a sensor line array simulation component, a sensor volume array simulation component, and a sensor cylindrical array simulation component; the network model includes a time slot control simulation model, a network management simulation model, and a link transmission simulation model.

[0059] Furthermore, the model establishment and application module includes a simulation component acquisition unit and a simulation component assembly unit.

[0060] The simulation component acquisition unit is used to call the corresponding simulation components from the high-resolution model library according to the simulation components required in the simulation requirements.

[0061] The simulation component assembly unit is used to assemble and associate the called simulation components according to the assembly relationship and association relationship in the simulation requirements to obtain a target detection model.

[0062] Furthermore, the detection information includes the position information, speed information, and status information of the detection target.

[0063] The detection information collaborative processing module includes: a detection information spatial alignment unit, a detection information comprehensive processing unit, and an underwater target threat ranking unit.

[0064] The detection information spatial alignment unit is used to perform spatial alignment on the obtained position information, speed information, and status information by using a data alignment function, and filter the spatially aligned data through a Kalman filtering algorithm to obtain the filtered data, so as to achieve the unity of feature expression.

[0065] Among them, the data alignment function adopted in this embodiment includes but is not limited to the possic_memalign() function of POSIX

[0066] The detection information comprehensive processing unit is used to perform track preprocessing, track association processing, and track fusion processing on the filtered data to obtain a target fusion track.

[0067] Among them, in this embodiment, the track preprocessing adopts a combination of the classic "distance segmentation" and "time segmentation" methods, segments the track points with a time difference greater than the preset time difference or a distance difference greater than the preset distance difference, removes the stop points, and filters the fluctuation points.

[0068] The track association processing adopts the nearest neighbor association method, using the geometric vector distance as the similarity metric. The main processing process is as follows: establish an association gate, determine the association threshold, threshold filtering, similarity metric, establish an association matrix, determine the association method, and form association pairs.

[0069] Track fusion processing refers to the process of fusing the target positions obtained by sensors to form a new and more accurate fusion track, which makes the target position information more stable and reliable. The track fusion algorithm in this embodiment specifically adopts the convex combination fusion algorithm.

[0070] The underwater target threat ranking unit is used to construct an underwater target threat ranking model based on the target fusion track, calculate the underwater target threat degree, and rank the underwater target threat degree to form a threat ranking result as the target detection result.

[0071] Among them, the underwater target threat ranking model refers to a model that calculates the underwater target threat degree based on the fusion track and ranks the magnitude of the threat degree.

[0072] Specifically, after obtaining the target fusion track, threat factors such as target distance, depth, heading angle, speed, and maneuverability are extracted from the target fusion track and calculated, and the weight calculation method is used to calculate each threat factor to obtain the target threat degree. After obtaining the target threat degree, it is sorted in descending order according to the magnitude of the target threat degree to form a threat ranking result.

[0073] Furthermore, the target detection result classification and recognition module includes: a data preprocessing unit, a feature extraction unit, a feature selection unit, and an effective feature processing unit.

[0074] The data preprocessing unit is used to call the preprocessing function, extract the original signal from the target detection result generated by the detection information collaborative processing module, and perform spatial alignment preprocessing on the extracted original signal to obtain preprocessed data.

[0075] In this embodiment, the function for performing spatial alignment preprocessing on the extracted original signal adopts the possic_memalign() function.

[0076] The feature extraction unit is used to call the feature extraction function to extract features from the preprocessed data in the time domain, frequency domain, and time-frequency joint domain to obtain multi-dimensional features.

[0077] The data obtained by extracting features from the preprocessed data in the time domain is used as time-domain features; the data obtained by extracting features from the preprocessed data in the frequency domain is used as frequency-domain features; the data obtained by extracting features from the preprocessed data in the time-frequency domain is used as time-frequency features. Among them, the time-domain features are used to represent the waveform structure features of the signal, the frequency-domain features are used to represent the spectral features of the signal, and the time-frequency domain features are used to represent the characteristics of the signal energy changing with time.

[0078] The feature extraction functions in this embodiment include the Fourier transform function and the short-time Fourier transform function.

[0079] Specifically, after obtaining the preprocessed data, perform mean square value, peak value, and variance processing on the time-domain data of the preprocessed data to obtain time-domain features, then use the Fourier function to convert the time-domain features into frequency-domain features, and finally use the short-time Fourier transform function to convert the frequency-domain features into time-frequency features.

[0080] The feature selection unit is used to perform feature selection on the extracted multi-dimensional features according to the feature selection conditions in the simulation requirements to obtain effective features.

[0081] The feature selection conditions in this embodiment refer to the feature selection conditions formed according to the mapping relationship between features and targets.

[0082] The effective feature processing unit is used to perform correlation analysis on the selected effective features to obtain correlation features, and call the target recognition algorithm to classify the correlation features to obtain the target classification result.

[0083] Specifically, this embodiment uses the Apriori algorithm to perform correlation analysis on the selected effective features to obtain correlation features, mine frequent item sets from the correlation features, and then extract strong association rules of things from the frequent item sets to assist in decision-making.

[0084] The target recognition algorithm in this embodiment uses a multi-class neural network model. Among them, the multi-class neural network model refers to a neural network model that can perform multiple category recognitions.

[0085] Specifically, collect a large number of correlation features and the labels corresponding to the correlation features as training samples, and divide the training samples into a training set, a test set, and a validation set, and perform training, testing, and validation on the selected multi-class neural network model to obtain a model that can classify correlation features as the target recognition algorithm in this embodiment. After obtaining the target recognition algorithm, classify the correlation features through the target recognition algorithm to obtain the target classification result.

[0086] Furthermore, the feature extraction unit includes: a time-domain feature extraction unit, a frequency-domain feature extraction unit, and a time-frequency feature extraction unit.

[0087] The time-domain feature extraction unit is used to perform mean square value, peak value, and variance processing on the time-domain data of the preprocessed data to obtain time-domain features;

[0088] The frequency-domain feature extraction unit is used to perform feature extraction on the preprocessed data by using the Fourier transform function to obtain frequency-domain features;

[0089] The time-frequency feature extraction unit is used to perform feature extraction on the preprocessed data by using the short-time Fourier transform function to obtain time-frequency features.

[0090] Furthermore, the data recording module is used to record all the data involved in the simulation system through the data storage access adapter according to the recording control instruction

[0091] Furthermore, the underwater target detection and recognition simulation system further includes a comprehensive situation display module.

[0092] The comprehensive situation display module is used to display the detection information, target detection results, and target recognition information of the target detection model. The display methods of the comprehensive situation display module in this embodiment include two-dimensional display method and three-dimensional display method.

[0093] Furthermore, the underwater target detection and recognition simulation system further includes a data storage module.

[0094] The data storage module is used to store the data in the model establishment and application module, the detection information collaborative processing module, the target detection result classification and recognition module, the simulation detection platform, and the data recording module.

[0095] Furthermore, from the perspective of reducing the software module coupling degree and facilitating software maintenance and update, the above-mentioned modules are scheduled and data-transmitted through an interface method.

[0096] Taking the detection target model as a warship, the detection node model as a passive detection sonar, and the sensor array model as a line array as an example, according to the ocean environment and combat missions, the model establishment and application module allocates network resources and determines the model topology. Suppose an enemy warship approaches our side, and the warship target radiates noise outward. This signal propagates through the sound field to our passive sonar. After the sensor array model receives the signal, it combines the ocean environment noise model, underwater acoustic propagation model, and Doppler effect model in the battlefield interference model to calculate the received signal and obtain detection information such as target position information, speed information, and status information. These detection information are transmitted to the detection information collaborative processing module through the simulation detection platform data bus. This module receives the detection information of each detection node, uses the fusion algorithm in the algorithm library to fuse multi-source information, then comprehensively processes the detection data, and reports intelligence externally. The detection information obtained by each detection node or the comprehensive information after fusion processing is transmitted to the target detection result classification and recognition module. This module calls the information preprocessing function in the algorithm library to perform data preprocessing work, then calls the feature extraction related functions in the algorithm library to extract features. After extracting multi-dimensional features, feature selection is performed, and correlation analysis is carried out on the selected multi-modal features. Next, the target recognition algorithm is called in the algorithm library for target classification. Finally, the target classification result is reported, displayed, and stored in the database. The data recording module records all the data involved in the simulation system through the data storage access adapter according to the recording control instruction. The comprehensive situation display module receives the target detection results reported by the detection information collaborative processing module, the target recognition information reported by the target detection result classification and recognition module, and the detection information output by the target detection model reported by the model establishment and application module.

[0097] The detection information collaborative processing module uses the functional simulation method to simulate the processing functions and technical characteristics of the detection information of distributed underwater detection nodes, integrates the main algorithms of information fusion, and provides the main interface and functions for underwater detection information fusion. The detection information collaborative processing module receives the detection information participating in the detection task, including position information, speed information, status information, etc. This information comes from the detection node models in the model library, and the detection node models output detection information according to the events and operation logics generated during the simulation deduction process. After the detection information collaborative processing module receives the detection information of each node, since the information expressed by each node is asymmetric and inconsistent, some detection nodes detect position information, some detect speed information, and some detect status information. Therefore, after obtaining the above information detected by the detection nodes, it is necessary to use the data alignment function to perform spatial alignment on the obtained position information, speed information, and status information, and filter the spatially aligned data through the Kalman filter algorithm to perform feature space alignment and achieve the unity of feature expression; then perform comprehensive processing of the detected information on the filtered data, including track preprocessing, track association, and track fusion, to obtain the target fusion track; next, construct an underwater target threat ranking model according to the target fusion track, calculate the underwater target threat level, and rank the underwater target threat levels to form an underwater target threat ranking result, and use this target threat ranking result as the final target detection result. According to the intelligence subscription or specified distribution strategy, distribute the underwater intelligence information and perform the comprehensive display of underwater intelligence for the information of each intelligence source, the final fusion track, and the underwater target threat ranking result.

[0098] The target detection result classification and recognition module is mainly used for identifying, processing, and classifying the target detection results generated by the detection information collaborative processing module. After receiving the information from the detection nodes, it calls the preprocessing function in the algorithm library to perform data preprocessing. Since the underwater target signal contains all the information of the underwater target radiation sound field, from the perspective of signal processing, it has time-domain characteristics, frequency-domain characteristics, and time-frequency domain joint characteristics, which reflect the essential characteristics of the underwater target sound field from three different perspectives. Next, it calls the function in the algorithm library according to the feature extraction operation instruction to perform feature extraction and judgment. If no feature extraction operation is required, the recognition result is directly output; otherwise, the feature extraction operation is performed. The distributed sonar array has the above time-frequency characteristics, frequency-domain characteristics, and time-frequency domain characteristics. For non-stationary signals, time-frequency characteristics are used for signal analysis. Among them, the time-domain characteristics can be directly analyzed through the signal sequence. Since the signal collected by the sonar is a mixed signal composed of useful signals and noise signals, the power spectrum analysis method is an extremely effective means to extract useful information in the noise background. It truly reflects the distribution of each frequency component and its energy size in the sound signal. Therefore, the power spectrum analysis method can be used to obtain the characteristics of the target in the frequency domain. In this embodiment, the power spectrum analysis method specifically uses the Fourier transform function to transform the signal from the time domain to the frequency domain. The time-frequency domain characteristics are analyzed and obtained by using the method based on the short-time Fourier transform. After extracting multi-dimensional characteristics, feature selection is performed. Using the mapping relationship between the features and the target, the selected multi-dimensional characteristics are selected to obtain effective features. Next, the Apriori algorithm is used to perform association analysis on the selected effective features to obtain association features. Finally, the target recognition algorithm is called in the algorithm library to perform target classification to obtain the target classification result. According to the intelligence subscription or specified distribution strategy, underwater recognition intelligence information is distributed, the target classification result is reported and displayed, and stored in the database.

[0099] After the simulation is completed, all the data involved in the simulation system is pushed to the data recording module, and the driving playback situation data is generated and sent to the two-dimensional and three-dimensional integrated situation display module for display.

[0100] An underwater target detection and recognition simulation system provided in this embodiment takes the detection information collaborative processing module and the target detection result classification and recognition module as the core, combines the simulation components in the high-resolution model library, completes the simulation of the process of underwater target collaborative detection and recognition, and realizes a simulation system for underwater target collaborative detection and recognition.

[0101] Embodiment 2

[0102] As Figure 3 shown, the present invention provides a detection method based on the above underwater target detection and recognition simulation system, which specifically includes the following steps:

[0103] S10: Obtain underwater detection information according to the detection node model, and perform fusion processing on the obtained underwater detection information to obtain a fusion trajectory.

[0104] S20: Construct an underwater target threat ranking model based on the fusion track, calculate the underwater target threat degree, and rank the underwater target threat degree to form a threat ranking result as the target detection result.

[0105] S30: Call the preprocessing function, extract the original signal from the target detection result generated by the detection information collaborative processing module, and perform spatial alignment preprocessing on the extracted original signal to obtain preprocessed data.

[0106] S40: Through the feature extraction function, perform feature extraction on the preprocessed data in the time domain, frequency domain, and time-frequency joint domain to obtain multi-dimensional features, and perform feature selection on the extracted multi-dimensional features according to the feature selection conditions in the simulation requirements to obtain effective features.

[0107] S50: Perform correlation analysis on the obtained effective features to obtain correlation features, and call the target recognition algorithm to classify the correlation features to obtain the target classification result.

[0108] Those skilled in the art can clearly understand that for the convenience and conciseness of description, only the above division of each functional unit and module is used as an example. In actual applications, the above functions can be allocated to different functional units and modules according to needs, that is, the internal structure of the device is divided into different functional units or modules to complete all or part of the functions described above.

[0109] The above specific implementation manners further elaborate the purpose, technical solutions, and beneficial effects of the present invention. It should be understood that the above are only specific implementation manners of the present invention and are not used to limit the protection scope of the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included in the protection scope of the present invention.

Claims

1. An underwater target detection and recognition simulation system, characterized in that, It includes: A model establishment and application module, which is used to call corresponding simulation components from a high-resolution model library according to simulation requirements to generate a target detection model; After obtaining the target detection model, input the data to be simulated in the simulation requirements into the target detection model for processing, and output detection information; A detection information collaborative processing module, which is used to perform fusion processing on the detection information output by the target detection model, generate a target detection result, and send it to the simulation detection platform; A target detection result classification and recognition module, which is used to perform recognition processing and classification on the target detection result generated by the detection information collaborative processing module to obtain target recognition information and send it to the simulation detection platform; A simulation detection platform, which is used to send the received detection information and target recognition information to the data recording module for recording; The detection information includes the position information, speed information, and status information of the detected target; The detection information collaborative processing module includes: A detection information spatial alignment unit, which is used to perform spatial alignment on the obtained position information, speed information, and status information by using a data alignment function, and filter the spatially aligned data through a Kalman filtering algorithm to obtain filtered data; A detection information comprehensive processing unit, which is used to perform track preprocessing, track association processing, and track fusion processing on the filtered data to obtain a target fusion track; An underwater target threat ranking unit, which is used to construct an underwater target threat ranking model according to the target fusion track, calculate the underwater target threat degree, and rank the underwater target threat degree to form a threat ranking result as the target detection result; The target detection result classification and recognition module includes: A data preprocessing unit, which is used to call a preprocessing function to extract the original signal from the target detection result generated by the detection information collaborative processing module, and perform spatial alignment preprocessing on the extracted original signal to obtain preprocessed data; A feature extraction unit, which is used to call a feature extraction function to extract features from the preprocessed data in the time domain, frequency domain, and time-frequency joint domain to obtain multi-dimensional features; A feature selection unit, which is used to perform feature selection on the extracted multi-dimensional features according to the feature selection conditions in the simulation requirements to obtain effective features; An effective feature processing unit, which is used to perform correlation analysis on the selected effective features to obtain correlation features, and call a target recognition algorithm to classify the correlation features to obtain a target classification result; The feature extraction unit includes: A time-domain feature extraction unit, which is used to perform mean square value, peak value, and variance processing on the time-domain data of the preprocessed data to obtain time-domain features; A frequency-domain feature extraction unit, which is used to extract features from the preprocessed data by using a Fourier transform function to obtain frequency-domain features; A time-frequency feature extraction unit, which is used to extract features from the preprocessed data by using a short-time Fourier transform function to obtain time-frequency features.

2. The underwater target detection and recognition simulation system according to claim 1, characterized in that, The model establishment and application module includes: A simulation component acquisition unit, which is used to call corresponding simulation components from a high-resolution model library according to the simulation components required in the simulation requirements; A simulation component assembly unit, which is used to assemble and associate the called simulation components according to the assembly relationship and association relationship in the simulation requirements to obtain a target detection model.

3. The underwater target detection and recognition simulation system according to claim 2, characterized in that, The high-resolution model library includes a detection node model, a detection target model, a battlefield interference model, a sensor array model, and a network model; Among them, the detection node model includes a sonar simulation component, a buoy and submersible buoy simulation component, an unmanned underwater vehicle simulation component, and a surface ship simulation component; The detection target model includes a ship simulation component, a frogman simulation component, a submarine simulation component, and an unmanned underwater vehicle simulation component; The battlefield interference model includes a Doppler effect model, a self-noise interference model, an underwater acoustic propagation model, a reverberation model, and an ocean ambient noise model; The sensor array model includes a sensor line array simulation component, a sensor volume array simulation component, and a sensor cylindrical array simulation component; The network model includes a time slot control simulation model, a network management simulation model, and a link transmission simulation model.

4. The underwater target detection and recognition simulation system according to claim 1, characterized in that, The underwater target detection and recognition simulation system further includes a comprehensive situation display module; The comprehensive situation display module is used to display the detection information, target detection results, and target recognition information of the target detection model.

5. The underwater target detection and recognition simulation system according to claim 1, characterized in that, The underwater target detection and recognition simulation system further includes a data storage module; The data storage module is used to store the data in the model establishment and application module, the detection information collaborative processing module, the target detection result classification and recognition module, the simulation detection platform, and the data recording module.

6. An underwater target detection and recognition simulation system according to any one of claims 1-5, characterized in that, The modules are scheduled and data is transmitted through an interface method.

7. A detection method based on the underwater target detection and recognition simulation system according to any one of claims 1-6, characterized in that, Including: Obtain underwater detection information according to the detection node model, and perform fusion processing on the obtained underwater detection information to obtain a fusion trajectory; Construct an underwater target threat ranking model according to the fusion track, calculate the underwater target threat degree, and rank the underwater target threat degree to form a threat ranking result as the target detection result; Call the preprocessing function, extract the original signal from the target detection result generated by the detection information collaborative processing module, and perform spatial alignment preprocessing on the extracted original signal to obtain preprocessed data; Through the feature extraction function, extract features from the preprocessed data in the time domain, frequency domain, and time-frequency joint domain to obtain multi-dimensional features, and perform feature selection on the extracted multi-dimensional features according to the feature selection conditions in the simulation requirements to obtain effective features; Perform correlation analysis on the obtained effective features to obtain correlation features, and call the target recognition algorithm to classify the correlation features to obtain the target classification result; The detection information includes the position information, speed information, and status information of the detection target; The detection information collaborative processing module includes: A detection information spatial alignment unit, which is used to perform spatial alignment on the obtained position information, speed information, and status information by using a data alignment function, and filter the spatially aligned data through a Kalman filter algorithm to obtain filtered data; A detection information comprehensive processing unit, which is used to perform track preprocessing, track association processing, and track fusion processing on the filtered data to obtain a target fusion track; An underwater target threat ranking unit, which is used to construct an underwater target threat ranking model according to the target fusion track, calculate the underwater target threat degree, and rank the underwater target threat degree to form a threat ranking result as the target detection result; Perform correlation analysis on the obtained effective features to obtain correlation features, and call the target recognition algorithm to classify the correlation features to obtain the target classification results, including: Call the preprocessing function to extract the original signal from the target detection results generated by the detection information collaborative processing module, and perform spatial alignment preprocessing on the extracted original signal to obtain preprocessed data; Call the feature extraction function to extract features from the preprocessed data in the time domain, frequency domain, and time-frequency joint domain to obtain multi-dimensional features; Perform feature selection on the extracted multi-dimensional features according to the feature selection conditions in the simulation requirements to obtain effective features; Perform correlation analysis on the selected effective features to obtain correlation features, and call the target recognition algorithm to classify the correlation features to obtain the target classification results; Feature extraction, including: Perform mean square value, peak value, and variance processing on the time-domain data of the preprocessed data to obtain time-domain features; Use the Fourier transform function to extract features from the preprocessed data to obtain frequency-domain features; Use the short-time Fourier transform function to extract features from the preprocessed data to obtain time-frequency features.

Images