An amphibious ship target detection system and method

Abstract

The present invention relates to the technical field of target detection, and particularly to an amphibious ship target detection system and method. The method includes the following steps: acquiring monitoring data of multi-modal sensors; performing phase alignment processing on the monitoring data of the multi-modal sensors to generate multi-modal sensor phase alignment data; constructing a three-dimensional monitoring field of the ship space according to the multi-modal sensor phase alignment data to obtain a three-dimensional monitoring field of the ship space; analyzing the ship motion trajectory of the three-dimensional monitoring field of the ship to generate ship motion trajectory analysis data; planning an avoidance path for the ship motion trajectory analysis data to generate ship avoidance path data; Therefore, the present invention solves the problems of low accuracy, slow response and poor environmental adaptability of the traditional ship target detection system by integrating multi-modal data processing and intelligent path planning technologies, and improves the target recognition ability and obstacle avoidance ability of the ship in multiple environments.

Description

Technical Field

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

[0002] Existing amphibious ship target detection systems mainly rely on traditional radars or single sensors for target detection, and are easily affected by noise interference in complex marine environments, resulting in a decrease in detection accuracy, especially performing poorly in bad weather or long-distance detection. In addition, the single-modal data processing method cannot effectively fuse multi-source information, fails to make full use of the complementary advantages of sensors, and affects the reliability of detection results. In terms of target recognition, traditional feature extraction and pattern matching methods are difficult to cope with dynamically changing ship behaviors and complex backgrounds. Especially when identifying new targets and dealing with ship avoidance behaviors, there is a lack of real-time update and adjustment mechanisms, resulting in inaccurate behavior prediction. Existing systems rely on manually set rules and thresholds, lack adaptability and robustness, cannot cope with environmental changes, and reduce the execution efficiency. In data processing and model training, traditional methods rely on manual annotation and static analysis, lack automated and intelligent support, increase the complexity of deployment and maintenance, and also limit the processing ability of large-scale data. Especially when real-time code generation and automatic tuning are required, traditional methods are cumbersome and inefficient and cannot meet the needs of modern complex tasks. Summary of the Invention

[0003] Based on this, it is necessary to provide an amphibious ship target detection system and method to solve at least one of the above technical problems.

[0004] To achieve the above object, an amphibious ship target detection method includes the following steps:

[0005] Step S1: Obtain multi-modal sensor monitoring data; perform monitoring signal phase alignment processing on the multi-modal sensor monitoring data to generate multi-modal sensor phase alignment data; construct a three-dimensional monitoring field of the ship space based on the multi-modal sensor phase alignment data to obtain a three-dimensional ship space monitoring field;

[0006] Step S2: Analyze the ship motion trajectory of the three-dimensional ship space monitoring field to generate ship motion trajectory analysis data; plan an avoidance path for the ship motion trajectory analysis data to generate ship avoidance path data;

[0007] Step S3: Obtain onshore area data; perform trajectory correlation and cross-analysis based on the ship motion trajectory analysis data and the ship avoidance path data to generate ship trajectory cross-analysis data; perform a confidence evaluation model on the onshore area data and the ship trajectory cross-analysis data to generate an amphibious ship target detection model;

[0008] Step S4: Conduct target detection and recognition learning on the amphibious ship target detection model to obtain object detection target data; embed the object detection target data into the radar system to perform the amphibious ship target detection task.

[0009] The beneficial effects of the present invention are as follows. By acquiring multi-modal sensor monitoring data and performing phase alignment processing on these data, the consistency of the monitoring signals of each sensor in time series is ensured, thereby reducing data deviation caused by signal delay or synchronization error between different sensors. Then, based on the phase-aligned data, a three-dimensional space monitoring field of the ship is constructed, and through the fusion of spatial data, a comprehensive and accurate ship monitoring model is formed, providing high-quality basic data for subsequent trajectory analysis and path planning. In the second step, by analyzing the motion trajectory of the three-dimensional space monitoring field of the ship, detailed ship motion trajectory data is generated. Combining the motion law of the ship and environmental changes, an avoidance path planning is implemented to ensure that the ship can autonomously avoid obstacles or other ships in different complex environments, thereby improving the safety and maneuverability of the ship. Next, by acquiring land area data and cross-analyzing it with the ship's motion trajectory data, ship trajectory cross-analysis data is formed, further optimizing the ship path planning, and a confidence evaluation model is constructed based on this to ensure the accuracy and reliability of the data in practical applications. Finally, an amphibious ship target detection model is generated. This model combines target recognition algorithms to learn and identify the detection data, thereby realizing real-time monitoring and recognition of targets on the water surface and on land. Finally, by embedding the object detection target data into the radar system, a complete detection system is formed to perform the amphibious ship target detection task. Through this series of technical means, the present invention improves the target detection ability and avoidance efficiency of the ship in complex environments. Therefore, the present invention solves the problems of low accuracy, slow response, and poor environmental adaptability of traditional ship target detection systems by integrating multi-modal data processing and intelligent path planning technologies, and improves the target recognition ability and obstacle avoidance ability of ships in multiple environments.

[0010] Preferably, step S1 includes the following steps:

[0011] Step S11: Acquire multi-modal sensor monitoring data;

[0012] Step S12: Perform monitoring signal alignment processing on the multi-modal sensor monitoring data to generate multi-modal sensor phase alignment data; perform monitoring object morphology recognition on the multi-modal sensor phase alignment data to obtain monitoring object morphology recognition data;

[0013] Step S13: Construct a three-dimensional space monitoring field of the ship according to the monitoring object morphology recognition data to obtain the ship three-dimensional space monitoring field.

[0014] The present invention obtains monitoring data from different types of sensors (such as radar, optical sensors, sonar, etc.). Since the data sources of different sensors have different time series, frequencies, and spatial resolutions, in order to ensure the consistency and comparability of the data, signal alignment processing is required. The signal alignment processing includes time series synchronization, frequency adjustment, and spatial position registration. Through these processing means, the relative time deviation and spatial error between sensors can be eliminated, thereby generating accurate multi-modal sensor phase-aligned data. The phase-aligned data set provides a reliable basis for subsequent object shape recognition, and then object shape recognition of the monitored object is carried out on this basis. This process utilizes advanced image processing and pattern recognition technologies to identify and classify target objects in the monitoring area by analyzing the shape characteristics (such as edges, corners, textures, etc.) of different objects in the sensor data, and obtain the shape recognition data of the monitored object. Through shape recognition, the system can extract feature data with key information from a large amount of data, so as to accurately judge the shape and position of the ship or other important objects. Finally, based on the shape recognition data. This process utilizes geometric modeling and data fusion technologies to integrate the recognition data of various sensors according to spatial coordinates, and through data fitting and optimization, an accurate three-dimensional monitoring field of the ship and its surrounding environment is formed. This three-dimensional monitoring field can provide key support for subsequent ship trajectory analysis, target detection, and path planning. Through this technical solution, the present invention significantly improves the accuracy and real-time performance of ship monitoring, and can achieve efficient and stable target monitoring and recognition in complex marine environments.

[0015] Preferably, step S13 includes the following steps:

[0016] Step S131: Perform geometric edge calculation on the shape recognition data of the monitored object to generate edge calculation data of the monitoring sensor;

[0017] Step S132: Analyze the geometric constraint relationship of the monitoring sensor based on the simultaneous localization and mapping technology to generate geometric constraint data of the monitoring sensor; perform spatio-temporal compensation on the common view area of the monitoring sensor for the geometric constraint data of the monitoring sensor to generate spatio-temporal calibration data of the monitoring sensor;

[0018] Step S133: Obtain the position of the coastal feature points; perform spatio-temporal region mapping on the spatio-temporal calibration data of the monitoring sensor and the position of the coastal feature points, and use particle swarm optimization to correct the sensor registration error to generate ship spatial sensor data;

[0019] Step S134: Obtain a BEV bird's-eye view; perform multi-modal distance fusion on the ship spatial sensor data using the BEV bird's-eye view, and construct a three-dimensional monitoring field of the ship space to obtain a three-dimensional monitoring field of the ship space.

[0020] The present invention performs geometric edge calculation on the morphological recognition data of the monitored object. This process analyzes the geometric features in the recognition data, extracts the edge information of the object, and then generates the edge calculation data of the monitoring sensor, providing a basis for subsequent spatial relationship analysis. Based on the Simultaneous Localization and Mapping (SLAM) technology, geometric constraint relationship analysis is performed on the edge calculation data of the monitoring sensor. The SLAM technology can synchronously calculate the spatial position of the sensor and the surrounding environmental features, optimize the position relationship of the sensor in space through geometric constraints, and thus generate the geometric constraint data of the monitoring sensor. To further improve the spatio-temporal accuracy of the data, spatio-temporal compensation of the co-visible area of these data is carried out, that is, spatio-temporal calibration is performed within the observation area of the sensor to ensure the consistency of different sensors in time and space, generating the spatio-temporal calibration data of the monitoring sensor. This calibration process compensates and corrects the spatio-temporal errors of the sensor through algorithm optimization, improving the accuracy of multi-sensor data fusion. The system obtains the position data of the coastal feature points, which have significant geographical identification functions in the marine environment. Then, through spatio-temporal region mapping technology, the spatio-temporally calibrated sensor data is aligned with the positions of the coastal feature points to ensure that the sensor data can be accurately mapped to the actual geographical area. On this basis, the Particle Swarm Optimization (PSO) algorithm is used to correct the sensor registration error. The particle swarm optimization can efficiently search for the optimal solution and correct the registration error caused by differences between sensors or environmental changes, and finally generate the ship space sensor data, providing accurate sensor positioning information for the construction of the three-dimensional monitoring field. The system obtains the Bird's Eye View (BEV) image, and through multi-modal distance fusion of the BEV image and the ship space sensor data, the spatial registration of the sensor data is optimized, and then the three-dimensional monitoring field of the ship space is constructed. The BEV image can provide rich ground view information, and the combination with the sensor data helps to further enhance the spatial perception ability of the ship and its surrounding environment. Through this series of technical means, the present invention can achieve accurate three-dimensional modeling of the ship and its surrounding environment, improving the real-time monitoring accuracy and stability of the ship in complex water areas and shoreline environments.

[0021] Preferably, step S2 includes the following steps:

[0022] Step S21: Perform spatial motion analysis of the target object based on the three-dimensional space monitoring field of the ship to obtain the object spatial movement data;

[0023] Step S22: Perform ship motion trajectory analysis on the object spatial movement data to generate ship motion trajectory analysis data;

[0024] Step S23: Calculate the motion fluid disturbance of the object space movement data and the ship motion trajectory analysis data to generate the ship collision probability disturbance data; based on the ship collision probability disturbance data, plan an avoidance path to generate the ship avoidance path data.

[0025] The present invention first performs target object space motion analysis according to the data of the ship three-dimensional space monitoring field. This process mainly dynamically analyzes the motion states of target objects (such as other ships, buoys, marine obstacles, etc.) in the monitoring field, extracts the spatial movement characteristics of the objects, including motion parameters such as speed, direction, and acceleration, and then generates the object space movement data. These data provide the basis for subsequent trajectory analysis and avoidance path planning. By further analyzing the object space movement data, calculate the ship motion trajectory. Using the spatial movement data of the target object and combining with the current motion state of the ship, a trajectory prediction model is used to predict the motion trajectories of the ship and the surrounding objects, generating the ship motion trajectory analysis data. This analysis data can show the relative position changes between the ship and the target object, providing data support for the next collision avoidance decision. Combine the object space movement data with the ship motion trajectory analysis data to perform motion fluid disturbance calculation. This calculation estimates the probability of collision of the ship in different motion states by analyzing the influence of external fluid environments such as water flow and waves on the ship during navigation, as well as the disturbances generated by the interaction between the ship and the surrounding objects. Using the fluid mechanics model and computational fluid dynamics (CFD) technology, the disturbance effect of the water flow on the ship navigation trajectory can be quantified, thereby generating the ship collision probability disturbance data. Based on these data, further plan the avoidance path. By establishing an avoidance path planning algorithm and combining the collision probability disturbance data, the motion path of the ship is intelligently optimized to generate the ship avoidance path data. This process takes into account the speed, direction, navigation environment of the ship and the dynamic changes of the obstacles, and can provide an optimal avoidance path for the ship to minimize the collision risk. Through this series of technical means, the present invention realizes the accurate trajectory analysis and intelligent collision avoidance decision of the ship in a complex dynamic environment, significantly improving the navigation safety and adaptability of the ship.

[0026] Preferably, step S22 includes the following steps:

[0027] Step S221: Perform non-linear Kalman filtering on the object space movement data to construct a motion state matrix, generating a ship motion state prediction matrix;

[0028] Step S222: Perform frequency domain energy spectrum analysis on the ship motion state prediction matrix and extract the main frequencies of roll and surge to generate a ship frequency domain energy distribution map;

[0029] Step S223: Perform ship motion stability boundary analysis on the ship motion state prediction matrix through the Lyapunov exponent stability criterion to generate the safety threshold range of acceleration increment; perform dynamic truncation correction processing on the ship frequency-domain energy distribution map according to the safety threshold range of acceleration increment to generate ship motion trajectory analysis data.

[0030] The present invention constructs a motion state matrix by performing nonlinear Kalman filtering on the object space movement data. Kalman filtering is a dynamic signal processing method that iteratively updates the estimated value, which can effectively eliminate noise interference and accurately estimate the motion state of the ship. Nonlinear Kalman filtering further expands this technology and can handle complex and nonlinear dynamic systems. In this process, the Kalman filter adjusts the predicted value in real time according to the sensor data and gradually generates the ship motion state prediction matrix, which contains the motion characteristics of the ship such as speed, position, and acceleration at each moment and is used for subsequent stability analysis and trajectory prediction. The ship motion state prediction matrix undergoes frequency-domain energy spectrum analysis to extract the main frequencies of rolling and surge, generating the ship frequency-domain energy distribution map. Frequency-domain energy spectrum analysis is a technology that processes signals by converting them to the frequency domain, which can reveal the periodic and fluctuating characteristics existing in ship motion. Rolling and surge are the main motion modes of the ship in the transverse and longitudinal directions, which play important roles during the ship's navigation. By extracting these main frequencies, the dynamic response characteristics of the ship under different navigation conditions can be identified, and the corresponding frequency-domain energy distribution map can be generated, providing a basis for further stability analysis and path correction. Apply the Lyapunov exponent stability criterion to analyze the ship motion state prediction matrix to generate the safety threshold range of acceleration increment. The Lyapunov exponent is an index for measuring the stability of a dynamic system, which can judge whether the system is stable by analyzing the change of the system's behavior over time. Based on the ship motion state prediction matrix, use the Lyapunov exponent criterion to quantitatively analyze the motion stability of the ship, thereby determining the safety boundary of the ship under different motion states and generating the safety threshold range of acceleration increment. This process provides a quantitative standard for the safety control of the ship. Finally, according to this safety threshold range, perform dynamic truncation correction processing on the ship frequency-domain energy distribution map to generate ship motion trajectory analysis data. This correction process can filter out the energy components beyond the safety range to ensure that the ship's motion trajectory meets the requirements of safety and stability. Through this series of technical means, the present invention can achieve accurate prediction, dynamic correction, and stability analysis of the ship motion state, thereby effectively improving the safety and stability of the ship in a complex navigation environment.

[0031] Preferably, performing ship motion stability boundary analysis on the ship motion state prediction matrix through the Lyapunov exponent stability criterion includes the following steps:

[0032] Extract the maximum Lyapunov exponent from the ship motion state prediction matrix to generate a characterization of the sensitivity of the ship's initial conditions;

[0033] Calculate the gradient field through the characterization of the ship's initial condition sensitivity to generate the flow field velocity gradient tensor;

[0034] Construct an implicit function relationship of the ship's critical instability acceleration based on the flow field velocity gradient tensor to generate the ship's critical instability implicit function;

[0035] Conduct a motion stability boundary analysis based on the ship's critical instability implicit function to generate the safety threshold range of acceleration increment.

[0036] When the present invention extracts the maximum Lyapunov exponent from the ship motion state prediction matrix, the Lyapunov exponent is used to quantitatively measure the sensitivity of the ship motion system. The Lyapunov exponent is a method for describing the stability of a dynamic system and can measure the sensitivity of the ship motion system to changes in initial conditions. By extracting the maximum Lyapunov exponent, the stability of the ship motion under initial conditions can be quantified, which provides a basis for subsequent stability analysis and path correction and generates the characterization data of the ship's initial condition sensitivity. Then, based on the characterization of the ship's initial condition sensitivity, a gradient field calculation is performed to generate the flow field velocity gradient tensor. In this process, the calculation method of the gradient field is used to evaluate the velocity change characteristics of the fluid environment around the ship, especially the influence of the velocity gradient of the fluid on the ship motion state during the ship's motion. The flow field velocity gradient tensor can describe the changes in the flow field in space, reflecting the interaction between the ship and the surrounding fluid, and thus providing data support for the construction of the implicit function relationship of the critical instability acceleration. Through the flow field velocity gradient tensor, an implicit function relationship of the ship's critical instability acceleration is constructed. During the ship's motion, affected by fluid disturbances, when the acceleration exceeds a certain threshold, the ship will enter the critical instability state, and this process can be described by establishing an implicit function. The implicit function can reflect the influence of factors such as the fluid environment, ship structure, and motion state on the instability acceleration through a series of complex nonlinear relationships and provide a mathematical basis for the safety boundary of the ship motion. Based on the ship's critical instability implicit function, a motion stability boundary analysis is conducted to generate the safety threshold range of acceleration increment. By analyzing the implicit function, the critical stability boundary of the ship under different motion states is determined, that is, the maximum acceleration range within which the ship can safely navigate under specific fluid environments and motion states. This safety threshold range of acceleration increment provides a clear standard for the dynamic control of the ship and can effectively avoid instability or accidents caused by excessive acceleration during the ship's navigation.

[0037] Preferably, step S3 includes the following steps:

[0038] Step S31: Obtain onshore area data;

[0039] Step S32: Conduct trajectory correlation and intersection analysis based on ship motion trajectory analysis data and ship avoidance path data to generate ship trajectory intersection analysis data;

[0040] Step S33: Perform early data-level fusion on the onshore area data to generate input data for the amphibious ship model; conduct model training based on the input data for the amphibious ship model, and perform calibration of model confidence parameters to generate a confidence amphibious detection model;

[0041] Step S34: Perform late feature fusion based on the ship trajectory intersection analysis data to generate amphibious ship feature fusion data; use the amphibious ship feature fusion data to perform trajectory matching and weighted optimization on the confidence amphibious detection model to generate an amphibious ship target detection model.

[0042] The present invention obtains onshore area data, which includes terrain information, geographical location, coastline characteristics, etc. This data provides the necessary environmental background information for the trajectory analysis and collision avoidance path planning of ships. Based on the ship movement trajectory analysis data and the ship avoidance path data, a trajectory correlation and cross-analysis is carried out. This analysis comprehensively considers the spatial and temporal relationships between the ship and other objects (such as land obstacles, other ships, etc.) to generate ship trajectory cross-analysis data. This data can reveal the potential risks of the ship crossing with the onshore area or other objects during navigation, providing an important basis for ship target detection. Based on the onshore area data, early data-level fusion is carried out to generate input data for the amphibious ship model. This process uses multi-source data fusion technology to comprehensively process the geographical data of the onshore area, ship movement trajectories, environmental changes, etc., and extracts feature data with high correlation and high reliability as input to ensure the data quality and reliability during model training. Based on these data, model training is carried out and confidence parameters are calibrated to generate a confidence amphibious detection model. By learning a large amount of historical data and environmental characteristics, this model can identify and detect amphibious ship targets, and further improve the accuracy and robustness of the detection results through the calibrated confidence parameters. Based on the ship trajectory cross-analysis data, late feature fusion is carried out to generate amphibious ship feature fusion data. Late feature fusion further processes and optimizes the analysis data obtained in the early stage, fusing multi-dimensional spatio-temporal information, motion characteristics, environmental factors, etc. to form a comprehensive feature data set. These data sets can more accurately reflect the motion patterns and potential dangers of ships in complex environments. Then, these amphibious ship feature fusion data are used to perform trajectory matching and weighted optimization on the confidence amphibious detection model. This optimization process enhances the model's response to key features through a weighting method, enabling the model to prioritize important risk factors when facing complex and changing environments, thereby generating the final amphibious ship target detection model. This model can not only effectively identify targets in a static environment but also intelligently predict and adjust the ship's movement path in a dynamic environment, improving the accuracy and response speed of target detection.

[0043] Preferably, step S32 includes the following steps:

[0044] Step S321: Perform ship avoidance marking on the ship avoidance path data to generate ship avoidance marking data; perform two-dimensional feature extraction of deceleration-avoidance time characteristics on the ship avoidance marking data to generate ship avoidance marking feature data;

[0045] Step S322: Perform time window matching on the ship avoidance mark feature data and the ship motion trajectory analysis data to generate ship time window avoidance-path matching data; perform Euclidean distance trajectory similarity metric matching on the ship time window avoidance-path matching data to generate ship motion trajectory-avoidance path matching data;

[0046] Step S323: Perform ship behavior pattern recognition on the ship motion trajectory-avoidance path matching data to generate ship motion behavior pattern feature data; perform avoidance type trajectory association and cross-analysis based on the ship motion behavior pattern feature data to generate ship trajectory cross-analysis data.

[0047] The present invention generates ship avoidance marking data by performing ship avoidance marking on ship avoidance path data. This step aims to identify the moments and positions where ships need to avoid during navigation. By marking the path data, key nodes are provided for subsequent analysis. Then, two-dimensional feature extraction of deceleration-avoidance time characteristics is performed on the ship avoidance marking data to generate ship avoidance marking feature data. This feature extraction process focuses on capturing the deceleration and the time information required for avoidance during the ship avoidance process. These information are of great significance for predicting the temporal and spatial distribution of ship avoidance behavior. The extraction of deceleration and avoidance time characteristics provides a reliable data basis for further trajectory matching and path optimization by quantifying the process of ship deceleration and the duration of avoidance actions. The ship avoidance marking feature data is matched with the ship motion trajectory analysis data in a time window to generate ship time window avoidance-path matching data. This step synchronously matches the ship avoidance actions with the motion trajectory data by setting a time window, so as to more accurately dock the ship's motion and avoidance behavior. Through the matching of the time window, the ship's avoidance trajectory can be more clearly defined in terms of time and space, ensuring the consistency between the avoidance path and the motion trajectory. Subsequently, Euclidean distance trajectory similarity measurement matching is performed on the ship time window avoidance-path matching data to generate ship motion trajectory-avoidance path matching data. The Euclidean distance measurement method is used to evaluate the trajectory similarity of the ship at different time periods, so as to determine whether the ship motion trajectory matches the avoidance path, and optimize the path selection through the measurement of similarity to ensure that the ship can complete the avoidance safely and in a timely manner. The ship motion trajectory-avoidance path matching data undergoes ship behavior pattern recognition to generate ship motion behavior pattern feature data. This step analyzes the ship's motion behavior in different situations through pattern recognition algorithms, identifies the typical behavior patterns of the ship in different avoidance scenarios, and generates corresponding feature data to provide a basis for subsequent avoidance decisions. By recognizing the ship's motion behavior patterns, the motion laws of the ship in specific avoidance situations can be revealed, providing guidance for the dynamic adjustment and optimization of the avoidance path. Then, based on the ship motion behavior pattern feature data, cross-analysis of avoidance type trajectory association is performed to generate ship trajectory cross-analysis data. By cross-analyzing different types of avoidance behaviors, the internal relationship between ship motion and avoidance path can be further explored, generating ship trajectory cross-analysis data, thereby optimizing the selection and adjustment of the avoidance path and enhancing the intelligence and real-time performance of ship avoidance decisions.

[0048] Preferably, step S4 includes the following steps:

[0049] Step S41: Based on the target detection and recognition learning of the amphibious ship target detection model, object detection target data is obtained;

[0050] Step S42: Automatically generate code based on the object detection target data to generate an amphibious ship target detection module;

[0051] Step S43: Embed the amphibious ship target detection module into the radar system to perform the amphibious ship target detection task.

[0052] In the present invention, object detection target data is obtained by performing target detection and recognition learning based on the amphibious ship target detection model. This step uses deep learning technology to train a large amount of sensor data (including radar, infrared, vision, etc.). Through feature extraction and pattern recognition, different target objects are automatically identified and classified. Through an efficient target recognition algorithm, the targets of the ship in a complex environment can be accurately classified and located, thereby generating object detection target data. These data are crucial for further target verification and task execution. Automatically generate code based on the object detection target data to generate an amphibious ship target detection module. Through the automatic code generation technology, the target detection results are converted into code modules that can actually run. The automatic code generation technology parses the target detection data and automatically creates corresponding functional modules. These modules can be docked into a larger system for data interaction and task execution. This process not only improves the development efficiency but also ensures the high consistency and accuracy of the target detection module, avoids potential errors or inconsistencies in manual code writing, and enhances the stability and maintainability of the system. The generated amphibious ship target detection module is embedded into the radar system to perform the amphibious ship target detection task. By embedding the already generated target detection module into the existing radar system, the system can perform the target detection task in real time. The radar system receives and processes data from different sensors, and uses the automatic recognition and processing capabilities of the detection module to classify, locate, and track potential targets. This process greatly improves the accuracy and efficiency of target detection and can respond to various target changes in real time in a dynamic environment to ensure the safe navigation of the ship.

[0053] In this specification, an amphibious ship target detection system is provided for performing the above-mentioned amphibious ship target detection method. The amphibious ship target detection system includes:

[0054] A data acquisition and space construction module, configured to obtain multi-modal sensor monitoring data; perform monitoring signal phase alignment processing on the multi-modal sensor monitoring data to generate multi-modal sensor phase alignment data; construct a three-dimensional monitoring field of the ship space according to the multi-modal sensor phase alignment data to obtain a three-dimensional monitoring field of the ship space;

[0055] The motion trajectory analysis and path planning module is used to analyze the ship motion trajectory in the ship three-dimensional space monitoring field, generate ship motion trajectory analysis data; perform avoidance path planning on the ship motion trajectory analysis data to generate ship avoidance path data;

[0056] The trajectory correlation and target detection model construction module is used to obtain onshore area data; perform trajectory correlation and cross-analysis based on the ship motion trajectory analysis data and the ship avoidance path data to generate ship trajectory cross-analysis data; perform a confidence evaluation model on the onshore area data and the ship trajectory cross-analysis data to generate an amphibious ship target detection model;

[0057] The target detection and execution module is used to perform target detection and recognition learning on the amphibious ship target detection model to obtain object detection target data; embed the object detection target data into the radar system to execute the amphibious ship target detection task.

[0058] The beneficial effects of the present invention are as follows. By obtaining multi-modal sensor monitoring data and performing phase alignment processing on these data, the consistency of the monitoring signals of each sensor in time series is ensured, thereby reducing the data deviation caused by signal delay or synchronization error between different sensors. Then, based on the phase-aligned data, a three-dimensional space monitoring field of the ship is constructed, and through the fusion of spatial data, a comprehensive and accurate ship monitoring model is formed, providing high-quality basic data for subsequent trajectory analysis and path planning. In the second step, by analyzing the motion trajectory in the ship three-dimensional space monitoring field, detailed ship motion trajectory data is generated, and combined with the motion law of the ship and environmental changes, avoidance path planning is implemented to ensure that the ship can autonomously avoid obstacles or other ships in different complex environments, thereby improving the safety and maneuverability of the ship. Next, by obtaining onshore area data and performing cross-analysis with the ship's motion trajectory data, ship trajectory cross-analysis data is formed, further optimizing the ship path planning, and based on this, a confidence evaluation model is constructed to ensure the accuracy and reliability of the data in practical applications, and finally an amphibious ship target detection model is generated. This model combines target recognition algorithms to learn and identify the detection data, thereby realizing real-time monitoring and recognition of targets on the water surface and on land. Finally, by embedding the object detection target data into the radar system, a complete detection system is formed to execute the amphibious ship target detection task. Through this series of technical means, the present invention improves the target detection ability and avoidance efficiency of the ship in complex environments. Therefore, the present invention solves the problems of low accuracy, slow response and poor environmental adaptability of the traditional ship target detection system by integrating multi-modal data processing and intelligent path planning technologies, and improves the target recognition ability and obstacle avoidance ability of the ship in multiple environments. Brief Description of the Drawings

[0059] Figure 1 Schematic diagram of the step flow of an amphibious ship target detection method;

[0060] Figure 2 Is Figure 1 Schematic diagram of the detailed implementation steps of step S2 in

[0061] Figure 3 Is Figure 1 Schematic diagram of the detailed implementation steps of step S3 in

[0062] Figure 4 Is Figure 1 Schematic diagram of the detailed implementation steps of step S4 in

[0063] The realization, functional features and advantages of the object of the present invention will be further described with reference to the embodiments and the accompanying drawings. Detailed implementation manners

[0064] The technical method of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of them. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0065] In addition, the accompanying drawings are only schematic diagrams of the present invention and are not necessarily drawn to scale. The same reference numerals in the drawings represent the same or similar parts, and thus their repeated description will be omitted. Some of the block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor methods and / or microcontroller methods.

[0066] It should be understood that although the terms "first", "second", etc. may be used here to describe various units, these units should not be limited by these terms. These terms are only used to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, the first unit may be called the second unit, and similarly the second unit may be called the first unit. The term "and / or" used here includes any and all combinations of one or more of the listed associated items.

[0067] To achieve the above object, please refer to Figures 1 to 4 , an amphibious ship target detection method, the method comprising the following steps:

[0068] Step S1: Obtain multi-modal sensor monitoring data; perform monitoring signal phase alignment processing on the multi-modal sensor monitoring data to generate multi-modal sensor phase alignment data; construct a three-dimensional monitoring field of the ship's space based on the multi-modal sensor phase alignment data to obtain a three-dimensional monitoring field of the ship's space;

[0069] Step S2: Analyze the ship's motion trajectory in the three-dimensional monitoring field of the ship to generate ship motion trajectory analysis data; plan an avoidance path for the ship motion trajectory analysis data to generate ship avoidance path data;

[0070] Step S3: Obtain onshore area data; perform trajectory correlation and intersection analysis based on the ship motion trajectory analysis data and the ship avoidance path data to generate ship trajectory intersection analysis data; perform a confidence evaluation model on the onshore area data and the ship trajectory intersection analysis data to generate an amphibious ship target detection model;

[0071] Step S4: Perform target detection and recognition learning on the amphibious ship target detection model to obtain object detection target data; embed the object detection target data into the radar system to perform the amphibious ship target detection task.

[0072] The beneficial effects of the present invention are as follows. By obtaining the monitoring data of multi-modal sensors and performing phase alignment processing on these data, the consistency of the monitoring signals of each sensor in time series is ensured, thereby reducing the data deviation caused by signal delay or synchronization error between different sensors. Then, based on the phase-aligned data, a three-dimensional space monitoring field of the ship is constructed. Through the fusion of spatial data, a comprehensive and accurate ship monitoring model is formed, providing high-quality basic data for subsequent trajectory analysis and path planning. By analyzing the movement trajectory of the ship's three-dimensional space monitoring field, detailed ship movement trajectory data is generated. Combining the movement law of the ship and environmental changes, an avoidance path is planned to ensure that the ship can autonomously avoid obstacles or other ships in different complex environments, thereby improving the safety and maneuverability of the ship. Next, by obtaining land area data and cross-analyzing it with the ship's movement trajectory data, ship trajectory cross-analysis data is formed, further optimizing the ship path planning. Based on this, a confidence evaluation model is constructed to ensure the accuracy and reliability of the data in practical applications, and finally an amphibious ship target detection model is generated. This model combines target recognition algorithms to learn and identify the detection data, thereby realizing real-time monitoring and identification of targets on the water surface and on land. Finally, by embedding the object detection target data into the radar system, a complete detection system is formed to perform the amphibious ship target detection task. Through this series of technical means, the present invention improves the target detection ability and avoidance efficiency of the ship in complex environments. Therefore, the present invention solves the problems of low accuracy, slow response, and poor environmental adaptability of traditional ship target detection systems by integrating multi-modal data processing and intelligent path planning technologies, and improves the target recognition ability and obstacle avoidance ability of ships in multiple environments.

[0073] In the embodiment of the present invention, refer to Figure 1 As shown, it is a schematic diagram of the step flow of an amphibious ship target detection method of the present invention. In this example, the amphibious ship target detection method includes the following steps:

[0074] Step S1: Obtain the monitoring data of multi-modal sensors; perform phase alignment processing on the monitoring signals of the multi-modal sensors to generate phase-aligned data of the multi-modal sensors; construct a three-dimensional ship space monitoring field based on the phase-aligned data of the multi-modal sensors to obtain a three-dimensional ship space monitoring field;

[0075] In the embodiments of the present invention, multi-modal sensor monitoring data is obtained. These data come from different types of sensors, such as radar, infrared, sonar, visual sensors, etc. These sensors can provide multi-dimensional information about the ship and its surrounding environment, including position, speed, shape, and other related features. By comprehensively utilizing this multi-modal data, more comprehensive and accurate ship information can be obtained, laying a foundation for subsequent analysis and modeling. After the data is obtained, for the data from different sensors, monitoring signal phase alignment processing is performed to generate multi-modal sensor phase alignment data. Phase alignment processing refers to aligning the data collected by different sensors in terms of time and space, so that the data from different sensors can be effectively compared and analyzed in the same coordinate system and time frame. This processing can eliminate the time delay and spatial errors existing between sensors, synchronize the timing information of multiple data sources, and thus ensure the accuracy and consistency of subsequent analysis results. Next, based on the multi-modal sensor phase alignment data, a three-dimensional monitoring field of the ship's space is constructed. In this process, by combining the spatial distribution and temporal changes of the sensor data, three-dimensional modeling algorithms (such as multi-view geometric modeling, spatial interpolation, etc.) are used to convert the multi-modal sensor data into a three-dimensional space model of the ship's environment. This three-dimensional space monitoring field can accurately reflect the relative position, movement trajectory of the ship, and its interaction with the surrounding environment, providing detailed spatial structure information for subsequent monitoring and prediction tasks.

[0076] Step S2: Analyze the ship's movement trajectory in the three-dimensional monitoring field of the ship to generate ship movement trajectory analysis data; plan an avoidance path for the ship movement trajectory analysis data to generate ship avoidance path data;

[0077] In the embodiments of the present invention, through trajectory analysis technology, the movement trajectory of a ship can be extracted and modeled. The data-level technologies used include trajectory prediction algorithms based on time series, such as Kalman filtering, particle filtering, or Bayesian filtering. These methods can effectively handle noise and uncertainty, thus accurately predicting the future movement trajectory of the ship. By processing the spatial monitoring data, the movement pattern of the ship within a given time range can be extracted, and then the dynamic information required for subsequent avoidance path planning can be provided. After generating the ship movement trajectory analysis data, the next step is to perform avoidance path planning based on these data. The core task of avoidance path planning is to calculate an optimal avoidance path according to the movement state of the ship and the situation of the surrounding environment. This process relies on path planning algorithms, such as those based on the Dijkstra algorithm or algorithms based on optimization theory. By extracting the spatial relationship between the ship and obstacles from the ship movement trajectory analysis data, and according to the avoidance requirements and the dynamic characteristics of the ship, an avoidance path is generated. To further optimize the path planning result, different constraint conditions, such as the minimum turning radius of the ship, speed limit, avoidance time window, etc., are considered in this process, so as to ensure the executability and safety of the path.

[0078] Step S3: Obtain onshore area data; perform trajectory correlation and cross-analysis based on the ship movement trajectory analysis data and the ship avoidance path data to generate ship trajectory cross-analysis data; perform a confidence evaluation model on the onshore area data and the ship trajectory cross-analysis data to generate an amphibious ship target detection model;

[0079] In the embodiments of the present invention, by acquiring onshore area data, rich environmental backgrounds can be provided for subsequent ship trajectory analysis and avoidance path planning. Next, based on the ship movement trajectory analysis data and ship avoidance path data, trajectory correlation and cross-analysis are carried out. This analysis mainly relies on multi-variable data fusion and correlation techniques. By comparing the ship movement trajectory data with the avoidance path data and combining with onshore area data, the spatial interaction between the ship and the surrounding environment is analyzed. To achieve the cross-analysis of the trajectory and the avoidance path, spatio-temporal correlation algorithms and data mining techniques are adopted, and the dynamic movement data of the ship and the static onshore data are fused to reveal potential collision areas, avoidance opportunities, and cross paths. Through this step, ship trajectory cross-analysis data can be generated, providing data support for the accuracy of subsequent target detection tasks. For the ship trajectory cross-analysis data and onshore area data, a confidence evaluation model is constructed. The confidence evaluation model uses machine learning methods, such as support vector machine (SVM), decision tree, or random forest, etc., for data training and evaluation. During this process, the model will generate credibility indicators for ship target detection according to the input data (such as trajectory cross information, environmental characteristics, etc.). These indicators can help the system evaluate the reliability and accuracy of the detection results, thus providing effective data support for ship target detection. Finally, based on the above analysis results, an amphibious ship target detection model is generated. This model will combine the trajectory analysis data, avoidance path data, and confidence evaluation results, and further improve the ship target detection accuracy through data fusion and pattern recognition techniques. Through this model, the system can effectively identify potential targets in a complex environment and complete the precise detection task of amphibious ship targets.

[0080] Step S4: Conduct target detection and recognition learning on the amphibious ship target detection model to obtain object detection target data; embed the object detection target data into the radar system, so as to perform the amphibious ship target detection task.

[0081] In the embodiments of the present invention, target detection and recognition learning is performed on the target detection model of the amphibious ship. This process mainly involves training the target detection model through machine learning algorithms such as convolutional neural network (CNN), deep learning algorithms, decision tree, etc. During the training process, the model gradually adjusts its internal parameters through optimization algorithms according to the pre-collected object detection target data (such as ship image data, radar echo signals, or sensor data, etc.), so as to improve the accuracy and efficiency of target recognition. Through this process, the model can identify different target features from multi-modal data, and can handle factors such as complex background noise and occlusions, improving the robustness and accuracy of detection. The key technologies of target detection and recognition learning lie in the use of a large amount of labeled data and the introduction of an adaptive learning mechanism during the training process to ensure that the model can adjust its recognition strategy according to the actual situation and gradually improve the accuracy of target recognition. After the learning is completed, the model can generate object detection target data, which includes not only the position and morphological features of the target, but also the dynamic change information of the target, such as speed, direction, etc. Next, these object detection target data will be embedded into the radar system for data fusion and processing. The core technology of embedding the radar system is the design of data interfaces and communication protocols to ensure that the target detection data can be seamlessly docked with other components of the radar system and provide accurate and real-time target information. By receiving these embedded data, the radar system can update the execution status of the ship target detection task in real time according to the spatial position, speed, and motion trend of the object. During this process, the radar system will also perform optimization and adjustment based on the detection target data, and further improve the execution effect of the amphibious ship target detection task by continuously updating the object detection data. The real-time detection performance of the radar system depends on the precise target data processing ability and efficient data fusion technology to ensure the real-time detection and dynamic tracking of ship targets in complex environments.

[0082] Preferably, step S1 includes the following steps:

[0083] Step S11: Obtain multi-modal sensor monitoring data;

[0084] Step S12: Perform monitoring signal alignment processing on the multi-modal sensor monitoring data to generate multi-modal sensor phase-aligned data; perform monitoring object morphology recognition on the multi-modal sensor phase-aligned data to obtain monitoring object morphology recognition data;

[0085] Step S13: Construct a three-dimensional monitoring field of the ship space based on the monitoring object morphology recognition data to obtain a three-dimensional monitoring field of the ship space.

[0086] In the embodiments of the present invention, monitoring data from multi-modal sensors is acquired. The multi-modal sensors include lidar, infrared imaging, radar, vision sensors, etc., which can collect rich data about target objects from different perception dimensions (such as space, time, spectrum, etc.). These data include multi-dimensional information such as the spatial position, shape, material, temperature, speed of the object, etc. The multi-modal data fusion technology improves the reliability and diversity of information by integrating the raw data from different sensors. In practical applications, data synchronization and denoising processing are very important steps. First, phase alignment processing of the monitoring signals of these multi-modal sensor monitoring data is performed, aiming to solve the data inconsistency problem caused by the synchronization errors of different sensors in time and space. The phase alignment processing can ensure that the monitoring signals from different sensors are aligned within the same time window, thus ensuring the consistency and accuracy of the data in subsequent processing. This process uses interpolation algorithms, timestamp matching, or spatio-temporal alignment based on the Simultaneous Localization and Mapping (SLAM) technology. After the phase alignment is completed, the generated multi-modal sensor phase alignment data provides a highly consistent and accurate data basis for subsequent object recognition and monitoring field construction. Based on the aligned data, the morphological recognition of the monitored object continues. Through image recognition algorithms and target detection methods (such as Convolutional Neural Network (CNN), deep learning, etc.), the data acquired by the sensors can be analyzed to identify the morphological features of the object. These morphological recognition data not only include the geometric shape of the object, but also other physical features such as material type, surface reflectivity, etc., which are crucial for subsequent monitoring and recognition. According to the morphological recognition data of the monitored object, the three-dimensional space monitoring field of the ship is further constructed. The construction of the three-dimensional space monitoring field relies on multi-sensor fusion technology and geometric modeling methods. Using the morphological recognition data and combining with the spatial positioning data of the sensors, a three-dimensional space monitoring environment around the ship is constructed through three-dimensional reconstruction technologies (such as stereo vision, point cloud data fusion, spatial transformation algorithms, etc.).

[0087] Preferably, step S13 includes the following steps:

[0088] Step S131: Perform geometric edge calculation on the morphological recognition data of the monitored object to generate monitoring sensor edge calculation data;

[0089] Step S132: Analyze the geometric constraint relationship of the monitoring sensor edge calculation data based on the Simultaneous Localization and Mapping technology to generate monitoring sensor geometric constraint data; perform spatio-temporal compensation on the monitoring sensor geometric constraint data in the co-visible area of the sensors to generate monitoring sensor spatio-temporal calibration data;

[0090] Step S133: Obtain the positions of coastal feature points; perform spatio-temporal region mapping on the spatio-temporal calibration data of the monitoring sensors and the positions of the coastal feature points, and use particle swarm optimization to correct the sensor registration error to generate ship spatial sensor data;

[0091] Step S134: Obtain the BEV bird's-eye view; perform multi-modal distance fusion on the ship spatial sensor data using the BEV bird's-eye view, and construct a ship spatial three-dimensional monitoring field to obtain a ship three-dimensional space monitoring field.

[0092] In the embodiments of the present invention, by extracting the geometric edge features of the target object, the edge computing data of the monitoring sensor is generated. This process relies on image processing algorithms, such as edge detection algorithms (e.g., Canny edge detection or Sobel operator), to accurately extract the geometric features of the object by calculating the spatial boundary information in the sensor data. The edge computing data provides important geometric information in object recognition and subsequent analysis, which helps to construct an accurate monitoring environment model. Based on the Simultaneous Localization and Mapping (SLAM) technology, the geometric constraint relationship analysis of the monitoring sensor edge computing data is carried out. The SLAM technology combines sensor data and spatial positioning information to analyze the geometric constraint relationship between different sensors, and then generates the geometric constraint data of the monitoring sensor. This process uses optimization algorithms (such as Kalman filtering or maximum likelihood estimation) to identify the relative positions and postures between sensors, so as to accurately describe the geometric constraints between sensors. Next, the spatio-temporal compensation of the co-visible area of the monitoring sensor geometric constraint data is carried out. Through spatio-temporal compensation, it is ensured that the data of different sensors can be calibrated consistently in time and space, and accurate spatio-temporal calibration data of the monitoring sensor is generated. Obtaining the positions of coastal feature points is the key spatial reference information, and satellite positioning systems (such as GPS) or Geographic Information System (GIS) technology are used to determine the positions of coastal line feature points. Then, the spatio-temporal calibration data of the monitoring sensor is mapped to the spatio-temporal region of the coastal feature point positions. This process aligns the sensor data with the geographical coordinate system to ensure the accuracy of the spatial data. In order to further correct the sensor registration error, the Particle Swarm Optimization (PSO) algorithm is used for optimization processing. The particle swarm optimization can effectively reduce the registration error caused by sensor errors by simulating the group intelligence search for the optimal solution, and finally generates the ship space sensor data. By obtaining the Bird's Eye View (BEV), the multi-modal distance fusion of the ship space sensor data is further carried out. The BEV image provides a global perspective from high altitude, which can combine sensor data from different sources (such as lidar, vision, infrared, etc.), and enhance the accuracy and reliability of the data through multi-modal data fusion methods (such as weighted average, Kalman filtering, etc.). Finally, the fused data is applied to the construction of the ship space three-dimensional monitoring field, and an accurate three-dimensional space model of the ship is generated using three-dimensional reconstruction algorithms (such as stereo vision, point cloud processing, etc.).

[0093] As an example of the present invention, refer to Figure 2 shown, in this example, step S2 includes:

[0094] Step S21: Perform spatial motion analysis of the target object according to the ship three-dimensional space monitoring field to obtain object spatial movement data;

[0095] Step S22: Analyze the ship's motion trajectory based on the object's spatial movement data to generate ship motion trajectory analysis data;

[0096] Step S23: Calculate the motion fluid disturbance of the object's spatial movement data and the ship motion trajectory analysis data to generate ship collision probability disturbance data; Plan an avoidance path based on the ship collision probability disturbance data to generate ship avoidance path data.

[0097] In the embodiment of the present invention, by obtaining the position, velocity, and acceleration data of the ship in space, combining the dynamic characteristics of the object and external environmental factors, and using a dynamic model or spatio-temporal interpolation technology to calculate the object's spatial movement data. These calculations rely on state estimation methods such as Kalman filtering or particle filtering to track the motion trajectory of the target object in real time. Through these technologies, the motion trajectory, velocity change, and acceleration characteristics of the object in three-dimensional space can be accurately extracted, thereby obtaining the object's spatial movement data. The object's spatial movement data is further used for ship motion trajectory analysis. Specifically, using the ship's historical trajectory data and real-time motion data, through technical means such as trajectory matching and path fitting, the motion trajectory of the object is modeled and analyzed. Common methods include the least squares method, polynomial fitting, etc. By analyzing the ship's state at different time points, ship motion trajectory analysis data is generated, which reflects the ship's moving path, velocity change, and turning points along the way. Trajectory analysis helps to provide a scientific basis for subsequent applications such as path planning and collision warning. The ship's motion trajectory analysis data is combined with the object's spatial movement data, and through the calculation of motion fluid disturbance, ship collision probability disturbance data is obtained. This calculation involves a fluid mechanics model, especially when considering the influence of fluid factors such as water flow, wind speed, and tides on the ship's motion. Using fluid dynamics simulation, combined with the ship's velocity, direction, and disturbance factors of the external environment, the disturbance during the ship's motion is calculated and the collision risk is predicted. In order to accurately calculate the collision probability of the ship under disturbed conditions, stochastic simulation methods such as Monte Carlo simulation need to be used to obtain accurate collision probability data. Finally, based on the ship collision probability disturbance data, an avoidance path is planned. This process involves data analysis and optimization algorithms based on collision probability, such as the Dijkstra algorithm, etc. Combining the ship's current position, velocity, and avoidance target, the optimal avoidance path is calculated.

[0098] Preferably, step S22 includes the following steps:

[0099] Step S221: Perform non-linear Kalman filtering on the object's spatial movement data to construct a motion state matrix and generate a ship motion state prediction matrix;

[0100] Step S222: Perform frequency-domain energy spectrum analysis on the ship motion state prediction matrix, extract the main frequencies of rolling and surge, and generate a ship frequency-domain energy distribution map;

[0101] Step S223: Conduct ship motion stability boundary analysis on the ship motion state prediction matrix through the Lyapunov exponent stability criterion to generate an acceleration increment safety threshold range; perform dynamic truncation correction processing on the ship frequency-domain energy distribution map according to the acceleration increment safety threshold range to generate ship motion trajectory analysis data.

[0102] In the embodiments of the present invention, the object space movement data is processed by non-linear Kalman filtering to construct a motion state matrix and generate a ship motion state prediction matrix. Kalman filtering is an optimal estimation method used to process dynamic systems with noise. During the non-linear Kalman filtering process, first, based on the dynamic model of the ship (such as the motion equation) and the actual observation data, the filtering algorithm is used to estimate the motion state of the ship. Especially when the system has non-linear characteristics, methods such as Extended Kalman Filtering (EKF) or Unscented Kalman Filtering (UKF) are used for state prediction. These methods continuously update the estimated values of the state variables to obtain an accurate ship motion state prediction matrix, which covers important parameters such as the position, speed, and acceleration of the ship. These prediction matrices provide key data support for subsequent motion analysis, trajectory prediction, and stability assessment. The ship motion state prediction matrix will be subjected to frequency-domain energy spectrum analysis. The frequency-domain energy spectrum analysis is based on the Fourier transform, which converts the time-domain signal into a frequency-domain representation, thereby enabling the revelation of the main frequency components during the ship's motion. During this process, by calculating the energy spectral density of the ship's motion, especially extracting the frequency components of rolling and surge, the motion characteristics of the ship in different directions can be identified. Through the energy spectrum analysis, the energy distribution of the ship at different frequencies can be obtained, and the main frequency components can be extracted. These main frequencies reflect the inherent vibration modes of the ship's motion. The generated ship frequency-domain energy distribution map provides frequency-domain information for subsequent stability analysis and motion state prediction, helping to reveal the dynamic behavior of the ship under external disturbances. The Lyapunov exponent stability criterion is used to conduct a motion stability boundary analysis on the ship motion state prediction matrix. The Lyapunov exponent is an important tool for judging the stability of a system. By analyzing the time evolution of the ship's motion state, the Lyapunov exponent can reveal whether the ship is in a stable state, especially in the presence of external disturbances. By calculating the Lyapunov exponent, the stability boundary of the ship's motion can be determined, that is, the maximum acceleration increment that the ship can withstand under different conditions. This safety threshold helps to provide a stability boundary for the ship's motion and avoid dangers such as ship out-of-control or collision caused by exceeding the safety threshold. Finally, according to the acceleration increment safety threshold range, the ship frequency-domain energy distribution map is dynamically truncated and corrected to further optimize the analysis data of the ship motion trajectory.

[0103] Preferably, the ship motion stability boundary analysis of the ship motion state prediction matrix by the Lyapunov exponent stability criterion includes the following:

[0104] Extract the maximum Lyapunov exponent of the ship motion state prediction matrix to generate a characterization of the ship's initial condition sensitivity;

[0105] Calculate the gradient field through the characterization of the ship's initial condition sensitivity to generate a flow field velocity gradient tensor;

[0106] Construct an implicit function relationship of the critical instability acceleration of a ship based on the velocity gradient tensor of the flow field to generate an implicit function of ship critical instability;

[0107] Perform motion stability boundary analysis based on the implicit function of ship critical instability to generate a safe threshold range of acceleration increment.

[0108] In the embodiment of the present invention, by extracting the maximum Lyapunov exponent from the ship motion state prediction matrix, it aims to characterize the sensitivity of the ship to the initial conditions through the Lyapunov exponent. As a classic stability analysis tool, the Lyapunov exponent can evaluate the sensitivity of the ship system to the initial conditions, that is, the stability of the system. If the Lyapunov exponent is positive, the system is unstable; conversely, if it is negative, the system is stable. By calculating the maximum Lyapunov exponent, the stability analysis result of the ship under the initial conditions can be obtained, which provides basic data support for subsequent ship motion state prediction and optimization. Using the characterization of the ship's initial condition sensitivity to perform gradient field calculation to generate the velocity gradient tensor of the flow field. The purpose of gradient field calculation is to capture the changes in the flow field during the ship's motion, especially the velocity distribution and variation law at different spatial positions. By calculating the gradient of the velocity field, the velocity gradient tensor of the flow field can be obtained, which reflects the velocity change trend during the ship's motion and the flow velocity changes in different directions. These data are crucial for describing the interaction between the ship and the surrounding fluid and analyzing the stability of the ship's motion. Based on the above-calculated velocity gradient tensor of the flow field, further construct an implicit function relationship of the critical instability acceleration of the ship. The critical instability acceleration refers to the situation where the ship becomes unstable when it exceeds a certain acceleration threshold under external disturbances. By analyzing the velocity gradient of the flow field and the ship's motion response, an implicit function relationship between acceleration and ship instability can be established. This implicit function describes the dynamic connection between ship acceleration and instability under specific conditions through mathematical modeling and numerical simulation, and can predict the critical point of instability according to the actual motion state. Finally, perform motion stability boundary analysis based on the implicit function of ship critical instability to generate a safe threshold range of acceleration increment. Motion stability boundary analysis is to evaluate the acceleration of the ship in different motion states through the established implicit function relationship, and then determine the safe operating range of the ship. The safe threshold range of acceleration increment provides a stable operating boundary for the ship when encountering external disturbances, ensuring that the ship's motion does not exceed the safe range, thereby avoiding safety hazards such as instability or collision.

[0109] As an example of the present invention, refer to Figure 3 As shown, in this example, step S3 includes:

[0110] Step S31: Obtain onshore area data;

[0111] Step S32: Conduct trajectory correlation and intersection analysis based on the ship motion trajectory analysis data and the ship avoidance path data to generate ship trajectory intersection analysis data;

[0112] Step S33: Perform early data-level fusion on the onshore area data to generate input data for the amphibious ship model; conduct model training based on the input data for the amphibious ship model, and perform model confidence parameter calibration to generate a confidence amphibious detection model;

[0113] Step S34: Perform late feature fusion based on the ship trajectory intersection analysis data to generate amphibious ship feature fusion data; perform trajectory matching and weighted optimization on the confidence amphibious detection model with the amphibious ship feature fusion data to generate an amphibious ship target detection model.

[0114] In the embodiments of the present invention, onshore area data is acquired and processed to extract environmental features related to ship target detection. These data include geographical information, meteorological data, building layouts, ground moving objects, etc., and these factors play a key role in subsequent detection and trajectory analysis. Subsequently, based on the ship movement trajectory analysis data and the ship avoidance path data, a trajectory correlation cross-analysis is performed. This analysis identifies the correlation between the movement behavior of a ship and its avoidance path under specific environments and conditions through time and space matching, generating ship trajectory cross-analysis data. This data provides a basis for subsequent behavior prediction and path optimization, and a scientific basis for understanding ship movement patterns. Early data-level fusion of onshore area data aims to combine the environmental data of the onshore area with the movement trajectory data of the ship to generate input data for the amphibious ship model. This fusion process involves the integration of multi-source data. Through technical means such as data cleaning, normalization, and time alignment, data from different sources is unified into a processable standard format for subsequent modeling and analysis. Based on these fused input data, a machine learning method is used to train the model, and the reliability of the model is further improved through confidence parameter calibration. During the training process, the model captures the movement laws of ships in various environments by learning a large amount of historical data, generating a confidence amphibious detection model. Late feature fusion is performed based on the ship trajectory cross-analysis data. This process uses various features extracted in the early analysis, such as the movement trajectory of the ship, the avoidance path, and changes in the surrounding environment, for feature extraction and fusion, thereby generating amphibious ship feature fusion data. These features include multi-dimensional fusion of data such as time series, spatial coordinates, speed, and acceleration, forming a comprehensive description of ship behavior. Finally, the confidence amphibious detection model is optimized by weighted trajectory matching using the generated feature fusion data, further improving the accuracy of the model and optimizing the target detection ability. In this process, a weighted optimization algorithm is used. By weighted adjustment of the importance of different features, the model can more accurately identify amphibious ship targets in practical applications.

[0115] Preferably, step S32 includes the following steps:

[0116] Step S321: Perform ship avoidance marking on the ship avoidance path data to generate ship avoidance marking data; perform two-dimensional feature extraction of deceleration-avoidance time features on the ship avoidance marking data to generate ship avoidance marking feature data;

[0117] Step S322: Perform time window matching on the ship avoidance marking feature data and the ship movement trajectory analysis data to generate ship time window avoidance-path matching data; perform Euclidean distance trajectory similarity metric matching on the ship time window avoidance-path matching data to generate ship movement trajectory-avoidance path matching data;

[0118] Step S323: Identify the ship behavior pattern from the ship movement trajectory-avoidance path matching data to generate ship movement behavior pattern feature data; perform association cross-analysis of avoidance types and trajectories based on the ship movement behavior pattern feature data to generate ship trajectory cross-analysis data.

[0119] In the embodiments of the present invention, by analyzing the path changes of the ship and motion characteristics such as speed and acceleration, the key nodes of ship avoidance and the occurrence times of avoidance behaviors are calibrated to generate ship avoidance marking data. Further, deceleration-avoidance time feature extraction is performed on these marking data, and combined with the changes in deceleration and time dimensions during ship avoidance, two-dimensional features are extracted. This process generates ship avoidance marking feature data through time series analysis and statistical feature extraction methods, comprehensively reflecting the key behavior patterns of ship avoidance. The ship avoidance marking feature data is matched with the ship movement trajectory analysis data in a time window. The time window is divided by segmenting the time series of the ship movement trajectory data to ensure comparison and matching within the same time range, thereby generating ship time window avoidance-path matching data. To further improve the matching accuracy, the Euclidean distance trajectory similarity measurement method is used to match the ship trajectories within the time window. This method calculates the Euclidean distance between the trajectories to measure the trajectory similarity, providing a quantitative basis for the matching between the ship movement trajectory and the avoidance path, and generating ship movement trajectory-avoidance path matching data. Finally, based on the ship movement trajectory-avoidance path matching data, ship behavior pattern recognition is performed. This process uses machine learning and pattern recognition technologies to analyze the matching data and identify the behavior patterns shown by the ship during avoidance. This analysis not only focuses on the spatio-temporal laws of avoidance actions but also involves the interaction patterns between the ship and other ships and obstacles in the complex marine environment, generating ship movement behavior pattern feature data. Subsequently, through the association cross-analysis of avoidance types and trajectories, the association relationships between different avoidance types and trajectories are further explored, thereby generating ship trajectory cross-analysis data.

[0120] As an example of the present invention, referring to Figure 4 as shown, in this example, step S4 includes:

[0121] Step S41: Based on the target detection and recognition learning of the amphibious ship target detection model, obtain object detection target data;

[0122] Step S42: Generate an amphibious ship target detection module according to the object detection target data;

[0123] Step S43: Embed the amphibious ship target detection module into the radar system to perform the amphibious ship target detection task.

[0124] In the embodiments of the present invention, through target detection and recognition learning of the amphibious ship target detection model, training is carried out based on a large amount of ship target data and marine environment data to extract the feature data of objects. During the training process, the model uses deep learning algorithms, especially technologies such as convolutional neural networks (CNNs), to automatically learn the relationship between the ship and the surrounding environment and identify different types of target objects. Through the gradual analysis of the training data set, the target detection and recognition learning obtains the object detection target data, which includes different ship types, the spatial positions, speeds, and their related feature information of the target objects. The core technology in this stage lies in optimizing the target recognition algorithm through a large amount of labeled data to make it have high recognition accuracy and generalization ability. Subsequently, according to the object detection target data, automated code generation is performed, and then the amphibious ship target detection module is generated. In this process, the object detection target data is used as input, and through automated programming tools (such as machine learning automation frameworks, code generation toolchains, etc.), algorithm codes related to target detection are generated. These codes can implement the entire process of operations from data collection to target recognition, including data preprocessing, feature extraction, classification prediction, and other links. Through templatized and modular design, the automated code generation enables the target detection system to be quickly deployed and adapt to different application scenarios and technical requirements. The key technology of code generation depends on advanced automated programming technologies and the design of model interfaces, making the detection module highly adaptable. Finally, the generated amphibious ship target detection module is embedded in the radar system to perform the amphibious ship target detection task. In this stage, the target detection module is embedded in the core processing unit of the radar system. The radar system receives data from sensors in real time and transmits it to the detection module for processing. At this time, the target detection module can quickly identify and track the detected targets according to the real-time data stream through the previously learned recognition model, generate corresponding target detection results and feedback them to the system, thereby effectively performing the detection task of amphibious ship targets.

[0125] In this specification, an amphibious ship target detection system is provided for performing the above-mentioned amphibious ship target detection method. The amphibious ship target detection system includes:

[0126] A data collection and space construction module, configured to obtain multi-modal sensor monitoring data; perform monitoring signal phase alignment processing on the multi-modal sensor monitoring data to generate multi-modal sensor phase alignment data; and construct a three-dimensional ship space monitoring field based on the multi-modal sensor phase alignment data to obtain a three-dimensional ship space monitoring field;

[0127] A motion trajectory analysis and path planning module, configured to perform ship motion trajectory analysis on the three-dimensional ship space monitoring field to generate ship motion trajectory analysis data; and perform avoidance path planning on the ship motion trajectory analysis data to generate ship avoidance path data;

[0128] A trajectory association and target detection model construction module, configured to obtain onshore area data; perform trajectory association and cross-analysis based on ship motion trajectory analysis data and ship avoidance path data to generate ship trajectory cross-analysis data; perform a confidence evaluation model on the onshore area data and the ship trajectory cross-analysis data to generate an amphibious ship target detection model.

[0129] A target detection and execution module, configured to perform target detection and recognition learning on the amphibious ship target detection model to obtain object detection target data; embed the object detection target data into a radar system to execute the amphibious ship target detection task.

[0130] The beneficial effects of the present invention are as follows: By obtaining multi-modal sensor monitoring data and performing phase alignment processing on these data, the consistency of the monitoring signals of each sensor in time series is ensured, thereby reducing data deviation caused by signal delay or synchronization error between different sensors. Then, based on the phase-aligned data, a three-dimensional space monitoring field of the ship is constructed, and through the fusion of spatial data, a comprehensive and accurate ship monitoring model is formed, providing high-quality basic data for subsequent trajectory analysis and path planning. In the second step, by analyzing the motion trajectory of the ship's three-dimensional space monitoring field, detailed ship motion trajectory data is generated, and combined with the ship's motion law and environmental changes, an avoidance path planning is implemented to ensure that the ship can autonomously avoid obstacles or other ships in different complex environments, thereby improving the safety and maneuverability of the ship. Next, by obtaining onshore area data and performing cross-analysis with the ship's motion trajectory data, ship trajectory cross-analysis data is formed, further optimizing the ship path planning, and a confidence evaluation model is constructed based on this to ensure the accuracy and reliability of the data in practical applications, and finally an amphibious ship target detection model is generated. This model combines target recognition algorithms to learn and identify the detection data, thereby realizing real-time monitoring and identification of targets on the water surface and on land. Finally, by embedding the object detection target data into the radar system, a complete detection system is formed to execute the amphibious ship target detection task. Through this series of technical means, the present invention improves the target detection ability and avoidance efficiency of the ship in complex environments. Therefore, the present invention solves the problems of low accuracy, slow response, and poor environmental adaptability of traditional ship target detection systems by integrating multi-modal data processing and intelligent path planning technologies, and improves the target recognition ability and obstacle avoidance ability of ships in multiple environments.

[0131] Therefore, in any regard, the embodiments should be regarded as exemplary and non-limiting. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, all changes falling within the meaning and scope of the equivalent elements of the application documents are intended to be encompassed by the present invention.

[0132] The above are only specific embodiments of the present invention, enabling those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be obvious to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to these embodiments shown herein, but rather to the widest scope consistent with the principles and novel features invented herein.

Claims

1. An amphibious ship target detection method, characterized in that It includes the following steps: Step S1: Obtain the monitoring data of the multimodal sensor; perform monitoring signal phase alignment processing on the monitoring data of the multimodal sensor to generate multimodal sensor phase-aligned data; construct a three-dimensional monitoring field of the ship's space based on the multimodal sensor phase-aligned data to obtain a three-dimensional monitoring field of the ship's space; Step S2: Analyze the ship's motion trajectory in the three-dimensional monitoring field of the ship to generate ship motion trajectory analysis data; plan an avoidance path for the ship motion trajectory analysis data to generate ship avoidance path data; Step S3: Obtain land area data; perform trajectory correlation and cross-analysis based on the ship motion trajectory analysis data and the ship avoidance path data to generate ship trajectory cross-analysis data; perform early data-level fusion on the land area data to generate input data for the amphibious ship model; Train the model based on the input data of the amphibious ship model and calibrate the model confidence parameters to generate a confidence amphibious detection model; Perform late feature fusion based on the ship trajectory cross-analysis data to generate amphibious ship feature fusion data; Use the amphibious ship feature fusion data to perform trajectory matching and weighted optimization on the confidence amphibious detection model to generate an amphibious ship target detection model; Step S4: Perform target detection and recognition learning on the amphibious ship target detection model to obtain object detection target data; Embed the object detection target data into the radar system to perform the amphibious ship target detection task.

2. The amphibious ship target detection method according to claim 1, characterized in that, Step S1 includes the following steps: Step S11: Obtain the monitoring data of the multimodal sensor; Step S12: Perform monitoring signal alignment processing on the monitoring data of the multimodal sensor to generate multimodal sensor phase-aligned data; perform monitoring object morphology recognition on the multimodal sensor phase-aligned data to obtain monitoring object morphology recognition data; Step S13: Construct a three-dimensional monitoring field of the ship's space based on the monitoring object morphology recognition data to obtain a three-dimensional monitoring field of the ship's space.

3. The amphibious ship target detection method according to claim 2, characterized in that, Step S13 includes the following steps: Step S131: Perform geometric edge calculation on the monitoring object morphology recognition data to generate monitoring sensor edge calculation data; Step S132: Analyze the geometric constraint relationship of the monitoring sensor based on the Simultaneous Localization and Mapping (SLAM) technology for the monitoring sensor edge calculation data to generate monitoring sensor geometric constraint data; perform spatio-temporal compensation on the common viewing area of the monitoring sensors for the monitoring sensor geometric constraint data to generate monitoring sensor spatio-temporal calibration data; Step S133: Obtain the position of the coastal feature points; perform spatio-temporal region mapping on the monitoring sensor spatio-temporal calibration data and the position of the coastal feature points, and use Particle Swarm Optimization (PSO) to correct the sensor registration error to generate ship space sensor data; Step S134: Obtain the Bird's Eye View (BEV); perform multimodal distance fusion on the ship space sensor data using the BEV and construct a three-dimensional monitoring field of the ship's space to obtain a three-dimensional monitoring field of the ship's space.

4. The amphibious ship target detection method according to claim 1, wherein Step S2 includes the following steps: Step S21: Analyze the spatial motion of the target object based on the three-dimensional monitoring field of the ship to obtain object spatial movement data; Step S22: Analyze the ship's motion trajectory from the object's spatial movement data to generate ship motion trajectory analysis data; Step S23: Calculate the motion fluid disturbance of the object's spatial movement data and the ship motion trajectory analysis data to generate ship collision probability disturbance data; Plan an avoidance path based on the ship collision probability disturbance data to generate ship avoidance path data.

5. The amphibious ship target detection method according to claim 1, characterized in that, Step S22 includes the following steps: Step S221: Perform non-linear Kalman filtering on the object's spatial movement data to construct a motion state matrix, and generate a ship motion state prediction matrix; Step S222: Conduct frequency-domain energy spectrum analysis on the ship motion state prediction matrix, and extract the main frequencies of rolling and surge to generate a ship frequency-domain energy distribution map; Step S223: Analyze the ship motion stability boundary of the ship motion state prediction matrix through the Lyapunov exponent stability criterion to generate an acceleration increment safety threshold range; Perform dynamic truncation correction on the ship frequency-domain energy distribution map according to the acceleration increment safety threshold range to generate ship motion trajectory analysis data.

6. The amphibious ship target detection method according to claim 5, wherein, Analyzing the ship motion stability boundary of the ship motion state prediction matrix through the Lyapunov exponent stability criterion includes the following: Extract the maximum Lyapunov exponent of the ship motion state prediction matrix to generate a ship initial condition sensitivity characterization; Calculate the gradient field through the ship initial condition sensitivity characterization to generate a flow field velocity gradient tensor; Construct an implicit function relationship of the ship's critical instability acceleration based on the flow field velocity gradient tensor to generate a ship critical instability implicit function; Based on the ship critical instability implicit function, conduct ship motion stability boundary analysis to generate an acceleration increment safety threshold range.

7. The amphibious ship target detection method according to claim 1, characterized in that, Performing trajectory correlation and intersection analysis based on the ship motion trajectory analysis data and the ship avoidance path data to generate ship trajectory intersection analysis data includes the following steps: Mark the ship avoidance on the ship avoidance path data to generate ship avoidance mark data; Extract two-dimensional features of deceleration-avoidance time characteristics from the ship avoidance mark data to generate ship avoidance mark feature data; Match the ship avoidance mark feature data and the ship motion trajectory analysis data in a time window to generate ship time window avoidance-path matching data; Perform Euclidean distance trajectory similarity metric matching on the ship time window avoidance-path matching data to generate ship motion trajectory-avoidance path matching data; Perform ship behavior pattern recognition on the ship motion trajectory-avoidance path matching data to generate ship motion behavior pattern feature data; Based on the ship motion behavior pattern feature data, conduct avoidance type trajectory correlation and intersection analysis to generate ship trajectory intersection analysis data.

8. The amphibious ship target detection method according to claim 1, characterized in that Step S4 includes the following steps: Step S41: Based on the target detection and recognition learning of the amphibious ship target detection model, obtain object detection target data; Step S42: Generate an amphibious ship target detection module according to the object detection target data; Step S43: Embed the amphibious ship target detection module into the radar system to perform the amphibious ship target detection task.

9. An amphibious ship target detection system, characterized in that, For performing the amphibious ship target detection method described in claim 1, the amphibious ship target detection system includes: A data acquisition and space construction module, configured to obtain multi-modal sensor monitoring data; perform monitoring signal phase alignment processing on the multi-modal sensor monitoring data to generate multi-modal sensor phase-aligned data; construct a three-dimensional monitoring field of the ship's space based on the multi-modal sensor phase-aligned data to obtain a three-dimensional monitoring field of the ship's space; A motion trajectory analysis and path planning module, configured to perform ship motion trajectory analysis on the three-dimensional monitoring field of the ship to generate ship motion trajectory analysis data; perform avoidance path planning on the ship motion trajectory analysis data to generate ship avoidance path data; A trajectory correlation and target detection model construction module, configured to obtain onshore area data; perform trajectory correlation and cross-analysis based on the ship motion trajectory analysis data and the ship avoidance path data to generate ship trajectory cross-analysis data; perform a confidence evaluation model on the onshore area data and the ship trajectory cross-analysis data to generate an amphibious ship target detection model; A target detection and execution module, configured to perform target detection and recognition learning on the amphibious ship target detection model to obtain object detection target data; embed the object detection target data into the radar system to perform the amphibious ship target detection task.

Images