Method and system for detecting beach swimming behavior based on radar tracks

Abstract

The application provides a detection method and system for seaside swimming behavior based on radar trajectory, comprising: receiving discrete point tracks detected by a radar as source data for analysis; aggregating the discrete point tracks to obtain radar continuous tracks; preprocessing the radar continuous tracks to filter out radar targets of non-swimming behavior; extracting features from the filtered radar continuous tracks to obtain speed features, heading features, position change features, distance information features from all radar devices and track appearance time distribution features; performing weighted calculation on the features; obtaining a swimming behavior probability; when the swimming behavior probability is greater than a preset value, it is considered that the seaside swimming behavior is met, and the detection is successful. The application proposes a method for analyzing and detecting swimming behavior by using radar trajectory suitable for a coastal water scene, detects people swimming on the coast, and ensures the safety and stability of the sea area.

Description

Technical Field

[0001] This invention relates to the technical field of marine detection, specifically to a method and system for detecting swimming behavior at the beach based on radar trajectories. Background Technology

[0002] Due to the complex coastal environment, dangerous swimming methods exist along unsupervised coastlines, and despite countries investing significant human and material resources in maritime supervision, no effective solution has yet been found.

[0003] Patent document CN112561970A discloses a method for managing trajectory segments in a particle filter estimation framework, comprising the following steps: performing trajectory segment prediction based on a list of previous trajectory segments (Tp,t) to determine surviving trajectory segments and new trajectory segments (Tnew); sampling new measurements used to initialize the new trajectory segments (Tnew) to determine the number of estimated new trajectory segments (Neti); determining the number of surviving trajectory segments (Np,t) based on the list of previous trajectory segments (Tp,t); and determining the number of surviving trajectory segments (Neti) based on the number of estimated new trajectory segments (Neti). i) The number of new trajectory segments to be sampled (Nnew) and the number of updated surviving trajectory segments (Np,t+1) are determined by the number of surviving trajectory segments (Np,t) and the memory limit (Nmax); the updated surviving trajectory segments (Tp,t+1) are sampled from the list of surviving trajectory segments (Tp,t); and the new trajectory segments (Tnew) are sampled from the unassociated measurement results, depending on the determined number of new trajectory segments (Nnew).

[0004] Regarding the aforementioned technologies, the inventors believe that the lack of existing detection methods for swimming leads to dangerous swimming practices on unsupervised coastlines, and despite significant investment of manpower and resources in maritime supervision, no effective solution has been found. Therefore, a new technical solution is needed to address these technical problems. Summary of the Invention

[0005] In view of the shortcomings of the prior art, the purpose of this invention is to provide a method and system for detecting swimming behavior at the beach based on radar trajectory.

[0006] According to the present invention, a radar trajectory-based detection method for swimming behavior at the beach is provided, the method comprising the following steps:

[0007] Step S1: Receive the discrete point traces detected by the radar as source data for analysis;

[0008] Step S2: Aggregate the discrete points detected by the radar to obtain a complete continuous radar trajectory;

[0009] Step S3: Preprocess the complete continuous radar trajectory to filter out radar targets that do not exhibit swimming behavior;

[0010] Step S4: Extract features from the filtered continuous radar trajectory to obtain speed features, heading features, position change features, distance information features to all radar devices, and trajectory occurrence time distribution features;

[0011] Step S5: The obtained speed characteristics, heading characteristics, position change characteristics, distance information characteristics with all radar devices, and trajectory occurrence time distribution characteristics are weighted according to the likelihood function to obtain the final swimming behavior probability;

[0012] Step S6: When the probability of swimming behavior is greater than the preset value, the radar target is considered to meet the swimming behavior at the beach; when the probability of swimming behavior is less than or equal to the threshold, the radar target is considered not to meet the swimming behavior characteristics.

[0013] Preferably, in step S1, the information of the discrete point traces detected by the radar includes the latitude and longitude of the location point, the heading and speed of the location point, the time of the location point, and the radar target ID number of the location point.

[0014] Preferably, in step S2, the discrete points detected by the radar are aggregated according to the radar target ID number.

[0015] Preferably, in step S3, the preprocessing method for the complete continuous radar trajectory is as follows: filtering out long-term stationary targets and high-speed moving targets; determining whether it is a long-term stationary target based on the trajectory's movement distance; and determining whether it is a high-speed moving target based on the trajectory's speed.

[0016] Preferably, in step S4, feature extraction is performed on the filtered radar continuous trajectory. The extracted features include speed features, heading features, position change features, distance features to all radar devices, and trajectory occurrence time distribution features.

[0017] The speed characteristics include maximum speed information, minimum speed information, average speed information, and average speed change information;

[0018] The heading characteristics include heading distribution information and average heading change information;

[0019] The location change characteristics include average movement distance information, maximum movement distance information, and total movement distance information;

[0020] The distance information features with all radar devices include distance information with each radar device and maximum distance information with all detection radars;

[0021] The time distribution characteristics of the trajectory include the time information of the first appearance of the trajectory.

[0022] The present invention also provides a radar trajectory-based detection system for swimming behavior at the beach, the system comprising the following modules:

[0023] Module M1: Receives discrete point traces detected by radar as source data for analysis;

[0024] Module M2: Aggregates the discrete points detected by radar to obtain a complete continuous radar trajectory;

[0025] Module M3: Preprocesses the complete continuous radar trajectory to filter out radar targets that do not exhibit swimming behavior;

[0026] Module M4: Extracts features from the filtered continuous radar trajectory to obtain speed features, heading features, position change features, distance information features to all radar devices, and trajectory occurrence time distribution features;

[0027] Module M5: The obtained speed characteristics, heading characteristics, position change characteristics, distance information characteristics with all radar devices, and trajectory occurrence time distribution characteristics are weighted according to the likelihood function to obtain the final swimming behavior probability;

[0028] Module M6: When the probability of swimming behavior is greater than the preset value, the radar target is considered to meet the swimming behavior at the beach; when the probability of swimming behavior is less than or equal to the threshold, the radar target is considered not to meet the swimming behavior characteristics.

[0029] Preferably, in module M1, the information of the discrete points detected by the radar includes the latitude and longitude of the location point, the heading and speed of the location point, the time of the location point, and the radar target ID number of the location point.

[0030] Preferably, in module M2, the discrete points detected by radar are aggregated according to the radar target ID number.

[0031] Preferably, in module M3, the preprocessing method for the complete continuous radar trajectory is as follows: filtering out long-term stationary targets and high-speed moving targets; determining whether it is a long-term stationary target based on the trajectory's movement distance; and determining whether it is a high-speed moving target based on the trajectory's speed.

[0032] Preferably, in module M4, feature extraction is performed on the filtered continuous radar trajectory. The extracted features include speed features, heading features, position change features, distance features to all radar devices, and trajectory occurrence time distribution features.

[0033] The speed characteristics include maximum speed information, minimum speed information, average speed information, and average speed change information;

[0034] The heading characteristics include heading distribution information and average heading change information;

[0035] The location change characteristics include average movement distance information, maximum movement distance information, and total movement distance information;

[0036] The distance information features with all radar devices include distance information with each radar device and maximum distance information with all detection radars;

[0037] The time distribution characteristics of the trajectory include the time information of the first appearance of the trajectory.

[0038] Compared with the prior art, the present invention has the following beneficial effects:

[0039] 1. This invention automatically detects swimming behavior by analyzing radar patterns, which can reduce the workload of manual coastline patrols and free up manpower and resources;

[0040] 2. This invention analyzes multidimensional information about the trajectory and takes all factors into account during detection through weighted calculation, which will effectively reduce the false alarm rate and improve the detection rate;

[0041] 3. In the actual detection process, the present invention will perform data preprocessing, first filtering out stationary targets and high-speed moving targets, which greatly reduces the amount of calculation.

[0042] 4. This invention proposes a method for detecting swimming behavior on the coast, providing a solution for the coastal monitoring system to promptly detect and stop dangerous swimming behavior, which can greatly facilitate the coastal monitoring department and ensure the safety and stability of the sea area. Attached Figure Description

[0043] Other features, objects, and advantages of the present invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0044] Figure 1 This is a flowchart of the swimming behavior detection method of the present invention. Detailed Implementation

[0045] The present invention will now be described in detail with reference to specific embodiments. These embodiments will help those skilled in the art to further understand the present invention, but do not limit the invention in any way. It should be noted that those skilled in the art can make several changes and improvements without departing from the concept of the present invention. These all fall within the protection scope of the present invention.

[0046] Example 1:

[0047] According to the present invention, a radar trajectory-based detection method for swimming behavior at the beach is provided, the method comprising the following steps:

[0048] Step S1: Receive the discrete point traces detected by radar as source data for analysis; the information of the discrete point traces detected by radar includes the latitude and longitude of the location point, the heading and speed of the location point, the time of the location point, and the radar target ID number of the location point.

[0049] Step S2: Aggregate the discrete points detected by the radar to obtain a complete continuous radar trajectory; the discrete points detected by the radar are aggregated according to the radar target ID number.

[0050] Step S3: Preprocess the complete continuous radar trajectory to filter out radar targets that do not swim. The preprocessing method for the complete continuous radar trajectory is as follows: filter out targets that remain stationary for a long time and targets that move at high speed; determine whether it is a target that remains stationary for a long time based on the trajectory's movement distance; determine whether it is a target that moves at high speed based on the trajectory's speed. "Long time" usually refers to 20 minutes, which can be adjusted appropriately based on the actual radar detection situation; "high speed" is usually 4 knots, which is the maximum speed of a normal person swimming.

[0051] Step S4: Extract features from the filtered continuous radar trajectory to obtain speed features, heading features, position change features, distance information features to all radar devices, and trajectory appearance time distribution features. The extracted features include speed features, heading features, position change features, distance information to all radar devices, and trajectory appearance time distribution features. The speed features include maximum speed information, minimum speed information, average speed information, and average speed change information. The heading features include heading distribution information and average heading change information. The position change features include average distance traveled information, maximum distance traveled information, and total distance traveled information. The distance information features to all radar devices include distance information to each radar device and maximum distance information to all detected radars. The trajectory appearance time distribution features include the time point when the trajectory first appears.

[0052] Step S5: The obtained speed characteristics, heading characteristics, position change characteristics, distance information characteristics with all radar devices, and trajectory occurrence time distribution characteristics are weighted and calculated according to the likelihood function to obtain the final swimming behavior probability.

[0053] Step S6: When the probability of swimming behavior is greater than the preset value, the radar target is considered to meet the swimming behavior at the beach; when the probability of swimming behavior is less than or equal to the threshold, the radar target is considered not to meet the swimming behavior characteristics.

[0054] The present invention also provides a radar trajectory-based detection system for swimming behavior at the beach. The radar trajectory-based detection system for swimming behavior at the beach can be implemented by executing the process steps of the radar trajectory-based detection method for swimming behavior at the beach. That is, those skilled in the art can understand the radar trajectory-based detection method for swimming behavior at the beach as a preferred embodiment of the radar trajectory-based detection system for swimming behavior at the beach.

[0055] Example 2:

[0056] The present invention also provides a radar trajectory-based detection system for swimming behavior at the beach, the system comprising the following modules:

[0057] Module M1: Receives discrete point traces detected by radar as source data for analysis; the information of the discrete point traces detected by radar includes the latitude and longitude of the location point, the heading and speed of the location point, the time of the location point, and the radar target ID number of the location point.

[0058] Module M2: Aggregates discrete points detected by radar to obtain a complete continuous radar trajectory; the method of aggregating discrete points detected by radar is to aggregate them according to the radar target ID number.

[0059] Module M3: Preprocesses the complete continuous radar trajectory to filter out radar targets that do not swim. The preprocessing method for the complete continuous radar trajectory is as follows: filtering out targets that have been stationary for a long time and targets that move at high speed; determining whether a target has been stationary for a long time based on the trajectory's distance; and determining whether a target moves at high speed based on the trajectory's speed.

[0060] Module M4: Extracts features from the filtered continuous radar trajectory to obtain speed features, heading features, position change features, distance information features to all radar devices, and trajectory appearance time distribution features. The extracted features include speed features, heading features, position change features, distance information to all radar devices, and trajectory appearance time distribution features. The speed features include maximum speed information, minimum speed information, average speed information, and average speed change information. The heading features include heading distribution information and average heading change information. The position change features include average distance traveled information, maximum distance traveled information, and total distance traveled information. The distance information features to all radar devices include distance information to each radar device and maximum distance information to all detected radars. The trajectory appearance time distribution features include the time point when the trajectory first appears.

[0061] Module M5: The obtained speed characteristics, heading characteristics, position change characteristics, distance information characteristics with all radar devices, and trajectory occurrence time distribution characteristics are weighted according to the likelihood function to obtain the final swimming behavior probability;

[0062] Module M6: When the probability of swimming behavior is greater than the preset value, the radar target is considered to meet the swimming behavior at the beach; when the probability of swimming behavior is less than or equal to the threshold, the radar target is considered not to meet the swimming behavior characteristics.

[0063] Example 3:

[0064] like Figure 1 As shown, a radar trajectory-based detection method for swimming behavior at the beach, according to the present invention, includes the following steps:

[0065] S1. Receive discrete point traces detected by radar as source data for analysis; wherein the discrete point information detected by radar includes at least the latitude and longitude of the location point; the heading and speed of the location point; the time of the location point; and the radar target ID number of the location point.

[0066] S2. Aggregate the discrete points detected by radar to obtain a complete continuous radar trajectory. The aggregation method for the discrete points detected by radar is to aggregate them according to the radar target ID number. The aggregation method is to arrange and connect the points with the same radar target ID in chronological order to obtain the state chain at different time points, i.e., the continuous trajectory.

[0067] S3. Preprocess the complete continuous radar trajectory to filter out radar targets that do not exhibit swimming behavior. The preprocessing method for the complete continuous radar trajectory is as follows: filter out targets that have been stationary for a long time and targets that are moving at high speeds; determine whether a target has been stationary for a long time based on the trajectory's movement distance; determine whether a target is moving at high speeds based on the trajectory's speed; the filtering method is to calculate the target's historical movement range. Targets with a movement range span of <0.0005° are considered stationary targets; targets with a speed >5 knots are considered high-speed moving targets. Neither of these types of targets can be generated by normal swimming behavior, so they can be directly filtered out.

[0068] S4. Perform feature extraction on the filtered continuous radar trajectory to obtain speed features, heading features, position change features, distance information features to all radar devices, and trajectory occurrence time distribution features; the feature extraction method is as follows:

[0069] The speed characteristics include maximum speed information, minimum speed information, average speed information, and average speed change information;

[0070] Heading characteristics include heading distribution information and average heading change information;

[0071] The location change characteristics include average movement distance information, maximum movement distance information, and total movement distance information;

[0072] The distance information feature with all radar devices includes distance information with each radar device and maximum distance information with all detected radars;

[0073] The time distribution characteristics of trajectory appearance include information on the time point when the trajectory first appears.

[0074] S5. The obtained features are weighted according to the likelihood function to obtain the final swimming behavior probability. The swimming behavior probability is calculated by weighting and summing the obtained features according to the likelihood function.

[0075] S6. When the probability of swimming behavior is greater than the preset value, the radar target is considered to meet the swimming behavior at the beach; when the probability of swimming behavior is less than or equal to the threshold, the radar target is considered not to meet the swimming behavior characteristics.

[0076] Those skilled in the art can understand this embodiment as a more specific description of Embodiment 1 and Embodiment 2.

[0077] Those skilled in the art will understand that, besides implementing the system and its various devices, modules, and units provided by this invention in the form of purely computer-readable program code, the same functions can be achieved entirely through logical programming of the method steps, making the system and its various devices, modules, and units of this invention function in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers. Therefore, the system and its various devices, modules, and units provided by this invention can be considered as a hardware component, and the devices, modules, and units included therein for implementing various functions can also be considered as structures within the hardware component; alternatively, the devices, modules, and units for implementing various functions can be considered as both software modules implementing the method and structures within the hardware component.

[0078] Specific embodiments of the present invention have been described above. It should be understood that the present invention is not limited to the specific embodiments described above, and those skilled in the art can make various changes or modifications within the scope of the claims, which do not affect the essence of the present invention. Unless otherwise specified, the embodiments and features described in this application can be arbitrarily combined with each other.

Claims

1. A radar trajectory-based detection method for swimming behavior at the beach, characterized in that, The method includes the following steps: Step S1: Receive the discrete point traces detected by the radar as source data for analysis; Step S2: Aggregate the discrete points detected by the radar to obtain a complete continuous radar trajectory; Step S3: Preprocess the complete continuous radar trajectory to filter out radar targets that do not exhibit swimming behavior; Step S4: Extract features from the filtered continuous radar trajectory to obtain speed features, heading features, position change features, distance information features to all radar devices, and trajectory occurrence time distribution features; Step S5: The obtained speed characteristics, heading characteristics, position change characteristics, distance information characteristics with all radar devices, and trajectory occurrence time distribution characteristics are weighted according to the likelihood function to obtain the final swimming behavior probability; Step S6: When the probability of swimming behavior is greater than the preset value, the radar target is considered to meet the swimming behavior at the beach; when the probability of swimming behavior is less than or equal to the preset value, the radar target is considered not to meet the swimming behavior characteristics.

2. The radar trajectory-based detection method for swimming behavior at the beach according to claim 1, characterized in that, In step S1, the information of the discrete points detected by the radar includes the latitude and longitude of the location point, the heading and speed of the location point, the time of the location point, and the radar target ID number of the location point.

3. The radar trajectory-based detection method for swimming behavior at the beach according to claim 1, characterized in that, In step S2, the discrete points detected by the radar are aggregated according to the radar target ID number.

4. The radar trajectory-based detection method for swimming behavior at the beach according to claim 1, characterized in that, In step S3, the preprocessing method for the complete continuous radar trajectory is as follows: filtering out long-term stationary targets and high-speed moving targets; determining whether it is a long-term stationary target based on the trajectory's movement distance; and determining whether it is a high-speed moving target based on the trajectory's speed.

5. The radar trajectory-based detection method for swimming behavior at the beach according to claim 1, characterized in that, In step S4, feature extraction is performed on the filtered radar continuous trajectory. The extracted features include speed features, heading features, position change features, distance features to all radar devices, and trajectory occurrence time distribution features. The speed characteristics include maximum speed information, minimum speed information, average speed information, and average speed change information; The heading characteristics include heading distribution information and average heading change information; The location change characteristics include average movement distance information, maximum movement distance information, and total movement distance information; The distance information features with all radar devices include distance information with each radar device and maximum distance information with all detection radars; The time distribution characteristics of the trajectory include the time information of the first appearance of the trajectory.

6. A radar trajectory-based detection system for swimming behavior at the beach, characterized in that, The system includes the following modules: Module M1: Receives discrete point traces detected by radar as source data for analysis; Module M2: Aggregates the discrete points detected by radar to obtain a complete continuous radar trajectory; Module M3: Preprocesses the complete continuous radar trajectory to filter out radar targets that do not exhibit swimming behavior; Module M4: Extracts features from the filtered continuous radar trajectory to obtain speed features, heading features, position change features, distance information features to all radar devices, and trajectory occurrence time distribution features; Module M5: The obtained speed characteristics, heading characteristics, position change characteristics, distance information characteristics with all radar devices, and trajectory occurrence time distribution characteristics are weighted according to the likelihood function to obtain the final swimming behavior probability; Module M6: When the probability of swimming behavior is greater than the preset value, the radar target is considered to meet the swimming behavior at the beach; when the probability of swimming behavior is less than or equal to the preset value, the radar target is considered not to meet the swimming behavior characteristics.

7. The radar trajectory-based detection system for swimming behavior at the beach according to claim 6, characterized in that, In module M1, the information of discrete points detected by radar includes the latitude and longitude of the location point, the heading and speed of the location point, the time of the location point, and the radar target ID number of the location point.

8. The radar trajectory-based detection system for swimming behavior at the beach according to claim 6, characterized in that, In module M2, the discrete points detected by the radar are aggregated according to the radar target ID number.

9. The radar trajectory-based detection system for swimming behavior at the beach according to claim 6, characterized in that, In module M3, the preprocessing method for the complete continuous radar trajectory is as follows: filtering out long-term stationary targets and high-speed moving targets; determining whether it is a long-term stationary target based on the trajectory's movement distance; and determining whether it is a high-speed moving target based on the trajectory's speed.

10. The radar trajectory-based detection system for swimming behavior at the beach according to claim 6, characterized in that, In module M4, feature extraction is performed on the filtered radar continuous trajectory. The extracted features include speed features, heading features, position change features, distance features to all radar devices, and trajectory occurrence time distribution features. The speed characteristics include maximum speed information, minimum speed information, average speed information, and average speed change information; The heading characteristics include heading distribution information and average heading change information; The location change characteristics include average movement distance information, maximum movement distance information, and total movement distance information; The distance information features with all radar devices include distance information with each radar device and maximum distance information with all detection radars; The time distribution characteristics of the trajectory include the time information of the first appearance of the trajectory.

Images