A sea target detection method, device, equipment and storage medium

Abstract

The application discloses a kind of offshore target detection method, device, equipment and storage medium, comprising: to initial coordinate point set using adaptive filtering algorithm is carried out rough filtration and obtains first screening coordinate point set, and according to first screening coordinate point set dynamic acquisition wake region;First screening coordinate point set is carried out precision filtration based on wake region and obtains second screening coordinate point set;Based on the physical parameters of laser radar, adaptive clustering algorithm is used to the second screening coordinate point set is clustered and obtains detection target.The comprehensiveness of laser radar detection information is improved by fusing point cloud data, and by rough filtering and precision filtering of the initial coordinate point set obtained after fusion, and based on the physical parameters of laser radar, the filtered results are aggregated to complete the detection of each target obstacle, so that the water surface target obstacle can be efficiently and real-time detected by laser radar in complex sea surface environment.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of unmanned ship control, and in particular to a sea target detection method, device, equipment and storage medium. BACKGROUND

[0002] As a multifunctional intelligent platform, the water unmanned ship has been widely used in the marine field in recent years. Due to the lightweight design, economy, long endurance, high mobility, fast navigation and adaptability to harsh environment of the unmanned ship, it has shown great application potential in the fields of marine environment monitoring, seabed topography mapping and resource development.

[0003] In the local path autonomous planning of the water unmanned ship, obstacle detection is one of the core tasks. Traditional ships usually rely on photoelectric devices such as visible light and infrared sensors, and marine radars for environmental perception. Among them, the photoelectric device can provide high refresh rate and high resolution image information, and the marine radar is known for its long detection distance and all-weather navigation capability. However, these two sensors have obvious limitations in complex water areas: the radar has low update frequency and insufficient resolution in the near distance and high-speed target scene, and is easily disturbed by sea clutter in the near shore area; the photoelectric device is unstable in performance when the weather and light change, and cannot accurately detect the distance information and real size of the target object, which is difficult to meet the needs of accurate detection and positioning.

[0004] As an active sensor, the laser radar combines the advantages of optical imaging and radar, and can provide distance, azimuth information and three-dimensional profile data of the target. The active laser ranging technology of the laser radar is not affected by light and weather, and the scanning speed is fast, which is suitable for detection and identification of medium and short range targets. The three-dimensional information acquisition capability of the laser radar makes it a key tool for unmanned ships to detect targets, measure sizes, distances and speeds, and has important value for navigation obstacle avoidance, target positioning and tracking.

[0005] However, the laser radar also faces challenges in practical application. The point cloud data of the laser radar may be sparse in complex sea conditions, and the detection capability for small targets is limited, and it is easily disturbed by sea clutter. The water unmanned ship usually operates in complex environments such as near sea, beach or inland river, and the background is variable and the interference is strong. Therefore, how to realize efficient target detection and identification based on laser radar under complex sea conditions and strong interference conditions has become a technical problem to be solved in the environmental perception technology of unmanned ships. SUMMARY

[0006] The present application provides a sea target detection method, device, equipment and storage medium to realize accurate detection of obstacles for water unmanned ships based on laser radar.

[0007] According to a first aspect of the present application, a sea target detection method is provided, comprising: scanning the surrounding environment of an unmanned surface vehicle by a laser radar to obtain point cloud data, and fusing and converting the point cloud data on a grid map to obtain an initial coordinate point set;

[0008] Coarse filtering the initial coordinate point set by using an adaptive filtering algorithm to obtain a first screened coordinate point set, and dynamically obtaining a wake region according to the first screened coordinate point set;

[0009] Fine filtering the first screened coordinate point set based on the wake region to obtain a second screened coordinate point set;

[0010] Clustering the second screened coordinate point set by using an adaptive clustering algorithm based on the physical parameters of the laser radar to obtain a detection target.

[0011] According to another aspect of the present application, a target detection device is provided, comprising: a fusion conversion module configured to scan the surrounding environment of an unmanned surface vehicle by a laser radar to obtain point cloud data, and fuse and convert the point cloud data on a grid map to obtain an initial coordinate point set;

[0012] A coarse filtering module configured to coarse filter the initial coordinate point set by using an adaptive filtering algorithm to obtain a first screened coordinate point set, and dynamically obtain a wake region according to the first screened coordinate point set;

[0013] A fine filtering module configured to fine filter the first screened coordinate point set based on the wake region to obtain a second screened coordinate point set;

[0014] A clustering module configured to cluster the second screened coordinate point set by using an adaptive clustering algorithm based on the physical parameters of the laser radar to obtain a detection target.

[0015] According to another aspect of the present application, a terminal device is provided, comprising: one or more processors;

[0016] A storage device configured to store one or more programs,

[0017] When the one or more programs are executed by the one or more processors, the one or more processors implement the method described in any of the embodiments of the present application.

[0018] According to another aspect of the present application, a storage medium of computer executable instructions is provided, and a computer program is stored on the storage medium, which is executed by a processor to implement the method described in any of the embodiments of the present application.

[0019] The technical scheme of the present application improves the comprehensiveness of laser radar detection information by fusing point cloud data, filters out significant clutter and wake area clutter in the process of detecting target motion through rough filtering and fine filtering of the initial coordinate point set obtained after fusion, and aggregates the filtered results based on the physical parameters of the laser radar to complete the detection of each target obstacle, thereby enabling efficient and real-time detection of water surface target obstacles using laser radar in a complex sea surface environment, providing environmental perception support for autonomous navigation of water surface unmanned vehicles.

[0020] It should be understood that the content described in this part is not intended to identify key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become apparent from the following description. BRIEF DESCRIPTION OF DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor on the basis of these drawings.

[0022] Figure 1 is a flow chart of a sea target detection method according to the first embodiment of the present application;

[0023] Figure 2 is a structural schematic diagram of an unmanned vehicle according to the first embodiment of the present application;

[0024] Figure 3 is a schematic diagram of a detected target according to the first embodiment of the present application;

[0025] Figure 4 is a flow chart of a sea target detection method according to the second embodiment of the present application;

[0026] Figure 5 is a structural schematic diagram of a sea target detection device according to the third embodiment of the present application;

[0027] Figure 6 is a structural block diagram of a terminal device according to the present application. DETAILED DESCRIPTION

[0028] In the following, the technical solutions in the embodiments of the present application will be described clearly and completely with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all the other embodiments obtained by a person of ordinary skill in the art without creative effort should belong to the protection scope of the present application.

[0029] It should be noted that the terms "first", "second" and the like in the description and claims of the present application and the above drawings are used to distinguish similar objects, and do not necessarily indicate a specific order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the application described herein can be implemented in an order other than that illustrated or described herein. In addition, "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, device, product or terminal device including a series of steps or units need not be limited to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products or terminal devices.

[0030] Embodiment one

[0031] Figure 1 A flowchart of a data synchronization method provided by the embodiment of the present application. The embodiment can be applied to the case of synchronizing data on a cloud database. The method can be executed by a data synchronization device, which can be realized in the form of hardware and / or software, and the device can be integrated into a terminal device. As shown in the figure, the method comprises the following steps. Figure 1

[0032] In step S101, the surrounding environment of the unmanned ship is scanned by a laser radar to obtain point cloud data, and the point cloud data is fused and converted on a grid map to obtain an initial coordinate point set.

[0033] Optionally, the point cloud data is fused and converted on the grid map to obtain the initial coordinate point set, comprising: converting the point cloud data on the pre-created grid map to obtain coordinate points; fusing the converted multiple frames of coordinate points on the grid map to obtain fused coordinate points, and constructing an initial coordinate point set according to the fused coordinate points, wherein each coordinate point in the initial coordinate point set is marked with height information.

[0034] In the embodiment, the laser radar is used to scan the surrounding environment of the unmanned ship to obtain the point cloud data, and the point cloud data is fused and converted on the grid map to obtain the initial coordinate point set. The laser radar can scan the surrounding environment of the unmanned ship in real time, and the point cloud data obtained by scanning can reflect the real-time changes of the surrounding environment of the unmanned ship. The point cloud data is fused and converted on the grid map to obtain the initial coordinate point set, which can reflect the real-time changes of the surrounding environment of the unmanned ship in a more intuitive and accurate manner. Figure 2 ​A structural schematic diagram of an unmanned ship is shown, and the embodiment adopts a small water unmanned ship as a laser radar experimental carrying platform. The platform adopts a single outboard hanging machine power mode, has good controllability, and is provided with a laser radar in the tail area of the unmanned ship. The specific installation position of the laser radar is not limited in the embodiment. The laser radar can accommodate multiple lasers and has a 360-degree horizontal scanning field of view and a 24-degree vertical scanning field of view, and can provide nearly one million measurement data points per second. Therefore, the surrounding environment of the unmanned ship is scanned in real time by the laser radar installed on the unmanned ship to obtain point cloud data in the embodiment.

[0035] Specifically, the embodiment takes the laser radar as the coordinate origin, the forward direction of the carrier, i.e., the unmanned ship, as the positive direction of the X axis, and the right-hand side perpendicular to the forward direction as the positive direction of the Y axis. According to the range setting of the laser radar, the X axis and the Y axis are truncated at distances A and B, respectively, the grid size is m1, and a grid map is established according to A, B, and m1. After obtaining the point cloud data, the obtained point cloud data is processed on the grid map. Specifically, the three-dimensional point cloud data is converted into two-dimensional coordinate points to realize fast processing of the laser radar point cloud data. For example, the coordinate point of any point p i (x i y i z i ) in the point cloud data is converted into a two-dimensional coordinate point (g x ,g y), and in this embodiment, the height information is also obtained according to the three-dimensional coordinate information of the three-dimensional point cloud data, and the height information is marked as an attribute on each coordinate point. Of course, this embodiment is only an example and does not limit the specific process of two-dimensional conversion. Since the point cloud data detected by the laser radar is relatively sparse under complex sea conditions, the corresponding coordinate points after two-dimensional conversion are also sparse, and the ability of offshore target detection is limited, so in this embodiment, the converted multiple frames of coordinate points are fused on the grid map to obtain fused coordinate points. Considering the speed of the unmanned ship and the speed of the dynamic obstacle, in order to ensure the safety of the autonomous navigation of the surface unmanned ship, it is required that the surface unmanned ship needs to have high real-time performance of offshore obstacle detection, so three frames of coordinate points can be used for fusion, which can ensure the accuracy of obstacle detection and good real-time performance. The fusion process of three frames of coordinate points is to fuse one frame of coordinate points again on the basis of the fusion process of two frames of coordinate points, that is, the results of three laser radar scans are fused for the same target point, so as to avoid the information loss caused by single laser radar scanning. In this embodiment, an initial coordinate point set is constructed according to the fused coordinate points, and each fused coordinate point in the initial coordinate point set is marked with height information, and the height information at this time is the average of the height information of the three coordinate points that are fused. Of course, this embodiment only takes three frames of coordinate points as an example to illustrate the fusion, and does not limit the specific number of frames for fusion. The specific number of frames can be set according to the accuracy requirement of offshore target detection and the computing power of computing resources.

[0036] In step S102, an adaptive filtering algorithm is used to coarsely filter the initial coordinate point set to obtain a first screening coordinate point set, and a wake region is dynamically obtained according to the first screening coordinate point set.

[0037] Optionally, the adaptive filtering algorithm is used to coarsely filter the initial coordinate point set to obtain a first screening coordinate point set, including: extracting coordinate points from the initial coordinate point set according to the height information to obtain an input coordinate point set; taking a specified number of coordinate points with the smallest height information in the input coordinate point set as initial seed points; using a random sample consensus (RANSAC) algorithm to estimate a plane model according to the initial seed points to obtain a filtering plane, and filtering the coordinate points on the filtering plane to obtain the first screening coordinate point set.

[0038] Specifically, after obtaining the initial coordinate point set, the random sample consensus (RANSAC) algorithm is used for rough filtering in the embodiment. Since each coordinate point in the initial coordinate point set is marked with height information, the wake is broken and horizontal, and the obstacle point cloud reflection is stable and has a height. Therefore, in the embodiment, the coordinate points with unknown horizontal height information are extracted, and the extracted coordinate points are used as the input coordinate point set of the RANSAC algorithm, so as to avoid processing of invalid data. In addition, in the embodiment, the initial seed points are dynamically selected according to the height distribution of the coordinate points in the local region. For example, the coordinate points in the input coordinate point set are sorted in descending order according to the height information, and the first three coordinate points in the sequence are selected as the initial seed points. The random sample consensus RANSAC algorithm is used to estimate the plane model according to the initial seed points to obtain a filtering plane, and the coordinate points on the filtering plane are filtered to obtain a first screened coordinate point set. Specifically, the following formula (1) can be used for plane fitting to obtain the filtering plane:

[0039]

[0040] wherein, is an estimated plane normal vector, is an intercept offset, is an arbitrary point on the plane, and the plane normal vector can be solved by calculating the covariance matrix of the coordinate points in the local region, as shown in the following formula (2):

[0041]

[0042] wherein, COV is the covariance matrix of the coordinate points in the local region, is the mean of the seed points, is the number of coordinate points in the input coordinate point set, is the number of coordinate points, is the coordinate point, the covariance matrix describes the distribution characteristics of the input coordinate point set. By solving the matrix to determine three eigenvalues and corresponding eigenvectors, three principal directions of the spatial distribution of the input coordinate point set can be obtained, and the eigenvector corresponding to the smallest eigenvalue is used as the plane normal vector.

[0043] Optionally, the wake region is dynamically obtained according to the first screening coordinate point set, including: detecting, by a sensor, a working parameter associated with the first screening coordinate point set, wherein the working parameter includes a speed, a ship length, a turning angle, a maneuverability turning index, and a maneuverability following index; determining a dynamic wake length, a dynamic wake width, and a dynamic wake radius of curvature according to the working parameter; and dynamically determining the wake region according to the dynamic wake length, the dynamic wake width, and the dynamic wake radius of curvature.

[0044] Specifically, after the first screening coordinate point set is obtained by performing coarse filtering on the filtered surface, although the obvious noise and the noise in the tail wave are roughly filtered, in order to ensure the accuracy of noise removal, the wake region is dynamically obtained according to the first screening coordinate point set in the embodiment, so as to perform subsequent fine filtering according to the wake region. In the embodiment, the dynamic wake length, the dynamic wake width, and the dynamic wake radius of curvature are obtained by using the following formula (3):

[0045]

[0046] wherein, is the dynamic wake length, is the dynamic wake width, and S is the dynamic wake radius of curvature, , and is a proportional coefficient, is a ship speed against the ground, is a ship length, is a turning angle, is a maneuverability turning index, is a maneuverability following index. In the embodiment, the above , , , and are used as working parameters, wherein the ship length is determined according to a distance between a coordinate point at a most front end and a coordinate point at a most end in a region surrounded by the first screening coordinate point set, and the remaining four parameters in the working parameter can be detected by other sensors installed on the unmanned ship, such as a speed sensor. Of course, the embodiment is only an example and does not limit the specific determination method of the working parameter. When the dynamic wake length, the dynamic wake width, and the dynamic wake radius of curvature are known, the range of the wake region can be basically determined.

[0047] In step S103, the first screening coordinate point set is fine filtered based on the wake region to obtain a second screening coordinate point set.

[0048] Optionally, the first screening coordinate point set is finely filtered based on the wake region to obtain a second screening coordinate point set, including: determining a first candidate coordinate point set in the first screening coordinate point set located in the wake region, and a second candidate coordinate point set located outside the wake region; determining a height difference threshold associated with the wake region, and deleting coordinate points with height information less than the height difference threshold in the first candidate coordinate point set to obtain a third candidate coordinate point set; and combining the second candidate coordinate point set and the third candidate coordinate point set to obtain the second screening coordinate point set.

[0049] Optionally, the height difference threshold associated with the wake region is determined, including: obtaining a basic target height difference threshold, wherein the basic target height difference threshold follows the dynamic change of the water surface wave height; obtaining a speed associated with the first screening coordinate point set, and determining the height difference threshold according to the speed and the basic target height difference threshold.

[0050] Specifically, after the wake region is determined, the first screening coordinate point set obtained by the coarse filtering can be finely filtered based on the wake region, and the fine filtering is mainly to accurately remove the remaining wake clutter in the wake region, so as to avoid misjudging the generated wake as an obstacle target during the navigation of the ship. In the embodiment, the first candidate coordinate point set in the first screening coordinate point set located in the wake region and the second candidate coordinate point set located outside the wake region are determined. The coordinate points outside the wake region can be determined as all valid information related to the detection target, so the second candidate coordinate point set is all retained. Most of the first candidate coordinate point set located in the wake region is wake clutter, but there is also a small amount of valid information related to the detection target, so the embodiment needs to screen the valid information from the first candidate coordinate point set.

[0051] In the embodiment, the height difference threshold associated with the wake region is obtained by using the following formula (4) when the first candidate coordinate point set is filtered:

[0052]

[0053] wherein, is the height difference threshold, is a proportionality coefficient, is the speed of the ship relative to the ground, is a basic target height difference threshold, and is a basic target height difference threshold, and The height difference threshold is dynamically adjusted according to the water surface wave height. Since the height of the tail wave from the horizontal plane is usually not too large, after the height difference threshold is dynamically determined based on the water surface wave condition, the coordinate points with height information less than the height difference threshold in the first candidate coordinate point set are deleted to obtain a third candidate coordinate point set. The third candidate coordinate point set contains a small number of effective coordinate points related to the detection target, such as the stern and other structural parts. After filtering the wake area to obtain the third candidate coordinate point set, the second candidate coordinate point set and the third candidate coordinate point set are combined to obtain a second screening coordinate point set. The combination here only refers to the process of combining two point sets into one point set. Therefore, in the present embodiment, when fine filtering the wake area, the wake area is determined in combination with the actual working parameters such as speed and turning angle, thereby ensuring the accuracy of the determination of the wake area. In addition, the height difference threshold associated with the wake area is dynamically adjusted to effectively distinguish the actual wake from the real target obstacle, thereby ensuring accurate filtering of invalid clutter.

[0054] In step S104, the second screening coordinate point set is clustered based on the physical parameters of the laser radar to obtain a detection target.

[0055] Optionally, the second screening coordinate point set is clustered based on the physical parameters of the laser radar to obtain a detection target, including: obtaining a dynamic target distance based on the second screening coordinate point set, determining a dynamic neighborhood radius based on the physical parameters of the laser radar and the dynamic target distance by using a density-based noise application spatial clustering algorithm DBSCAN; calculating a distance-related density of the second screening coordinate point set according to the maximum detection distance of the laser radar and the atmospheric attenuation coefficient; dynamically calculating a minimum point threshold based on the dynamic neighborhood radius and the distance-related density; clustering the second screening coordinate point set based on the minimum point threshold to obtain a clustering cluster, and determining a detection target based on the clustering cluster.

[0056] Specifically, in the present embodiment, an adaptive clustering algorithm with the physical parameters of the laser radar as a constraint is proposed to solve the problem of coordinate point density decay with distance, for example, a density-based noise application spatial clustering algorithm (Density-Based Spatial Clustering of Applications with Noise, DBSCAN). After obtaining the second screening coordinate point set through coarse filtering and fine filtering, a dynamic target distance D is obtained based on the second screening coordinate point set, and a dynamic neighborhood radius is determined based on the physical parameters of the laser radar and the dynamic target distance by using the following formula (5) based on DBSCAN according to the ranging principle:

[0057]

[0058] wherein, is a dynamic neighborhood radius, is a dynamic target distance, is a laser radar divergence angle, is a ranging noise standard deviation, is an acceptable false detection probability, and the above can be obtained according to the standard deviation calculation of the second screening coordinate point set, and can be set in advance according to the detection accuracy. In addition, the embodiment introduces real-time weather data in the DBSCAN algorithm, for example, the atmospheric attenuation coefficient m dynamically updates the distance-related density, and the calculation formula of the distance-related density is shown in the following formula (6):

[0059]

[0060] wherein, is a distance-related density, is a laser radar maximum detection distance, m is an atmospheric attenuation coefficient, is a dynamic target distance, is the total number of coordinate points in the second screening coordinate point set, and m is updated in real time according to the weather conditions. In the case where the dynamic neighborhood radius and the distance-related density are known, the minimum point threshold can be calculated and obtained by using the following formula (7) :

[0061]

[0062] Specifically, after obtaining the minimum point threshold, the second screening coordinate point set can be clustered based on the minimum point threshold to obtain clustering clusters, for example, when the minimum point threshold is determined to be 30, the second screening coordinate point set can be clustered to obtain a plurality of clustering clusters, and each clustering cluster contains 30 coordinate points, and the obtained clustering clusters are taken as detected obstacle targets, as shown in the detection target schematic diagram of the laser radar. Figure 3

[0063] The technical scheme of the embodiment of the application improves the comprehensiveness of the laser radar detection information by fusing the point cloud data, filters the initial coordinate point set obtained after fusion through coarse filtering and fine filtering to filter out significant clutter and wake area clutter in the process of detecting target movement, and aggregates the filtered results based on the physical parameters of the laser radar to complete the detection of each target obstacle, so that the laser radar can be used to efficiently and real-timely detect water surface target obstacles in a complex sea surface environment, and provide environmental perception support for autonomous navigation of a water surface unmanned vehicle.

[0064] Embodiment two

[0065] ​Figure 4 is a flow chart of a sea target detection method provided by an embodiment of the present application, and based on the above-mentioned embodiment, after the adaptive clustering algorithm is used to cluster the second screening coordinate point set based on the physical parameters of the laser radar to obtain the detection target, the embodiment further includes: when it is determined that the weather state meets the preset requirement, image information of the environment around the unmanned ship is captured, and the detection target is corrected according to the image information. As shown in Figure 4 , the method comprises:

[0066] In step S201, the point cloud data is obtained by scanning the environment around the unmanned ship through the laser radar, and the point cloud data is fused and converted on the grid map to obtain the initial coordinate point set.

[0067] Optionally, the point cloud data is fused and converted on the grid map to obtain the initial coordinate point set, which comprises: the point cloud data is two-dimensionally converted on the pre-created grid map to obtain the coordinate points; the fused coordinate points are obtained by fusing the converted multiple frames of coordinate points on the grid map, and the initial coordinate point set is constructed according to the fused coordinate points, wherein each coordinate point in the initial coordinate point set is marked with height information.

[0068] In step S202, the adaptive filtering algorithm is used to coarsely filter the initial coordinate point set to obtain the first screening coordinate point set, and the wake region is dynamically obtained according to the first screening coordinate point set.

[0069] Optionally, the adaptive filtering algorithm is used to coarsely filter the initial coordinate point set to obtain the first screening coordinate point set, which comprises: the input coordinate point set is obtained by extracting the coordinate points from the initial coordinate point set according to the height information; the specified number of coordinate points with the minimum height information in the input coordinate point set are taken as the initial seed points; the plane model is estimated according to the initial seed points by using the random sample consensus (RANSAC) algorithm to obtain the filtering surface, and the coordinate points located on the filtering surface are filtered to obtain the first screening coordinate point set.

[0070] Optionally, the wake region is dynamically obtained according to the first screening coordinate point set, which comprises: the working parameters associated with the first screening coordinate point set are detected by the sensor, wherein the working parameters include speed, ship length, turning angle, maneuverability turning index and maneuverability following index; the dynamic wake length, dynamic wake width and dynamic wake curvature radius are determined according to the working parameters; and the wake region is dynamically determined according to the dynamic wake length, dynamic wake width and dynamic wake curvature radius.

[0071] In step S203, the second screening coordinate point set is obtained by finely filtering the first screening coordinate point set based on the wake region.

[0072] Optionally, the first screening coordinate point set is finely filtered based on the wake region to obtain a second screening coordinate point set, including: determining a first candidate coordinate point set in the first screening coordinate point set located in the wake region, and a second candidate coordinate point set located outside the wake region; determining a height difference threshold associated with the wake region, and deleting coordinate points in the first candidate coordinate point set with height information less than the height difference threshold to obtain a third candidate coordinate point set; and combining the second candidate coordinate point set and the third candidate coordinate point set to obtain the second screening coordinate point set.

[0073] Optionally, the height difference threshold associated with the wake region is determined, including: obtaining a basic target height difference threshold, wherein the basic target height difference threshold changes dynamically with the water surface wave height; obtaining a speed associated with the first screening coordinate point set, and determining the height difference threshold based on the speed and the basic target height difference threshold.

[0074] In step S204, the second screening coordinate point set is clustered based on the physical parameters of the laser radar to obtain a detection target using an adaptive clustering algorithm.

[0075] Optionally, the second screening coordinate point set is clustered based on the physical parameters of the laser radar to obtain a detection target using an adaptive clustering algorithm, including: obtaining a dynamic target distance based on the second screening coordinate point set, determining a dynamic neighborhood radius based on the physical parameters of the laser radar and the dynamic target distance using a density-based noise application spatial clustering algorithm DBSCAN; calculating a distance-related density of the second screening coordinate point set based on a maximum detection distance of the laser radar and an atmospheric attenuation coefficient; dynamically calculating a minimum point threshold based on the dynamic neighborhood radius and the distance-related density; clustering the second screening coordinate point set based on the minimum point threshold to obtain a clustering cluster, and determining the detection target based on the clustering cluster.

[0076] In step S205, when it is determined that the weather state meets the preset requirement, image information of the environment around the unmanned ship is captured, and the detection target is corrected based on the image information.

[0077] Specifically, after detecting the detection target by the laser radar, the unmanned ship can be automatically avoided based on the detection target, but the detection target detected by the laser radar is also corrected to ensure the safety of automatic navigation. When the weather state is good, for example, the visibility is more than 50 meters, the image information of the environment around the unmanned ship can be captured by the camera. Since the air state is good, the captured image information has certain reference value, so the image information is used to check whether the detection target is missed or misdetected in the embodiment. When it is determined that the detection target is missed, the detection target is added based on the image information. When the detection target is misdetected, the misdetected detection target is deleted to realize the accuracy of the detection target displayed on the grid map. In the embodiment, the detection target detected by the laser radar is corrected by the image information with high reliability captured in the good weather state, so as to ensure the accuracy of the finally displayed detection target.

[0078] It should be noted that in the process of correcting the detection target by the image information in the embodiment, the proportion of missed detection and misdetected detection target is also obtained. When the proportion of missed detection and misdetected detection target is too large, it indicates that the laser radar device may have a fault or the software module may have a running fault. At this time, an alarm information is generated and displayed on the man-machine interaction interface, so as to timely prompt the staff to repair and maintain the laser radar or the software module, so as to ensure the accuracy of the marine target detection.

[0079] The technical scheme of the embodiment of the application improves the comprehensiveness of the laser radar detection information by fusing the point cloud data. The initial coordinate point set obtained after fusion is filtered coarsely and finely to filter out significant clutter and wake area clutter during the movement of the detection target. The filtered results are aggregated based on the physical parameters of the laser radar to complete the detection of each target obstacle, so that the laser radar can be used to efficiently and real-timely detect the water surface target obstacle in a complex sea surface environment, and provide environmental perception support for the autonomous navigation of the water surface unmanned ship.

[0080] Embodiment three

[0081] Figure 5 A structure diagram of a marine target detection device provided by the embodiment of the application is shown in FIG. 1. Figure 5 As shown in the figure, the device comprises a fusion conversion module 310, a coarse filtering module 320, a fine filtering module 330 and a clustering module 340.

[0082] The fusion conversion module 310 is configured to scan the environment around the unmanned ship by the laser radar to obtain point cloud data, and fuse and convert the point cloud data on the grid map to obtain an initial coordinate point set.

[0083] a coarse filtering module 320, configured to perform coarse filtering on the initial coordinate point set by using an adaptive filtering algorithm to obtain a first screened coordinate point set, and dynamically obtain a wake region according to the first screened coordinate point set;

[0084] a fine filtering module 330, configured to perform fine filtering on the first screened coordinate point set based on the wake region to obtain a second screened coordinate point set;

[0085] a clustering module 340, configured to perform clustering on the second screened coordinate point set by using an adaptive clustering algorithm based on physical parameters of the laser radar to obtain a detection target.

[0086] Optionally, the fusion conversion module 310 is configured to perform two-dimensional conversion on the point cloud data on a pre-created grid map to obtain coordinate points.

[0087] The converted multiple frames of coordinate points are fused on the grid map to obtain fused coordinate points, and the initial coordinate point set is constructed according to the fused coordinate points, wherein each coordinate point in the initial coordinate point set is marked with height information.

[0088] Optionally, the coarse filtering module 320 includes a coarse filtering subunit, configured to perform coordinate point extraction from the initial coordinate point set according to the height information to obtain an input coordinate point set.

[0089] A specified number of coordinate points with the smallest height information in the input coordinate point set are taken as initial seed points.

[0090] A random sample consensus (RANSAC) algorithm is used to perform plane model estimation according to the initial seed points to obtain a filtering plane, and coordinate points located on the filtering plane are filtered to obtain the first screened coordinate point set.

[0091] Optionally, the coarse filtering module 320 includes a wake region obtaining subunit, configured to detect working parameters associated with the first screened coordinate point set by using a sensor, wherein the working parameters include speed, ship length, turning angle, maneuverability turning index, and maneuverability following index.

[0092] The dynamic wake length, the dynamic wake width, and the dynamic wake curvature radius are determined according to the working parameters.

[0093] The wake region is dynamically determined according to the dynamic wake length, the dynamic wake width, and the dynamic wake curvature radius.

[0094] Optionally, the fine filtering module 330 is configured to determine a first candidate coordinate point set located in the wake region and a second candidate coordinate point set located outside the wake region in the first screened coordinate point set.

[0095] Determine a height difference threshold associated with the wake region, and delete coordinate points in the first candidate coordinate point set whose height information is less than the height difference threshold to obtain a third candidate coordinate point set;

[0096] Combine the second candidate coordinate point set and the third candidate coordinate point set to obtain a second screening coordinate point set.

[0097] Optionally, the fine filtering module 330 is further configured to obtain a basic target height difference threshold, wherein the basic target height difference threshold changes dynamically with the water surface wave height;

[0098] Obtain a speed associated with the first screening coordinate point set, and determine a height difference threshold according to the speed and the basic target height difference threshold.

[0099] Optionally, the clustering module 340 is configured to obtain a dynamic target distance according to the second screening coordinate point set, and determine a dynamic neighborhood radius according to physical parameters of the laser radar and the dynamic target distance by using a density-based spatial clustering algorithm DBSCAN.

[0100] Calculate a distance-related density of the second screening coordinate point set according to a maximum detection distance of the laser radar and an atmospheric attenuation coefficient.

[0101] Dynamically calculate a minimum point number threshold according to the dynamic neighborhood radius and the distance-related density.

[0102] Cluster the second screening coordinate point set based on the minimum point number threshold to obtain a clustering cluster, and determine a detection target according to the clustering cluster.

[0103] The offshore target detection device provided in the embodiments of the present application can perform the offshore target detection method provided in any of the embodiments of the present application, and has the corresponding function modules and beneficial effects of the execution method.

[0104] Embodiment Four

[0105] Figure 6 A structural schematic diagram of a terminal device 10 that can be used to implement embodiments of the present application is shown. The terminal device is intended to represent various forms of digital computers, such as laptops, desktops, tablets, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The terminal device can also represent various forms of mobile devices. The components shown here, their connections and relationships, and their functions, are meant to be examples only, and are not intended to limit implementations of the applications described and / or claimed in this document.

[0106] The components shown here, their connections and relationships, and their functions, are meant to be examples only, and are not intended to limit implementations of the applications described and / or claimed in this document.

[0107] like Figure 6 As shown, the terminal device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer programs stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 can also store various programs and data required for the operation of the terminal device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0108] Multiple components in terminal device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows terminal device 10 to exchange information / data with other terminal devices through computer networks such as the Internet and / or various telecommunications networks.

[0109] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as maritime target detection methods.

[0110] In some embodiments, the maritime target detection method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on terminal device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the maritime target detection method described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to perform the maritime target detection method by any other suitable means (e.g., by means of firmware).

[0111] Various implementations of the above described apparatus and techniques of this document can be realized in digital electronic circuitry, integrated circuitry, a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), a system on a chip (SOC), a complex programmable logic device (CPLD), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include implementation in one or more computer programs that are executable and / or interpretable on a programmable apparatus including at least one programmable processor, which can be special or general purpose, coupled to receive data and instructions from, and to transmit data and instructions to, a storage apparatus, at least one input apparatus, and at least one output apparatus.

[0112] Computer programs used to implement the offshore target detection method of the present application can be written in any combination of one or more programming languages. These computer programs can be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the computer program, when executed, implements the functions / acts specified in the flowcharts and / or block diagrams. The computer program can be executed entirely on a machine, partially on a machine, partially on a machine as part of a standalone software package, or entirely on a remote machine or server.

[0113] In the context of the present application, a computer readable storage medium can be a tangible medium that can contain or store the computer program for use by or in connection with the instruction execution apparatus, device or terminal. The computer readable storage medium can include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or terminal, or any suitable combination of the foregoing. Alternatively, the computer readable storage medium can be a machine readable signal medium. More specific examples of the machine readable storage medium will include one or more lines of electrical connections, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage terminal, a magnetic storage terminal, or any suitable combination of the foregoing.

[0114] To provide for interaction with a user, the devices and techniques described here can be implemented on a terminal device having a display device (e.g., a touch screen) for displaying information to the user and a keyboard, a mouse, or a touch screen by which the user can provide input to the terminal device. Other kinds of devices can be used to provide for interaction with a user as well; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form, including acoustic, speech, or tactile input.

[0115] It should be understood that the steps shown in the various forms above can be reordered, added to, or deleted from without departing from the scope of the application. For example, the steps recited in the application can be performed in parallel, in series, or in a different order, without departing from the desired results of the technical solutions of the application, which are not limited herein.

[0116] The specific implementation described above does not constitute a limitation on the protection scope of the present application. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent replacements, and improvements made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A method for detecting maritime targets, characterized in that, The method includes: The unmanned surface vessel (USV) is scanned by lidar to obtain point cloud data, and the point cloud data is fused and transformed on a grid map to obtain an initial set of coordinate points. An adaptive filtering algorithm is used to coarsely filter the initial coordinate point set to obtain a first filtered coordinate point set, and the wake region is dynamically obtained based on the first filtered coordinate point set. Based on the wake region, the first set of filtered coordinate points is finely filtered to obtain the second set of filtered coordinate points; Based on the physical parameters of the lidar, an adaptive clustering algorithm is used to cluster the second set of selected coordinate points to obtain the detection target; The step of dynamically acquiring the wake region based on the first set of selected coordinate points includes: detecting operating parameters associated with the first set of selected coordinate points using sensors, wherein the operating parameters include speed, ship length, turning angle, maneuverability turning index, and maneuverability following index; determining the dynamic wake length, dynamic wake width, and dynamic wake curvature radius based on the operating parameters; and dynamically determining the wake region based on the dynamic wake length, the dynamic wake width, and the dynamic wake curvature radius. The step of refining the first set of selected coordinate points based on the wake region to obtain a second set of selected coordinate points includes: determining a first set of candidate coordinate points located within the wake region and a second set of candidate coordinate points located outside the wake region; determining a height difference threshold associated with the wake region and deleting coordinate points in the first set of candidate coordinate points whose height information is less than the height difference threshold to obtain a third set of candidate coordinate points; and combining the second set of candidate coordinate points and the third set of candidate coordinate points to obtain the second set of selected coordinate points.

2. The method according to claim 1, characterized in that, The step of fusing and transforming the point cloud data on the grid map to obtain an initial set of coordinate points includes: The point cloud data is transformed into two dimensions on a pre-created grid map to obtain coordinate points; The converted multi-frame coordinate points are fused on the grid map to obtain fused coordinate points, and the initial coordinate point set is constructed based on the fused coordinate points, wherein each coordinate point in the initial coordinate point set is marked with height information.

3. The method according to claim 2, characterized in that, The step of using an adaptive filtering algorithm to coarsely filter the initial set of coordinate points to obtain the first set of filtered coordinate points includes: Based on the height information, the input coordinate point set is obtained by extracting coordinate points from the initial coordinate point set; Use a specified number of coordinate points with the smallest height information in the input coordinate point set as initial seed points; The Random Sampling Consensus (RANSAC) algorithm is used to estimate the plane model based on the initial seed points to obtain the filter surface. The coordinate points located on the filter surface are then filtered to obtain the first set of filtered coordinate points.

4. The method according to claim 1, characterized in that, The determination of the height difference threshold associated with the wake region includes: Obtain a basic target height difference threshold, wherein the basic target height difference threshold dynamically changes with the water surface wave height; Obtain the velocity associated with the first set of filtered coordinate points, and determine the height difference threshold based on the velocity and the basic target height difference threshold.

5. The method according to claim 1, characterized in that, The physical parameters based on lidar are used to cluster the second set of selected coordinate points using an adaptive clustering algorithm to obtain the detection target, including: The dynamic target distance is obtained based on the second set of selected coordinate points, and the density-based noise application spatial clustering algorithm DBSCAN is used to determine the dynamic neighborhood radius based on the physical parameters of the lidar and the dynamic target distance. The distance correlation density of the second set of selected coordinate points is calculated based on the maximum detection range of the lidar and the atmospheric attenuation coefficient. The minimum number of points threshold is obtained by dynamically calculating based on the dynamic neighborhood radius and the distance correlation density. Clustering is performed on the second set of selected coordinate points based on the minimum number of points threshold to obtain clusters, and the detection target is determined based on the clusters.

6. A target detection device, characterized in that, The device includes: The fusion and conversion module is used to scan the surrounding environment of the unmanned surface vessel with lidar to obtain point cloud data, and to fuse and convert the point cloud data on the grid map to obtain an initial coordinate point set. The coarse filtering module is used to perform coarse filtering on the initial coordinate point set using an adaptive filtering algorithm to obtain a first filtered coordinate point set, and to dynamically obtain the wake region based on the first filtered coordinate point set. The fine filtering module is used to perform fine filtering on the first set of filtered coordinate points based on the wake region to obtain a second set of filtered coordinate points; The clustering module is used to cluster the second set of selected coordinate points based on the physical parameters of the lidar using an adaptive clustering algorithm to obtain the detection target; The coarse filtering module is further configured to detect operating parameters associated with the first set of filtered coordinate points via sensors, wherein the operating parameters include speed, ship length, turning angle, maneuverability turning index, and maneuverability following index; determine the dynamic wake length, dynamic wake width, and dynamic wake curvature radius based on the operating parameters; and dynamically determine the wake region based on the dynamic wake length, the dynamic wake width, and the dynamic wake curvature radius. The fine filtering module is further configured to determine a first candidate coordinate point set located within the wake region and a second candidate coordinate point set located outside the wake region from the first set of filtered coordinate points; determine a height difference threshold associated with the wake region; delete coordinate points in the first set of candidate coordinate points whose height information is less than the height difference threshold to obtain a third candidate coordinate point set; and combine the second candidate coordinate point set and the third candidate coordinate point set to obtain the second set of filtered coordinate points.

7. A terminal device, characterized in that, The terminal device includes: One or more processors; Storage device for storing one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the method as described in any one of claims 1-5.

8. A storage medium for computer-executable instructions, wherein a computer program is stored thereon, characterized in that, When the program is executed by the processor, it implements the method as described in any one of claims 1-5.

Images