Water surface floating object detection method, electronic device and storage medium

Abstract

The present application discloses a method for detecting floating objects on the water surface, an electronic device, and a storage medium. The method for detecting floating objects on the water surface includes: clustering radar data collected by a radar device to obtain a target cluster, wherein the radar data records relevant information of several target objects on the water surface; determining the characteristics of the target cluster based on the relevant information of each target object in the target cluster; determining whether the characteristics of the target cluster are within the convex hull of the floating object, wherein the convex hull of the floating object includes several target points, each target point is used to represent the characteristics of a floating object cluster, and each floating object cluster is determined by using radar data about floating objects collected on the water surface at multiple times; in response to the characteristics of the target cluster being within the convex hull of the floating object, determining that several target objects are floating objects on the water surface. The above scheme can improve the accuracy of floating object detection.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a method for detecting floating objects on a water surface, an electronic device, and a storage medium. Background Art

[0002] Floating objects / garbage on the water surface seriously affect the water quality and ecology. In urban water networks, canals, tourist attractions and other places close to living areas, floating objects are often found on the water surface in large areas due to domestic garbage, vegetation garbage, mining garbage and aquatic plants. In the hot summer, if they are not cleaned up in time, they are likely to cause serious problems such as river siltation, river overflow, stagnant water and stench.

[0003] Currently, it is mainly determined by obtaining the flow velocity of each object on the water surface. For example, if the flow velocity of the object is close to the flow velocity of the water, each object is determined to be a floating object. However, this method of determining whether an object is a floating object based solely on flow velocity is likely to cause false detection. Summary of the invention

[0004] The present application at least provides a method for detecting floating objects on the water surface, an electronic device, and a storage medium.

[0005] The present application provides a method for detecting floating objects on the water surface, comprising: clustering radar data collected by a radar device to obtain a target clustering cluster, wherein the radar data records relevant information of a number of target objects on the water surface; determining the characteristics of the target clustering cluster based on the relevant information of each target object in the target clustering cluster; determining whether the characteristics of the target clustering cluster are within a convex hull of the floating object, wherein the convex hull of the floating object includes a number of target points, each target point being used to represent the characteristics of a floating object clustering cluster, and each floating object clustering cluster is determined by using radar data on floating objects collected on the water surface at multiple moments; in response to the characteristics of the target clustering cluster being within the convex hull of the floating object, determining that the number of target objects are floating objects on the water surface.

[0006] The present application provides a device for detecting floating objects on a water surface, comprising: a clustering module, a feature determination module, a judgment module and a floating object determination module; the clustering module is used to cluster radar data collected by a radar device to obtain a target clustering cluster, wherein the radar data records relevant information of a number of target objects on the water surface; the feature determination module is used to determine the features of the target clustering cluster based on the relevant information of each target object in the target clustering cluster; the judgment module is used to determine whether the features of the target clustering cluster are within a convex hull of a floating object, wherein the convex hull of the floating object includes a number of target points, each of which is used to represent the features of a floating object clustering cluster, and each floating object clustering cluster is determined by using radar data on floating objects collected on the water surface at multiple moments; the floating object determination module is used to determine that a number of target objects are floating objects on the water surface in response to the fact that the features of the target clustering cluster are within the convex hull of the floating object.

[0007] The present application provides an electronic device, including a memory and a processor, wherein the processor is used to execute program instructions stored in the memory to implement the above-mentioned method for detecting floating objects on the water surface.

[0008] The present application provides a computer-readable storage medium on which program instructions are stored. When the program instructions are executed by a processor, the above-mentioned method for detecting floating objects on the water surface is implemented.

[0009] The above scheme obtains floating object clusters related to floating objects by clustering radar data about floating objects collected on the water surface, and determines the floating object convex hull based on the characteristics of multiple floating object clusters, so that after the radar device collects radar data on the water surface, the target cluster can be determined according to the collected radar data. If the characteristics of the target cluster are within the convex hull of the floating object, it is determined that the radar data collected by the radar device is from the floating object, and it can be determined that the target object on the water surface is a floating object. Compared with simply determining whether the target object is a floating object based on the degree of proximity between the target object's moving speed and the flow velocity of the water flow, this scheme has more accurate detection results of floating objects on the water surface.

[0010] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] The drawings herein are incorporated into the specification and constitute a part of the specification. These drawings illustrate embodiments consistent with the present application and are used together with the specification to illustrate the technical solution of the present application.

[0012] Figure 1 This is a flow chart of an embodiment of a method for detecting floating objects on a water surface of the present application;

[0013] Figure 2 yes Figure 1 Schematic diagram of the sub-process of step S11;

[0014] Figure 3 yes Figure 1 Schematic diagram of the sub-process of step S13;

[0015] Figure 4 This is another schematic diagram of the process of an embodiment of the method for detecting floating objects on the water surface of the present application;

[0016] Figure 5 It is a schematic diagram of a convex hull of a floating object shown in an embodiment of a method for detecting floating objects on a water surface of the present application;

[0017] Figure 6 This is another flow chart of an embodiment of the method for detecting floating objects on the water surface of the present application;

[0018] Figure 7 It is a structural schematic diagram of an embodiment of a water surface floating object detection device of the present application;

[0019] Figure 8 It is a structural schematic diagram of an embodiment of the electronic device of the present application;

[0020] Fig. 9 It is a structural diagram of an embodiment of a computer-readable storage medium of the present application. DETAILED DESCRIPTION

[0021] The scheme of the embodiment of the present application is described in detail below in conjunction with the drawings of the specification.

[0022] In the following description, for the purpose of explanation rather than limitation, specific details such as specific system structures, interfaces, and technologies are provided to facilitate a thorough understanding of the present application.

[0023] The term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there may be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the objects associated before and after are in an "or" relationship. In addition, "many" in this article means two or more than two. In addition, the term "at least one" in this article means any combination of at least two of any one or more of a plurality of, for example, including at least one of A, B, and C, can mean including any one or more elements selected from the set consisting of A, B, and C.

[0024] Considering that radar has the advantage of being able to work all day and all night, regardless of wind or rain, day or night, the present solution is more applicable to determine whether the target on the water surface is a floating object through radar data obtained by the radar. Optionally, the radar device described in the present solution may be a millimeter wave radar, which has a high detection probability and high frame rate to obtain measurement information of moving targets.

[0025] In addition, when the size and number of floating objects are similar, the data characteristics of floating objects on the water surface are similar. The characteristics of floating objects on the water surface last year and the corresponding radar data characteristics of floating objects on the water surface this year are highly similar; the radar data characteristics corresponding to floating objects in different sections of the same river are also highly similar; the corresponding radar data characteristics of different rivers are also highly similar; the radar data characteristics corresponding to floating objects at different times and in different weather conditions at the same location are also highly similar. The radar data characteristics corresponding to the same floating object flowing through the same river and collected by different radar devices in different river sections are also highly similar.

[0026] In some application scenarios, the method for detecting floating objects on the water surface provided by the present application can be applied to the computer field, etc. The execution subject for implementing the method for detecting floating objects on the water surface described in the present application can be a floating object detection device on the water surface, an electronic device, etc. For example, the execution subject can be a device that establishes a connection with a radar device, or if the radar device has data processing capabilities, the execution subject can be a radar device. For example, the floating object detection device on the water surface can be set in a terminal device or a server or other processing device, wherein the terminal device can be an audio device, an electronic device, a device (User Equipment, UE), a handheld device, a computing device, a vehicle-mounted device, a wearable device, etc. In some possible implementations, the method for detecting floating objects on the water surface can be implemented by a processor calling a computer-readable instruction stored in a memory.

[0027] See also Figure 1 , Figure 1 This is a flow chart of an embodiment of the method for detecting floating objects on the water surface of the present application. Figure 1 As shown, the method for detecting floating objects on the water surface provided by the embodiment of the present disclosure may include the following steps:

[0028] Step S11: clustering the radar data collected by the radar device to obtain target clusters.

[0029] Radar data may be radar point cloud data. Radar data records relevant information of several target objects on the water surface. Some of the collected target objects may be floating objects on the water surface, such as domestic garbage, vegetation garbage, mining garbage, and aquatic plants, etc. Some target objects may also be ships or objects playing on the water surface, etc. The relevant information of the target object may include but is not limited to the distance, angle, speed (RadialSpeed), radar reflection area (RCS), etc. of the target object relative to the radar device.

[0030] Several may be one or more. Exemplarily, the radar data collected by the radar device each time may include relevant information of multiple target objects. The radar data collected by the radar device may be radar data collected once or radar data collected multiple times. Exemplarily, if the radar data is radar data collected by the radar device at a single time, the target clustering cluster may be obtained by clustering the radar data collected once; if the radar data is radar data collected by the radar device multiple times, the target clustering cluster may include multiple, that is, radar data collected at different times are clustered independently to obtain multiple target clustering clusters; or if the radar data is radar data collected by the radar device multiple times, the number of target clustering clusters may also be one, that is, the target clustering cluster is obtained by clustering the radar data at each time.

[0031] The radar data may be clustered by using the distance between target objects, or the speed between target objects, or the radar reflection area of ​​target objects, or may be clustered by combining two or more information of target objects.

[0032] Step S12: Determine the characteristics of the target cluster based on the relevant information of each target object in the target cluster.

[0033] Optionally, statistics can be performed on the relevant information of each target object in the target cluster to obtain the characteristics of the target cluster. Optionally, if the relevant information of the target object can include multiple ones, statistics can be performed on one or more relevant information to obtain the characteristics of the target cluster. In some application scenarios, the speed of each target object in the cluster can be counted to obtain the characteristics of the target cluster. In some application scenarios, the distance between each target object in the target cluster and the radar device can be counted to obtain the characteristics of the target cluster. In some application scenarios, the radar reflection area of ​​each target object in the target cluster can be counted to obtain the characteristics of the target cluster.

[0034] Step S13: Determine whether the features of the target cluster are within the convex hull of the floating object.

[0035] The floating object convex hull includes a number of target points, each of which is used to represent the characteristics of a floating object cluster, and each floating object cluster is determined by using radar data about floating objects collected on the water surface at multiple times. That is, the radar device can first collect radar data of floating objects at multiple times, and then the radar data at each time determines a cluster and the characteristics of each cluster constitute a target point in the floating object convex hull. The radar device that collects radar data of floating objects and the radar device in step S11 can be the same radar device or a different radar device.

[0036] Exemplarily, the convex hull of floating objects can be considered as a polygon formed by multiple floating object cluster features in space, and the feature of the target cluster can be considered as a point in the space. If the target cluster is within the polygon, it is determined that the feature of the target cluster is within the convex hull of the floating object.

[0037] Step S14: In response to the fact that the features of the target cluster are within the convex hull of the floating objects, determining that the plurality of target objects are floating objects on the water surface.

[0038] The characteristics of the target cluster clusters are within the convex hull of the floating object, including the characteristics of the target cluster clusters being within and on the boundary of the convex hull of the floating object. In some application scenarios, if the radar data obtained in step S11 is radar data collected from the water surface at multiple times, the corresponding convex hulls can be determined for the characteristics of the multiple target cluster clusters first, and then it is determined whether the convex hull is within the convex hull of the floating object. If so, it means that the target object is a floating object on the water surface. In some application scenarios, if the radar data obtained in step S11 is radar data collected from the water surface at multiple times, it can be determined in turn whether the characteristics of each target cluster cluster are within the convex hull of the floating object. If the target cluster cluster at a certain moment is within the convex hull of the floating object, it is determined that the target object collected at this moment is a floating object. If the target cluster cluster at a certain moment is not within the convex hull of the floating object, it is determined that the target object collected at this moment is not a floating object, or only when all the target cluster clusters are within the convex hull of the floating object, it is determined that the target objects collected at multiple moments are all floating objects.

[0039] Step S15: In response to the fact that the features of the target cluster are outside the convex hull of the floating objects, it is determined that the target objects are not floating objects on the water surface.

[0040] The fact that the characteristics of the target cluster are outside the convex hull of the floating object can be considered to mean that there is a certain distance between the characteristics of the target cluster and the boundary of the convex hull of the floating object.

[0041] The above scheme obtains floating object clusters related to floating objects by clustering radar data about floating objects collected on the water surface, and determines the floating object convex hull based on the characteristics of multiple floating object clusters, so that after the radar device collects radar data on the water surface, the target cluster can be determined according to the collected radar data. If the characteristics of the target cluster are within the convex hull of the floating object, it is determined that the radar data collected by the radar device is from the floating object, and it can be determined that the target object on the water surface is a floating object. Compared with simply determining whether the target object is a floating object based on the degree of proximity between the target object's moving speed and the flow velocity of the water flow, this scheme has more accurate detection results of floating objects on the water surface.

[0042] In some application scenarios, the structure of radar data can refer to Table 1, which is a schematic table of radar data. The first row in Table 1 represents the frame number, the number of measurements in the frame, and the measurement time of the frame, respectively. The data in Table 1 is only an example. The number of measurements can be the number of target objects measured. After the first row is the specific radar measurement point, the first column is the measurement ID (the ID of each target object), the 2nd and 3rd columns are the X and Y coordinates, the unit can be meters, the 4th column is the intensity information (RCS), the unit can be db, the intensity unit, and the 5th column is the speed (RadialSpeed), the unit can be m / s.

[0043] Table 1: Radar data diagram

[0044]

[0045] In some embodiments, the relevant information of the target objects includes speed information. Considering that the floating objects on the water surface are generally in the same direction as the water flow, if the moving direction of the target object is opposite to the water flow direction, the target object is likely to be a ship rather than a floating object. Optionally, before executing the above step S11, the following steps may be performed:

[0046] Determine the direction of water flow on the water surface and determine the moving direction of each target object based on the speed information of each target object. Then, remove the radar data with different directions from the water flow. Step S11 may be clustering the retained radar data to obtain target clusters.

[0047] The water flow direction can be preset, or the data collected by the radar device can be analyzed to determine the water flow direction. Optionally, the radar data can be grouped according to the speed of the target object (column 5 from row 2 to row 7 in Table 1), divided into target measurement point cloud data of the destination and target measurement point cloud data of the incoming direction. The water flow direction may be the incoming direction or the outgoing direction. According to the water flow direction, the radar data with positive or negative speed is retained, or the radar data with a speed of 0 can also be retained. By removing the radar data with different directions from the water flow, the disturbance of weeds and trees on the water bank and the influence of retrograde ships can be filtered out.

[0048] For floating objects, their speed is basically determined by the water flow. That is, the speed of floating objects is in a fixed ratio to the speed of water flow. When there is a cruise ship or a cleaning ship, the movement of the ship will cause waves in the water, and the speed of these waves is much greater than the speed of water flow. Therefore, setting a reasonable speed threshold can overcome the interference of water waves. Considering that the speed of floating objects is similar to the speed of water flow, if the moving speed of the target object is too fast, it is likely to be a ship. Before executing the above step S11, the following steps can also be performed:

[0049] Remove radar data whose speed is greater than or equal to a preset speed, wherein the preset speed is greater than the water flow velocity of the water surface. The preset speed and the water flow velocity can be preset or obtained by analyzing the radar data. Step S11 can be clustering the retained radar data to obtain a target cluster.

[0050] In some embodiments, the collected radar data may first be subjected to the step of removing radar data having a direction different from that of the water flow, and then the radar data having a speed greater than or equal to a preset speed may be removed; or the collected radar data may first be subjected to the step of removing radar data having a speed greater than or equal to a preset speed, and then the step of removing radar data having a direction different from that of the water flow may be performed.

[0051] In some embodiments, the relevant information of the target object includes position information, speed information and / or strength information. The strength information may be the radar reflection area mentioned above. The position information may include the distance and angle from the radar device. Figure 2 , the above step S11 may include the following steps:

[0052] Step S111: based on the position information, speed information and / or strength information of each target object, determine the position relationship, speed relationship and / or strength relationship between the target objects.

[0053] In some application scenarios, the position relationship can be determined based on the position information of each target object. In some application scenarios, the speed relationship can be determined based on the speed information of each target object. In some application scenarios, the intensity relationship can be determined based on the intensity information of each target object. In some application scenarios, the corresponding relationship can be determined based on two or three of the position information, speed information and intensity information. This embodiment takes the determination of the position relationship, speed relationship and intensity relationship between each target object based on the position information, speed information and intensity information of each target object as an example.

[0054] A coordinate system can be established with any point on the water surface as the origin. The distance and angle between each target object and the radar device can determine the horizontal and vertical coordinates of each target object in the coordinate system. The speed relationship includes the speed difference between any two target objects, and the position relationship includes the distance between any two target objects, which can be an "elliptical" distance. The intensity information relationship can be the difference in intensity information. For example, the position relationship is obtained. The method can refer to formula (1):

[0055] = Formula (1);

[0056] in, Represents a target object, whose position is ( , ), represents a target object in the target cluster k, and its location is ( , ), represents the “ellipse” distance between the two target objects, that is, the positional relationship between the two target objects. Represents a distance parameter, which can be preset.

[0057] Get speed relationship The method can refer to formula (2):

[0058] = Formula (2);

[0059] in, represents a target object whose velocity is , represents a target object in the target cluster k, and its speed is .

[0060] Get the intensity relationship The method can be referred to formula (3):

[0061] = Formula (3);

[0062] in, Represents a target object, whose intensity information is , represents a target object in the target cluster k, and its intensity information is .

[0063] Step S112: clustering the collected radar data based on the position relationships, speed relationships and / or intensity relationships to obtain target clusters.

[0064] This embodiment takes the method of clustering the collected radar data by combining various position relationships, speed relationships and intensity relationships to obtain target clusters as an example. The method of clustering the collected radar data by combining various position relationships, speed relationships and intensity relationships to obtain target clusters can refer to formula (4):

[0065] , , Formula (4);

[0066] in, is the distance threshold, and Respectively represent the speed difference threshold and the intensity difference threshold, which are The spatial position / wind speed in the environment is related. k represents the kth moment, represents the kth target cluster, That is, the target cluster is obtained by clustering the radar data collected at the kth time. In other words, if the target object When the position relationship, speed relationship and intensity relationship with any target object in the target cluster all meet the conditions, it is determined that the target object is in the target cluster. The target object being in the target cluster may mean that the radar data corresponding to the target object is in the target cluster.

[0067] Optionally, the target cluster includes relevant information of multiple target objects, and the above step S111 may include the following steps: determining clustering accuracy, where the number of target objects included in the clusters obtained with different clustering accuracies is different. Adjusting the method for obtaining the positional relationship according to the clustering accuracy. Determining the positional relationship between the target objects based on the adjusted method for obtaining the positional relationship and the position information of the target objects.

[0068] For the target cluster For example, the number of target objects contained in the target cluster can be set to be greater than or equal to a preset number, and the preset number can be understood as a target number threshold that satisfies the clustering. Greater than or equal to , Representation clustering The number of targets in Indicates the target number threshold for clustering.

[0069] Considering that the strictness of clustering may be different in different scenarios, clustering level 1, level 2, and level 3 can be pre-set. Different levels have different clustering accuracy. Before application, one level can be selected from multiple levels as the clustering accuracy. The stricter the clustering, the different the number of target objects in the final cluster. As mentioned above, the method for obtaining the position relationship is as shown in the above formula (1). In different methods for obtaining the position relationship, Can be different, you can pre-set the corresponding clustering accuracy . Optionally, The size of is positively correlated with the clustering accuracy.

[0070] In some other embodiments, the above step S111 may include the following steps: determining clustering conditions according to clustering accuracy, the clustering conditions including position relationship conditions, speed relationship conditions and / or strength relationship conditions. Based on the clustering conditions, position relationship, speed relationship and strength relationship, a target cluster is obtained.

[0071] Exemplarily, the speed difference threshold, distance threshold, and intensity difference threshold may be adjusted according to clustering accuracy.

[0072] In some embodiments, the relevant information of the target object includes position information, speed information and / or intensity information, and the characteristics of the target cluster include average speed information, average intensity information, intensity variance and / or the area of ​​the target convex hull. That is, the relevant information of the target object may include one or more of the position information, speed information and intensity information, and the characteristics of the target cluster include one or more of the average speed information, average intensity information, intensity variance and the area of ​​the target convex hull. In this embodiment, the characteristics of the target cluster include average speed information, average intensity information, intensity variance and the area of ​​the target convex hull as an example. Among them, the target convex hull area of ​​the target cluster can be determined based on the position of each target object in the target cluster. Exemplarily, the position of the outermost target object in the target cluster is surrounded to form a target convex hull, and the enclosed area is the area of ​​the target convex hull.

[0073] For example, the characteristics of the target cluster are = ,in, , , as well as The calculation method of can refer to formula (5) to formula (8):

[0074] = Formula (5);

[0075] = Formula (6);

[0076] = Formula (7);

[0077] = Formula (8);

[0078] in Represents the target cluster The average speed information in Represents the target cluster The average intensity information in Represents the target cluster The intensity variance in Represents the target cluster The target convex hull area.

[0079] See also Figure 3 , the above step S13 may include the following steps:

[0080] Step S131: Determine the boundary target point in the convex hull of the floating object.

[0081] Each boundary target point is located on the boundary of the floating object convex hull. For example, the boundary target point set can be determined as = , among which , ,… , are the boundary target points on the convex hull of the floating objects, and each boundary target point is a feature of a floating object cluster.

[0082] Step S132: In response to the existence of a set of target data sets satisfying the preset constraint condition, it is determined that the characteristics of the target cluster are within the convex hull of the floating object.

[0083] Among them, the preset constraints include: the sum of the products of each target data in the target data group and each boundary target point is the characteristic of the target cluster, the sum of each target data in the target data group is 1, and each target data in the target data group is greater than or equal to 0 and less than or equal to 1.

[0084] Target data set = For example, , ,… Respectively represent each target data in the target data group. Please refer to formula (9) and formula (10):

[0085] = * Formula (9);

[0086] = Formula (10);

[0087] represents the characteristics of the target cluster, Represents the matrix formed by each boundary target point.

[0088] If the target data set also satisfies formula (11) and formula (12) at the same time, it is determined that the characteristics of the target cluster are within the convex hull of the floating object:

[0089] Formula (11);

[0090] =1 formula (12);

[0091] Among them, formula (11) means that each target data is greater than or equal to 0 and less than or equal to 1, and formula (12) means that the sum of each target data is 1.

[0092] That is, there exists a set of target data sets that can satisfy formula (9), formula (11) and formula (12), and it is determined that the characteristics of the target cluster are within the convex hull of the floating object.

[0093] Optionally, see Figure 4 , the method may further comprise the following steps:

[0094] Step S21: Acquire radar data on floating objects on the water surface collected by the radar device at multiple times.

[0095] The radar device may be the same as or different from the radar device in step S11. For example, the radar device here may be a millimeter wave radar.

[0096] Step S22: respectively obtaining the characteristics of the floating object clusters corresponding to the radar data about the floating objects at each moment.

[0097] Each moment corresponds to a floating object cluster. For each moment, the method for obtaining the floating object cluster at that moment can refer to the method for obtaining the target cluster, which will not be described here. The composition of the features of the floating object cluster can be the same as the composition of the features of the target cluster, for example, it can include average speed information, average intensity information, intensity variance and the area of ​​the target convex hull in the floating object cluster.

[0098] Step S23: taking the features of each floating object cluster as a target point in a space of preset dimensions.

[0099] The dimension of the space is related to the dimension of the features of the floating object clusters. The dimension of the features of the floating object clusters can be the same as the dimension of the features of the target clusters. The dimension of the space is greater than or equal to the dimension of the features of the clusters, such as the features of the target clusters. = ,It can be seen that the characteristics of the clustering cluster can be four-dimensional characteristics, and the preset dimension can be four-dimensional.

[0100] Step S24: The convex hull formed by each target point is used as the convex hull of the floating object.

[0101] For example, the convex hull of the floating object can be determined by enclosing the largest area of ​​the horse race. Because the multi-dimensional convex hull is difficult to describe using drawings, this embodiment uses a two-dimensional drawing as an example of the convex hull of the floating object. For details, please refer to Figure 5 , Figure 5 The floating object convex hull shown in includes 10 target points, and each target point connected by a boundary line is a boundary target point.

[0102] In some other embodiments, the method may further include the following steps:

[0103] In response to the fact that the characteristics of the target clusters are within the convex hull of the ship, several target objects are determined to be ships, and the convex hull of the ship is determined by using radar data of the ship collected on the water surface at multiple times.

[0104] Among them, the ship convex hull is obtained by clustering the characteristics of the ship clusters obtained by clustering the radar data collected from the ships on the water surface at multiple times. The specific acquisition method can refer to the above-mentioned method of obtaining the convex hull of floating objects, which will not be repeated here. The method for determining whether the characteristics of the target cluster are within the convex hull of the ship can refer to the above-mentioned method of determining whether the characteristics of the target cluster are within the convex hull of floating objects, which will not be repeated here.

[0105] In some other embodiments, the method may further include the following steps:

[0106] The target convex hull area of ​​the target cluster is obtained, and the target convex hull area of ​​the target cluster is determined based on the position of each target object in the cluster. The method of obtaining the target convex hull area of ​​the target cluster can be referred to above, and will not be repeated here.

[0107] In response to the target convex hull area of ​​the target cluster being greater than or equal to the preset area, prompt information is sent, and the target convex hull area of ​​the target cluster is determined based on the position of each target object in the cluster.

[0108] In some embodiments, in response to the target convex hull area of ​​the target cluster being smaller than a preset area, it is determined whether the sum of multiple historical convex hull areas and the target convex hull area of ​​the target cluster is greater than or equal to a preset area, where the historical convex hull area is the target convex hull area of ​​the target cluster determined based on radar data collected from the water surface at multiple historical moments; in response to the sum of multiple historical convex hull areas and the target convex hull area of ​​the target cluster being greater than or equal to the preset area, a prompt message is sent.

[0109] In some embodiments, the method for detecting floating objects on the water surface may further include the following steps:

[0110] First, install the equipment, such as installing the radar device on a stable rack, facing the water surface area where floating objects need to be detected. Then adjust the parameters, such as setting the working time, drawing the river area, and detecting sensitivity. After the preparation work is done, data collection is carried out. The radar device needs to operate normally and continuously collect radar data. Then you can selectively perform the annotation step, that is, annotate parameters such as clustering accuracy. These parameters can use the default values ​​(radar manufacturers do a lot of tests on floating objects on the water surface, and preset data and general parameters in the algorithm configuration as default configuration items). Or due to the width of the river, the flow rate of the river water, and the type of floating objects, the preset default parameters will also change, and manual annotation (correction) can be performed. Then the content of floating objects is detected. If the area of ​​the floating objects is greater than or equal to the preset area, it can be determined that the content of the floating objects exceeds the threshold. If it exceeds the threshold, a prompt message is issued. If it does not exceed the threshold, the radar data is continued to be collected. In addition, a review can be performed to determine whether the prompt message is sent by mistake. In addition, the floating object detection situation can be uploaded to the cloud and shared with other systems or other radar equipment systems upstream and downstream of the river.

[0111] For some examples, see Figure 6 The method for detecting floating objects on the water surface may further include the following steps:

[0112] Step S31: Initialization.

[0113] The initialization content includes: target clustering parameters, cumulative data frame number, speed threshold, water flow direction, floating object accumulation reporting threshold, setting river area, etc.

[0114] Step S32: The radar device senses the environment.

[0115] Exemplarily, a radar target detection algorithm is used to measure the environment and output environmental point cloud data in real time.

[0116] Step S33: Obtain radar point cloud data.

[0117] Only the measured point cloud data within the river area is retained.

[0118] Step S34: filter out reverse water flow targets.

[0119] The radar point cloud data are grouped according to the direction of departure and arrival, and the radar point cloud data in the opposite direction of the water flow are filtered out.

[0120] Step S35: Filter out targets that exceed a speed threshold.

[0121] Filter out radar point cloud data that exceeds the speed threshold.

[0122] Step S36: Clustering.

[0123] The radar point cloud data is clustered to obtain the target clusters. The specific clustering method can be referred to above and will not be described here.

[0124] Step S37: Find the eigenvector.

[0125] Get the feature vector of the target cluster The method can be referred to above and will not be described here.

[0126] Step S38: Is it a ship?

[0127] If the characteristic vector of the target cluster is within the convex hull of the ship, each target object is determined to be a ship. If the result of step S38 is yes, step S32 is executed, otherwise step S39 is executed.

[0128] Step S39: Is there any floating object?

[0129] If the characteristic vector of the target cluster is within the convex hull of the floating object, each target object is determined to be a floating object. If the execution result of step S39 is yes, step S40 is executed, otherwise step S32 is executed.

[0130] Step S40: extracting the area of ​​the convex hull.

[0131] The method of obtaining the target convex hull area of ​​the target cluster can be referred to above and will not be described in detail here.

[0132] Step S41: accumulate for a period of time.

[0133] That is, the target convex hull area at a single moment may not reach the reporting threshold (preset area), and the target convex hull area over a period of time can be accumulated.

[0134] Step S42: Accumulated amount of floating objects.

[0135] The area of ​​all floating objects over a period of time is accumulated, or the accumulated amount of floating objects can be corrected based on a correction factor.

[0136] Step S43: Whether the reporting threshold is reached.

[0137] Determine whether the floating objects exceed the reporting threshold. If so, execute step S44, otherwise execute step S32.

[0138] Step S44: Report.

[0139] The location and content of floating objects can be reported to the preset recipient, so that the preset recipient can clean up the floating objects on the water surface in time.

[0140] See also Figure 7 , Figure 7It is a structural schematic diagram of an embodiment of the water surface floating object detection device of the present application. The water surface floating object detection device 40 can execute the above-mentioned water surface floating object detection method. The water surface floating object detection device 40 includes a clustering module 41, a feature determination module 42, a judgment module 43 and a floating object determination module 44; the clustering module 41 is used to cluster the radar data collected by the radar device to obtain a target clustering cluster, and the radar data records the relevant information of several target objects on the water surface; the feature determination module 42 is used to determine the characteristics of the target clustering cluster based on the relevant information of each target object in the target clustering cluster; the judgment module 43 is used to determine whether the characteristics of the target clustering cluster are in the convex hull of the floating object, and the convex hull of the floating object includes several target points, each target point is used to represent the characteristics of a floating object clustering cluster, and each floating object clustering cluster is determined by using the radar data about floating objects collected on the water surface at multiple times; the floating object determination module 44 is used to determine that several target objects are floating objects on the water surface in response to the characteristics of the target clustering cluster being in the convex hull of the floating object.

[0141] The above scheme obtains floating object clusters related to floating objects by clustering radar data about floating objects collected on the water surface, and determines the floating object convex hull based on the characteristics of multiple floating object clusters, so that after the radar device collects radar data on the water surface, the target cluster can be determined according to the collected radar data. If the characteristics of the target cluster are within the convex hull of the floating object, it is determined that the radar data collected by the radar device is from the floating object, and it can be determined that the target object on the water surface is a floating object. Compared with simply determining whether the target object is a floating object based on the degree of proximity between the target object's moving speed and the flow velocity of the water flow, this scheme has more accurate detection results of floating objects on the water surface.

[0142] Among them, the functions of each module can be found in the embodiment of the method for detecting floating objects on the water surface, and will not be repeated here.

[0143] See also Figure 8 , Figure 8 : is a schematic diagram of the structure of an embodiment of an electronic device of the present application. The electronic device 50 includes a memory 51 and a processor 52, and the processor 52 is used to execute program instructions stored in the memory 51 to implement the steps in any of the above-mentioned embodiments of the method for detecting floating objects on the water surface. In a specific implementation scenario, the electronic device 50 may include but is not limited to: a microcomputer, a server, and in addition, the electronic device 50 may also include a carrier device such as a laptop computer and a tablet computer, which is not limited here.

[0144] Specifically, the processor 52 is used to control itself and the memory 51 to implement the steps in any of the above-mentioned embodiments of the method for detecting floating objects on the water surface. The processor 52 can also be called a CPU (Central Processing Unit). The processor 52 may be an integrated circuit chip with signal processing capabilities. The processor 52 can also be a general-purpose processor, a digital signal processor (Digital Signal Processor, DSP), an application-specific integrated circuit (Application Specific Integrated Circuit, ASIC), a field-programmable gate array (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. In addition, the processor 52 can be implemented by an integrated circuit chip.

[0145] The above scheme obtains floating object clusters related to floating objects by clustering radar data about floating objects collected on the water surface, and determines the floating object convex hull based on the characteristics of multiple floating object clusters, so that after the radar device collects radar data on the water surface, the target cluster can be determined according to the collected radar data. If the characteristics of the target cluster are within the convex hull of the floating object, it is determined that the radar data collected by the radar device is from the floating object, and it can be determined that the target object on the water surface is a floating object. Compared with simply determining whether the target object is a floating object based on the degree of proximity between the target object's moving speed and the flow velocity of the water flow, this scheme has more accurate detection results of floating objects on the water surface.

[0146] See also Fig. 9 , Fig. 9 The computer-readable storage medium 60 stores program instructions 601, which, when executed by a processor, implement the steps in any of the above-mentioned methods for detecting floating objects on a water surface.

[0147] The above scheme obtains floating object clusters related to floating objects by clustering radar data about floating objects collected on the water surface, and determines the floating object convex hull based on the characteristics of multiple floating object clusters, so that after the radar device collects radar data on the water surface, the target cluster can be determined according to the collected radar data. If the characteristics of the target cluster are within the convex hull of the floating object, it is determined that the radar data collected by the radar device is from the floating object, and it can be determined that the target object on the water surface is a floating object. Compared with simply determining whether the target object is a floating object based on the degree of proximity between the target object's moving speed and the flow velocity of the water flow, this scheme has more accurate detection results of floating objects on the water surface.

[0148] In some embodiments, the functions or modules included in the device provided by the embodiments of the present disclosure can be used to execute the method described in the above method embodiments. The specific implementation can refer to the description of the above method embodiments, and for the sake of brevity, it will not be repeated here.

[0149] The above description of various embodiments tends to emphasize the differences between the various embodiments. The same or similar aspects can be referenced to each other, and for the sake of brevity, they will not be repeated herein.

[0150] In the several embodiments provided in the present application, it should be understood that the disclosed methods and devices can be implemented in other ways. For example, the device implementation described above is only schematic. For example, the division of modules or units is only a logical function division. There may be other division methods in actual implementation. For example, units or components can be combined or integrated into another system, or some features can be ignored or not executed. In another image position, the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, and the indirect coupling or communication connection of devices or units can be electrical, mechanical or other forms.

[0151] In addition, each functional unit in each embodiment of the present application can be integrated into a processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of a software functional unit. If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or part of the contribution to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor (processor) to perform all or part of the steps of each implementation method of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program code.

Claims

1. A method for detecting floating objects on water surface, It is characterized in that include: Clustering radar data collected by the radar device to obtain target clusters, wherein the radar data records relevant information of several target objects on the water surface; Determining the characteristics of the target cluster based on the relevant information of each target object in the target cluster; Determine whether the feature of the target cluster is within the convex hull of the floating object, the convex hull of the floating object includes a plurality of target points, the convex hull of the floating object is a convex hull surrounded by the target points, each of the target points is used to represent the feature of a floating object cluster, each floating object cluster is determined by using radar data about floating objects collected on the water surface at multiple times by a radar device, wherein the radar data about floating objects at each time corresponds to a floating object cluster and the features of each floating object cluster in a preset dimensional space are respectively used as the target points; In response to the feature of the target cluster being within the convex hull of the floating object, the plurality of target objects are determined to be the floating objects on the water surface.

2. The method according to claim 1, It is characterized in that The relevant information of the target object includes position information, speed information and / or intensity information. The radar data collected by the radar device is clustered to obtain the target cluster, including: Determining a position relationship, a speed relationship and / or a strength relationship between the target objects based on the position information, speed information and / or strength information of the target objects; The collected radar data is clustered in combination with each of the position relationships, speed relationships and / or intensity relationships to obtain the target clusters, wherein the characteristics of the target clusters include average speed information, average intensity information, intensity variance and / or the area of ​​the target convex hull.

3. The method according to claim 2, It is characterized in that The relevant information of the target object includes the position information, and determining the position relationship, speed relationship and / or strength relationship between the target objects based on the position information, speed information and / or strength information of each target object includes: Determining clustering precision, where different clustering precisions result in different numbers of target objects included in clusters; adjusting a method for obtaining the positional relationship according to the clustering accuracy; Based on the adjusted method for acquiring the positional relationship and the position information of each of the target objects, the positional relationship between the target objects is determined.

4. The method according to any one of claims 1 to 3, It is characterized in that The step of determining whether the feature of the target cluster is within the convex hull of the floating object comprises: Determine boundary target points in the convex hull of the floating object, each of the boundary target points is located on a boundary in the convex hull of the floating object; In response to the existence of a set of target data groups satisfying a preset constraint condition, determining that the characteristics of the target cluster are within the convex hull of the floating object; Among them, the preset constraints include: the sum of the products of each target data in the target data group and each boundary target point is the characteristic of the target cluster, the sum of each target data in the target data group is 1, and each target data in the target data group is greater than or equal to 0 and less than or equal to 1.

5. The method according to claim 4, It is characterized in that The method further comprises: Acquire radar data on floating objects on the water surface collected by the radar device at multiple times; Obtain the characteristics of the floating object clusters corresponding to the radar data about the floating objects at each moment respectively; The features of each of the floating object clusters are respectively used as target points in a space of preset dimensions, wherein the dimension of the space is related to the dimension of the features of the floating object clusters; The convex hull formed by each of the target points is used as the convex hull of the floating object.

6. The method according to any one of claims 1 to 3, It is characterized in that The relevant information of the plurality of target objects includes speed information. Before clustering the radar data collected by the radar device to obtain the target clusters, the method further includes: Determining the direction of water flow on the water surface and determining the moving direction of each target object based on the speed information of each target object; The radar data having a direction different from that of the water flow is removed from the radar data.

7. The method according to any one of claims 1 to 3, It is characterized in that The method further comprises: In response to the fact that the feature of the target cluster is within the convex hull of the ship, the plurality of target objects are determined to be ships, and the convex hull of the ship is determined by using radar data of the ship collected on the water surface at multiple times.

8. The method according to any one of claims 1 to 3, It is characterized in that After determining that the plurality of target objects are floating objects on the water surface in response to the feature of the target cluster being within the convex hull of the floating object, the method further includes: Acquire a target convex hull area of ​​the target cluster, where the target convex hull area of ​​the target cluster is determined based on the position of each target object in the cluster; In response to the target convex hull area of ​​the target cluster being greater than or equal to a preset area, sending a prompt message; and / or, In response to the target convex hull area of ​​the target cluster being smaller than the preset area, determining whether a sum of a plurality of historical convex hull areas and the target convex hull area of ​​the target cluster is greater than or equal to the preset area, wherein the historical convex hull area is the target convex hull area of ​​the target cluster determined based on the radar data collected on the water surface at the plurality of historical moments; In response to the sum of the areas of the multiple historical convex hulls and the target convex hull area of ​​the target cluster being greater than or equal to a preset area, a prompt message is sent.

9. An electronic device, It is characterized in that The invention comprises a memory and a processor, wherein the processor is used to execute program instructions stored in the memory to implement the method according to any one of claims 1 to 8.

10. A computer-readable storage medium having program instructions stored thereon, It is characterized in that When the program instructions are executed by a processor, the method according to any one of claims 1 to 8 is implemented.

Images