Flood event detection method and device, storage medium and electronic device

Abstract

The present application discloses a method and device for detecting flood events, a storage medium and an electronic device. The method includes: using a target radar to detect a target area and determine multiple frames of point cloud data; determining multiple frames of same-direction point cloud data whose radial velocity meets preset conditions from the multiple frames of point cloud data, sampling the multiple frames of same-direction point cloud data at intervals, determining multiple groups of target point cloud data and multiple target feature vectors corresponding to the multiple groups of target point cloud data; determining multiple positive sample weight values ​​according to multiple target feature vectors and a predetermined positive sample feature matrix, and determining multiple negative sample weight values ​​according to multiple target feature vectors and a predetermined negative sample feature matrix; determining whether a flood event has occurred in the target area according to whether the multiple positive sample weight values ​​meet the first preset condition and whether the multiple negative sample weight values ​​meet the second preset condition. The present application solves the technical problem of being unable to accurately detect flood events.

Description

Technical Field

[0001] The present application relates to the field of computers, and in particular, to a method and device for detecting flood events, a storage medium, and an electronic device. Background Art

[0002] At present, the relevant technologies mainly rely on meteorological data collected through historical statistics, data from ground monitoring stations, etc. to detect flood events. That is, traditional flood event detection methods require a long time and a large range of data collection and analysis, which makes the detection results unable to meet the requirements of real-time and accuracy. In other words, since flood events are often sudden, only rough detection based on historical statistics leads to the technical problem that related technologies cannot accurately detect flood events.

[0003] To address the above-mentioned problems, no effective solution has been proposed yet. Summary of the invention

[0004] The embodiments of the present application provide a method and device for detecting a flood event, a storage medium, and an electronic device, so as to at least solve the technical problem that a flood event cannot be accurately detected.

[0005] According to one aspect of an embodiment of the present application, a method for detecting flood events is provided, comprising: using a target radar to detect a target area and determine multiple frames of point cloud data, wherein the detection direction of the target radar is set to be the same as the flow direction of a river in the target area; determining multiple frames of same-direction point cloud data whose radial velocities meet preset conditions from the multiple frames of point cloud data, sampling the multiple frames of same-direction point cloud data at intervals, determining multiple groups of target point cloud data and multiple target feature vectors corresponding to the multiple groups of target point cloud data, wherein one of the target feature vectors is composed of the radial velocity and radar reflection area corresponding to each frame of point cloud data in a group of the target point cloud data. Determine; determine a plurality of positive sample weight values ​​according to the plurality of target feature vectors and a predetermined positive sample feature matrix, and determine a plurality of negative sample weight values ​​according to the plurality of target feature vectors and a predetermined negative sample feature matrix, wherein one of the positive sample weight values ​​is used to indicate the degree of similarity between one of the target feature vectors and the positive sample feature matrix, and one of the negative sample weight values ​​is used to indicate the degree of similarity between one of the target feature vectors and the negative sample feature matrix; determine whether a flood event has occurred in the target area according to whether the plurality of positive sample weight values ​​satisfy a first preset condition and whether the plurality of negative sample weight values ​​satisfy a second preset condition.

[0006] According to another aspect of the embodiment of the present application, a flood event detection device is also provided, including: a detection module, used to use a target radar to detect a target area and determine multiple frames of point cloud data, wherein the detection direction of the target radar is set to be the same as the flow direction of the river in the target area; a processing module, used to determine multiple frames of same-direction point cloud data whose radial velocities meet preset conditions from the multiple frames of point cloud data, sample the multiple frames of same-direction point cloud data at intervals, determine multiple groups of target point cloud data and multiple target feature vectors corresponding to the multiple groups of target point cloud data, wherein one target feature vector is composed of radial velocities and radar reflection surfaces corresponding to each frame of point cloud data in a group of target point cloud data The invention relates to a method for determining a plurality of target feature vectors and a predetermined positive sample feature matrix according to the plurality of target feature vectors. The method comprises the following steps: an execution module for determining a plurality of positive sample weight values ​​according to the plurality of target feature vectors and a predetermined positive sample feature matrix, and determining a plurality of negative sample weight values ​​according to the plurality of target feature vectors and a predetermined negative sample feature matrix, wherein one of the positive sample weight values ​​is used to indicate the degree of similarity between one of the target feature vectors and the positive sample feature matrix, and one of the negative sample weight values ​​is used to indicate the degree of similarity between one of the target feature vectors and the negative sample feature matrix; a determination module for determining whether a flood event has occurred in the target area according to whether the plurality of positive sample weight values ​​satisfy a first preset condition and whether the plurality of negative sample weight values ​​satisfy a second preset condition.

[0007] Optionally, the device is used to determine whether a flood event has occurred in the target area according to whether the multiple positive sample weight values ​​meet a first preset condition and whether the multiple negative sample weight values ​​meet a second preset condition in the following manner: when the multiple positive sample weight values ​​meet the first preset condition and the multiple negative sample weight values ​​do not meet the second preset condition, determine that the flood event has occurred in the target area; when the multiple positive sample weight values ​​meet the first preset condition and the multiple negative sample weight values ​​meet the second preset condition, determine whether the flood event has occurred in the target area according to a pre-set sensitivity parameter.

[0008] Optionally, the device is used to determine whether the flood event has occurred in the target area according to a preset sensitivity parameter in the following manner when the multiple positive sample weight values ​​meet the first preset condition and the multiple negative sample weight values ​​meet the second preset condition: when the multiple positive sample weight values ​​meet the first preset condition and the multiple negative sample weight values ​​meet the second preset condition, the positive sample feature matrix and the negative sample feature matrix are mapped to the target plane to determine the positive sample convex hull and the negative sample convex hull, wherein the positive sample convex hull represents the area surrounding each positive sample in the positive sample feature matrix. The minimum convex polygon of the feature point, the negative sample convex hull represents the minimum convex polygon surrounding each negative sample feature point in the negative sample feature matrix; determining the first convex hull centroid corresponding to the positive sample convex hull and the second convex hull centroid corresponding to the negative sample convex hull; determining a positive-negative sample two-dimensional dividing line according to the first convex hull centroid and the second convex hull centroid, wherein the positive-negative sample two-dimensional dividing line is used to divide the intersection area of ​​the positive sample convex hull and the negative sample convex hull; using the positive-negative sample two-dimensional dividing line to divide the intersection area, and determining whether the target area has generated the flood event based on the division result and the sensitivity parameter.

[0009] Optionally, the device is used to determine the two-dimensional dividing line between positive and negative samples according to the first convex hull centroid and the second convex hull centroid in the following manner: connect the first convex hull centroid and the second convex hull centroid to obtain a target line segment; divide the target line segment according to the sensitivity parameter to determine the target midpoint; determine the two-dimensional dividing line between positive and negative samples with the horizontal coordinate of the target midpoint, wherein the two-dimensional dividing line between positive and negative samples is parallel to the longitudinal axis of the target plane.

[0010] Optionally, the device is used to divide the intersection area using the positive and negative sample two-dimensional dividing line in the following manner, and determine whether the target area has caused the flood event based on the division result and the sensitivity parameter: use the positive and negative sample two-dimensional dividing line to divide the intersection area to obtain a first intersection area and a second intersection area; when the area ratio of the first intersection area to the intersection area is greater than the sensitivity parameter, determine that the flood event has occurred in the target area; when the area ratio of the first intersection area to the intersection area is less than or equal to the sensitivity parameter, determine that the target area has not caused the flood event.

[0011] Optionally, the device is used to determine multi-frame same-direction point cloud data whose radial velocity satisfies preset conditions from the multi-frame point cloud data in the following manner: obtaining the radial velocity of each point of each frame of point cloud data in the multi-frame point cloud data; determining the points whose radial velocity satisfies the preset conditions as data points in the same-direction point cloud data to obtain the multi-frame same-direction point cloud data; sampling the multi-frame same-direction point cloud data at intervals of frames to determine multiple groups of target point cloud data and multiple target feature vectors corresponding to the multiple groups of target point cloud data.

[0012] Optionally, the device is used to perform frame-by-frame sampling on the multiple frames of same-direction point cloud data to determine multiple groups of target point cloud data and multiple target feature vectors corresponding to the multiple groups of target point cloud data in the following manner: obtain a predetermined set of tail frame sequence numbers, wherein the tail frame sequence number is used to indicate the target point cloud data corresponding to the last frame in a group of target point cloud data; perform frame-by-frame sampling according to the set of tail frame sequence numbers to determine the multiple groups of target point cloud data, wherein a group of target point cloud data includes the same-direction point cloud data from different frames, and the different frames are not adjacent to each other, and any two data points in a group of target point cloud data satisfy a preset distance condition, a preset speed difference condition, and a preset radar reflection area difference condition; determine the multiple target feature vectors based on the radial velocity mean, radial velocity variance, radar reflection area mean, and radar reflection area variance of each group of target point cloud data.

[0013] Optionally, the device is used to determine whether a flood event has occurred in the target area according to whether the multiple positive sample weight values ​​meet a first preset condition and whether the multiple negative sample weight values ​​meet a second preset condition in the following manner: when the multiple positive sample weight values ​​do not meet the first preset condition and the multiple negative sample weight values ​​meet the second preset condition, determine that the flood event has not occurred in the target area; when the multiple positive sample weight values ​​do not meet the first preset condition and the multiple negative sample weight values ​​do not meet the second preset condition, determine that at least one corresponding sample data in the positive sample feature matrix and the negative sample feature matrix is ​​abnormal.

[0014] According to another aspect of the embodiments of the present application, a computer-readable storage medium is provided, in which a computer program is stored, wherein the computer program is configured to execute the above-mentioned flood event detection method when running.

[0015] According to another aspect of the embodiment of the present application, a computer program product or a computer program is provided, the computer program product or the computer program includes computer instructions, the computer instructions are stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device performs the above flood event detection method.

[0016] According to another aspect of the embodiments of the present application, there is also provided an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor is configured to execute the above-mentioned method for detecting flood events through the computer program.

[0017] In an embodiment of the present application, a target radar is used to detect a target area to determine multiple frames of point cloud data, wherein the detection direction of the target radar is set to be the same as the flow direction of a river in the target area; multiple frames of same-direction point cloud data whose radial velocities meet preset conditions are determined from the multiple frames of point cloud data, and the multiple frames of same-direction point cloud data are sampled at intervals to determine multiple groups of target point cloud data and multiple target feature vectors corresponding to the multiple groups of target point cloud data, wherein a target feature vector is determined by the radial velocity and radar reflection area corresponding to each frame of point cloud data in a group of target point cloud data; multiple positive sample weight values ​​are determined based on multiple target feature vectors and a predetermined positive sample feature matrix, and multiple negative sample weight values ​​are determined based on multiple target feature vectors and a predetermined negative sample feature matrix, wherein a positive sample weight value is used to indicate the similarity of a target feature vector to the positive sample feature matrix. The method comprises the following steps: a negative sample weight value is used to indicate the similarity between a target feature vector and a negative sample feature matrix; a method for determining whether a flood event has occurred in a target area is performed according to whether multiple positive sample weight values ​​satisfy a first preset condition and whether multiple negative sample weight values ​​satisfy a second preset condition, that is, using a radar to detect the target area along the river direction to obtain multiple frames of point cloud data; screening out the same-direction point cloud data whose radial velocity meets specific conditions, performing interval sampling, and forming multiple groups of target point cloud data and corresponding target feature vectors; then, using the target feature vector and the preset positive and negative sample feature matrices, calculating the corresponding weight value; accurately detecting whether a flood event has occurred in the target area based on the weight value, thereby achieving the technical effect of high-precision detection of flood events, and effectively improving the accuracy of detecting flood events, thereby solving the technical problem of being unable to accurately detect flood events. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0019] Figure 1 is a schematic diagram of an application environment of an optional flood event detection method according to an embodiment of the present application;

[0020] Figure 2 is a flow chart of an optional flood event detection method according to an embodiment of the present application;

[0021] Figure 3 is a schematic diagram of an optional flood event detection method according to an embodiment of the present application;

[0022] Figure 4 is a schematic diagram of another optional flood event detection method according to an embodiment of the present application;

[0023] Figure 5 is a schematic diagram of another optional flood event detection method according to an embodiment of the present application;

[0024] Figure 6 is a schematic diagram of another optional flood event detection method according to an embodiment of the present application;

[0025] Figure 7 is a schematic diagram of another optional flood event detection method according to an embodiment of the present application;

[0026] Figure 8 is a schematic diagram of another optional flood event detection method according to an embodiment of the present application;

[0027] Fig. 9 is a schematic diagram of another optional flood event detection method according to an embodiment of the present application;

[0028] Fig.10 is a schematic diagram of another optional flood event detection method according to an embodiment of the present application;

[0029] Fig.11 is a schematic diagram of another optional flood event detection method according to an embodiment of the present application;

[0030] Fig.12 is a schematic diagram of another optional flood event detection method according to an embodiment of the present application;

[0031] Fig.13 is a schematic diagram of another optional flood event detection method according to an embodiment of the present application;

[0032] Fig.14 is a schematic structural diagram of an optional flood event detection device according to an embodiment of the present application;

[0033] Fig.15is a schematic structural diagram of an optional flood event detection product according to an embodiment of the present application;

[0034] Fig.16 It is a schematic diagram of the structure of an optional electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0035] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0036] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0037] Specifically, the embodiment of the present application can solve the cloth self-collision problem in neural network simulation. The loss function of the target motion model does not rely on the collision-free assumption, and is therefore suitable for the network training process of deep learning.

[0038] The present application is described below in conjunction with embodiments:

[0039] According to one aspect of an embodiment of the present application, a method for detecting a flood event is provided. Optionally, in this embodiment, the method for detecting a flood event can be applied to: Figure 1 In the hardware environment composed of the server 101 and the terminal device 103 shown in FIG. Figure 1As shown, the server 101 is connected to the terminal device 103 via a network, and can be used to provide services for the terminal device or an application 107 installed on the terminal device. The application can be a video application, an instant messaging application, a browser application, an educational application, a game application, etc. A database 105 may be set up on the server or independently of the server to provide data storage services for the server 101, for example, a game data storage server. The above-mentioned network may include but is not limited to: a wired network, a wireless network, wherein the wired network includes: a local area network, a metropolitan area network and a wide area network, and the wireless network includes: Bluetooth, WIFI and other networks that realize wireless communication. The terminal device 103 may be a terminal configured with an application, and may include but is not limited to at least one of the following: a mobile phone (such as an Android phone, an iOS phone, etc.), a laptop computer, a tablet computer, a PDA, a MID (Mobile Internet Devices), a PAD, a desktop computer, a smart TV, an intelligent voice interaction device, a smart home appliance, a vehicle-mounted terminal, an aircraft, a virtual reality (Virtual Reality, VR for short) terminal, an augmented reality (Augmented Reality, AR for short) terminal, a mixed reality (Mixed Reality, MR for short) terminal and other computer devices. The above-mentioned server may be a single server, or a server cluster consisting of multiple servers, or a cloud server.

[0040] Combination Figure 1 As shown, the above-mentioned method for detecting flood events can be executed by an electronic device, which can be a terminal device or a server. The above-mentioned method for detecting flood events can be implemented by the terminal device or the server separately, or by the terminal device and the server together.

[0041] The above is only an example and is not specifically limited in this embodiment.

[0042] Optionally, as an optional implementation, as Figure 2 As shown, the above-mentioned flood event detection method includes:

[0043] S202, using a target radar to detect the target area and determine multiple frames of point cloud data, wherein the detection direction of the target radar is set to be the same as the flow direction of the river in the target area;

[0044] Optionally, in an embodiment of the present application, the above-mentioned target radar may include but is not limited to millimeter wave radar, synthetic aperture radar, lidar, etc., the above-mentioned target area may include but is not limited to rivers, mountainous areas, forests, etc., and the above-mentioned multi-frame point cloud data can be understood as point set data obtained at different time points and representing the three-dimensional spatial information of the above-mentioned target area, wherein each frame of point cloud data represents a scanning result of the radar on the target area at that time point, and may include but is not limited to the position, speed, reflection intensity and other information of the measurement points in the target area (for example, mud in the river, pedestrians on the road, branches, etc.).

[0045] Exemplarily, taking the target area as a river as an example, the above-mentioned target radar is deployed on stable racks on both sides of the river, facing the river area to be detected. At this time, the detection direction of the target radar is the same as the flow direction of the river.

[0046] It should also be noted that, assuming that the target area is a river, it can be used in different areas of the river according to the actual application scenario, for example, Figure 3 is a schematic diagram of an optional flood event detection method according to an embodiment of the present application, such as Figure 3 As shown, the above-mentioned target radar is deployed every 500 meters, and the detection direction of each target radar is set to be the same as the river flow direction.

[0047] S204, determining multiple frames of same-direction point cloud data whose radial velocities meet preset conditions from multiple frames of point cloud data, sampling the multiple frames of same-direction point cloud data at intervals, and determining multiple groups of target point cloud data and multiple target feature vectors corresponding to the multiple groups of target point cloud data, wherein one target feature vector is determined by the radial velocity and radar reflection area corresponding to each frame of point cloud data in a group of target point cloud data;

[0048] Optionally, in an embodiment of the present application, the radial velocity refers to the velocity component of each measurement point detected by the target radar contained in the multi-frame point cloud data, and the velocity component is the velocity along the direction of radar wave propagation (i.e., radial direction). The preset condition can be understood as, among the measurement points detected by the target radar, the measurement point with a radial velocity value greater than 0, that is, the measurement point whose motion trajectory is away from the target radar, and the moving direction of the measurement point is also the same as the above-mentioned river flow direction. Therefore, the point cloud data corresponding to the measurement point will be determined as the above-mentioned same-direction point cloud data.

[0049] It should be noted that the target radar will detect several measurement points in each frame, each measurement point has its own point cloud data, and the number of measurement points detected in different time frames is not limited. Figure 4 is a schematic diagram of another optional flood event detection method according to an embodiment of the present application, such as Figure 4As shown, 6 measurement points are detected in the 4501st frame, and the ID is used as the point cloud data number of the measurement point. The point cloud data with ID number 0 is taken as an example here, including the coordinate position of the measurement point (the deployment position of the target radar is used as the coordinate origin to establish the coordinate system, and the extension directions of the X-axis, Y-axis, and Z-axis are not limited), the echo intensity RCS of the radar wave of the measurement point to the target radar, and the radial speed RadialSpeed. Similarly, the point cloud data corresponding to other measurement points can include their respective coordinate positions, echo intensity RCS, radial speed RadialSpeed, etc., which will not be repeated here.

[0050] Furthermore, after determining the above-mentioned multiple frames of same-direction point cloud data, the above-mentioned frame-interval sampling will be performed on these multiple frames of same-direction point cloud data. Specifically, assuming that 10 frames of same-direction point cloud data are included, it may include but is not limited to determining the same-direction point cloud data of the first frame, the same-direction point cloud data of the third frame, and the same-direction point cloud data of the fifth frame as a group of target point cloud data among the above-mentioned multiple groups of target point cloud data. Similarly, the same-direction point cloud data of the second frame, the same-direction point cloud data of the fourth frame, and the same-direction point cloud data of the sixth frame may also be determined as a group of target point cloud data among the above-mentioned multiple groups of target point cloud data, and so on.

[0051] In an exemplary embodiment, it may include but is not limited to first determining multiple tail frames from the time frames corresponding to the above-mentioned multiple frames of same-direction point cloud data, denoted as K=[4, 7...], then determining the same-direction point cloud data corresponding to non-adjacent time frames from the same-direction point cloud data corresponding to the three time frames before the 4th frame, and the same-direction point cloud data corresponding to the 4th frame as a group of target point cloud data in the above-mentioned multiple groups of target point cloud data, for example, determining the same-direction point cloud data corresponding to the 1st frame and the 4th frame as a group of target point cloud data in the above-mentioned multiple groups of target point cloud data; then, relative to when the value of K is 7, due to the corresponding same-direction point cloud data of the 1st frame and the 4th frame The same-direction point cloud data has been determined as one of the multiple groups of target point cloud data. Therefore, for the 7th frame, the same-direction point cloud data corresponding to the non-adjacent time frames can be determined from the same-direction point cloud data corresponding to the 2nd frame, the 3rd frame, the 5th frame, and the 6th frame, and the same-direction point cloud data corresponding to the 7th frame are determined as one of the multiple groups of target point cloud data. For example, the same-direction point cloud data corresponding to the 2nd frame, the 5th frame, and the 7th frame are determined as another group of target point cloud data in the multiple groups of target point cloud data. The determination process of the target point cloud data of the respective groups corresponding to the remaining tail frames is the same, which will not be repeated here.

[0052] It should be noted that the above frame-interval sampling method may also include but is not limited to periodic sampling, etc., and this application does not impose any limitation on this.

[0053] Exemplarily, after obtaining the above-mentioned multiple groups of target point cloud data, each group of target point cloud data can determine the target feature vector corresponding thereto, thereby obtaining the above-mentioned multiple target feature vectors. The determination method of the target feature vector corresponding to one group of target point cloud data is used as an example for explanation. The specific implementation process includes but is not limited to:

[0054] S1, obtain a set of target point cloud data and record it as cluster set C(k), where k refers to the last frame in several time frames corresponding to this set of point cloud data, including point cloud data of several measurement points, including the measurement point e i Point cloud data, that is, e i ∈C(k);

[0055] S2, determine whether there is a point e in this set of point cloud data. i Another different measurement point e j Point cloud data, that is, e j ∈C(k) and e j ≠e i , if it exists, execute S3, otherwise, execute S6;

[0056] S3, using measuring point e i 、Measurement point e j Coordinate position information recorded in each point cloud data , calculate the measurement point e i 、Measurement point e j The "ellipse" distance , the "ellipse" distance refers to the measurement point e i 、Measurement point e j The shortest path distance between them, and may include but is not limited to pre-setting a , indicating the distance threshold, judging Is it true? If it is true, execute S4. Otherwise, execute S6, where , The value is not limited;

[0057] S4, using the measuring point e i 、Measurement point e j The radial velocity recorded in the respective point cloud data , calculate the measurement point e i 、Measurement point e j Speed ​​difference , and may include but is not limited to pre-setting a , indicating the speed difference threshold, judging Is it true? If it is true, execute S5. Otherwise, execute S6, where ;

[0058] S5, using the measuring point e i 、Measurement point e j Radar reflection areas recorded in the respective point cloud data , calculate the measurement point e i 、Measurement point e j The echo strength , and may include but is not limited to pre-setting a , indicating the echo strength difference threshold, judging Is it true? If it is true, execute S7. Otherwise, execute S6. ;

[0059] S6, delete measurement point e i point cloud data to update this set of point cloud data;

[0060] S7, re-determine a new measurement point from the set of point cloud data and execute steps S2-S5 until all the measurement points in the set of point cloud data are traversed and executed, and then execute S8;

[0061] S8, determining the number of measurement points corresponding to the point cloud data in this set of point cloud data Is it greater than the target number threshold? , that is , where the target number threshold It can be set flexibly. If it is satisfied, S9 is executed. Otherwise, this set of point cloud data is deleted from the above multiple sets of point cloud data.

[0062] S9, determine the target feature vector V(k) corresponding to this set of point cloud data by the following formula = :

[0063] , , , , ,in, represents the average velocity in C(k), represents the velocity variance in C(k), represents the average intensity in C(k), represents the intensity variance in C(k), Represents the area of ​​the convex hull in C(k), which refers to the area of ​​the smallest convex polygon that contains all the measurement points in the cluster set C(k).

[0064] For example, Figure 5 is a schematic diagram of another optional flood event detection method according to an embodiment of the present application, wherein the convex hull area is as follows: Figure 5 As shown, Figure 5 A point in represents a measurement point.

[0065] S206, determining a plurality of positive sample weight values ​​according to a plurality of target feature vectors and a predetermined positive sample feature matrix, and determining a plurality of negative sample weight values ​​according to a plurality of target feature vectors and a predetermined negative sample feature matrix, wherein a positive sample weight value is used to indicate a similarity between a target feature vector and a positive sample feature matrix, and a negative sample weight value is used to indicate a similarity between a target feature vector and a negative sample feature matrix;

[0066] Optionally, the positive sample feature matrix includes positive sample point cloud data formed when a flood event occurs at a historical moment, and the positive sample point cloud data may include but is not limited to manually annotated and corrected data; the negative sample feature matrix includes negative sample point cloud data formed when no flood event occurs at a historical moment, and the negative sample point cloud data may include but is not limited to manually annotated and corrected data.

[0067] For example, Figure 6 is a schematic diagram of another optional flood event detection method according to an embodiment of the present application. The positive sample feature matrix can be as follows Figure 6 The convex hull area on the two-dimensional plane shown, similarly, Figure 7 is a schematic diagram of another optional flood event detection method according to an embodiment of the present application. The negative sample feature matrix can be as follows Figure 7 From the area of ​​the convex hull on the two-dimensional plane shown, it can be understood that the size of the negative sample feature matrix is ​​much larger than the positive sample feature matrix.

[0068] Specifically, the positive sample feature matrix includes several water flow feature vectors (equivalent to the target feature vectors) corresponding to the point cloud data formed when the flood event occurred at the historical moment, which are used to determine the water flow feature vectors corresponding to the convex hull vertices of the convex hull area, such as Figure 6 The water flow feature vector corresponding to point 602 in is an element in the positive sample feature matrix, and point 604 in the convex hull is not a convex hull vertex used to determine the area of ​​the convex hull. Therefore, the positive sample feature matrix does not include the water flow feature vector corresponding to point 604; similarly, the negative sample feature matrix includes several water flow feature vectors corresponding to the point cloud data constructed when no flood event occurred in history, which are used to determine the convex hull vertex.

[0069] It should be noted that the water flow feature vector can be a multi-dimensional vector. The water flow feature vector can be reduced in dimension through a preset dimensional mapping relationship, as shown above. Figure 6 , Figure 7 A point on a two-dimensional plane.

[0070] In an exemplary embodiment, the positive sample feature matrix can be used to determine the positive sample weights corresponding to the target feature vectors through the following steps. The positive sample weight α of a target feature vector is used as an example. First, assume that the positive sample feature matrix is ​​expressed as , p represents the number of convex hull vertices, and the target feature vector is represented as V(k):

[0071] S1, obtain the target feature vector V(k), and the positive sample feature matrix is ​​expressed as , p takes a value greater than 1;

[0072] S2, using V(k) = , where the positive sample weight is , x represents the size of α;

[0073] Furthermore, the above-mentioned multiple target feature vectors may be calculated to obtain the corresponding positive sample weight values.

[0074] In another exemplary embodiment, the positive sample feature matrix can be used to determine the positive sample weights corresponding to the target feature vectors by the following steps: To illustrate, first, assume that the positive sample feature matrix is ​​expressed as , q represents the number of convex hull vertices, and the target feature vector is represented as V(k):

[0075] S1, obtain the target feature vector V(k), and the positive sample feature matrix is ​​expressed as , q takes a value greater than 1;

[0076] S2, using V(k) = , where the negative sample weight value , y represents Size;

[0077] Furthermore, the above-mentioned multiple target feature vectors may be calculated to obtain the corresponding negative sample weight values.

[0078] S208 , determining whether a flood event occurs in the target area according to whether the weight values ​​of the multiple positive samples meet a first preset condition and whether the weight values ​​of the multiple negative samples meet a second preset condition.

[0079] Optionally, in an embodiment of the present application, the first preset condition is used to judge the weight value of the positive sample, and the second preset condition is used to judge the weight value of the negative sample, specifically:

[0080] Assume that the weight values ​​of multiple positive samples are expressed as [α A , α B……], the weight values ​​of multiple negative samples are expressed as [β A , β B ……], then, judge [α A , α B ...] whether all weight values ​​satisfy and =1, here refers to [α A , α B …] any weight value α= For example, α A =[0.4, 0.6], at this time, i takes the value of 0 or 1, that is, =0.4 or 0.6, at this time, judge , Is it established, and =1 is true, if both are true, then it means α A satisfy and =1, then continue to judge α B Is it satisfied? and =1, until the weight values ​​of the above positive samples all meet and =1; At the same time, it is also necessary to determine whether the negative sample weight value satisfies and =1, here refers to A , B Any weight value in = The specific judgment logic is the same as the judgment logic of the positive sample weight value mentioned above.

[0081] Furthermore, if the above positive sample weights all satisfy and =1, it means that the positive sample weight value meets the first preset condition, if the negative sample weight values ​​meet and =1, it means that the negative sample weight value meets the second preset condition.

[0082] In an exemplary embodiment, if the positive sample weight value satisfies the first preset condition and the negative sample weight value does not meet the second preset condition, it indicates that a flood event has occurred in the target area; conversely, if the negative sample weight value satisfies the second preset condition and the positive sample weight value does not meet the first preset condition, it indicates that a flood event has not occurred in the target area.

[0083] As an optional scheme, the above method also includes: determining whether a flood event has occurred in the above target area according to whether the above multiple positive sample weight values ​​meet the first preset condition and whether the above multiple negative sample weight values ​​meet the second preset condition: when the above multiple positive sample weight values ​​do not meet the above first preset condition and the above multiple negative sample weight values ​​meet the above second preset condition, determining that the above target area has not occurred in the above flood event; when the above multiple positive sample weight values ​​do not meet the above first preset condition and the above multiple negative sample weight values ​​do not meet the above second preset condition, determining that at least one corresponding sample data in the above positive sample feature matrix and the above negative sample feature matrix is ​​abnormal.

[0084] In another exemplary embodiment, if the positive sample weight value does not meet the first preset condition and the positive sample weight value does not meet the second preset condition, it means that there are errors in the positive sample feature matrix and the negative sample feature matrix, and adjustment is required; or, Figure 8 is a schematic diagram of another optional flood event detection method according to an embodiment of the present application. If there is at least one of the above-mentioned target feature vectors A corresponding to the positive sample weight value A among the above-mentioned multiple positive sample weight values, and the target feature vector A also corresponds to the negative sample weight value A among the above-mentioned multiple negative sample weight values, the positive sample weight value A satisfies: 0≤α i ≤1 and =1 and the negative sample weight value A satisfies and =1, it means that the weight values ​​of the above-mentioned multiple positive samples meet the above-mentioned first preset condition, and the weight values ​​of the above-mentioned multiple negative samples meet the above-mentioned second preset condition. It can be understood that the point of the target feature vector A after dimensionality reduction processing falls on the intersection area of ​​the convex hull area of ​​the above-mentioned positive sample feature matrix and the convex hull area of ​​the negative sample feature matrix on the two-dimensional plane. ,like Figure 8 shown.

[0085] Furthermore, the centroid positions of the convex area of ​​the positive sample feature matrix and the convex area of ​​the negative sample feature matrix are determined respectively, and the centroids of the two are connected by a line. Then, based on the intersection area of ​​the convex area Determine whether a flood event has occurred, including but not limited to determining a threshold boundary on a two-dimensional plane using a preset sensitivity parameter, and then Split into two parts according to the threshold boundary and , respectively calculate the positive sample area ratio / And negative sample area ratio / ,like / (Probability ratio) is greater than the preset sensitivity parameter, it means that a flood event occurs in the target area.

[0086] In another exemplary embodiment, Fig. 9 is a schematic diagram of another optional flood event detection method according to an embodiment of the present application. The flood event detection method proposed in the present application can be applied to geological disaster detection application scenarios, including but not limited to: Fig. 9 As shown:

[0087] S902, initializing target clustering parameters (setting relevant parameters for sampling multiple frames of point cloud data in the same direction at intervals), number of accumulated data frames, sensitivity, threshold, river direction, etc.;

[0088] S904, the target radar senses the environment, uses a radar target detection algorithm to measure the environment, and outputs radar point cloud data (the above-mentioned multi-frame point cloud data) in real time;

[0089] S906, obtaining radar point cloud data, grouping the radar point cloud data according to the direction opposite to or in the same direction as the river, and only retaining multi-frame point cloud data in the same direction as the river;

[0090] S908, accumulates multiple frames of same-direction point cloud data. Since the amount of single-frame radar point cloud data is limited, multiple frames can be accumulated to increase the richness of the data and enhance the data anti-interference ability;

[0091] S910, sampling multiple frames of same-direction point cloud data at intervals to achieve clustering of the multiple frames of same-direction point cloud data to obtain multiple groups of target point cloud data;

[0092] S912, determining a plurality of positive sample weight values ​​according to a plurality of target feature vectors and a predetermined positive sample feature matrix, and determining a plurality of negative sample weight values ​​according to a plurality of target feature vectors and a predetermined negative sample feature matrix;

[0093] S914, if the weight values ​​of the multiple positive samples meet the first preset condition, and the weight values ​​of the multiple negative samples do not meet the second preset condition, it is determined that a flood event has occurred in the target area, and S924 is executed; otherwise, S918 is executed;

[0094] S916, if the weight values ​​of the multiple positive samples do not meet the first preset condition, and the weight values ​​of the multiple negative samples meet the second preset condition, it is determined that no flood event occurs in the target area, and S926 is executed; otherwise, S918 is executed;

[0095] S918, if at least one of the multiple positive sample weight values ​​satisfies the first preset condition, and the negative sample weight value determined by the target feature vector corresponding to the positive sample weight value also satisfies the second preset condition, then determine the intersection area based on the positive sample feature matrix and the positive sample feature matrix , if there are one or more target feature vectors whose dimensionality reduction processing points fall on the intersection area of ​​the convex area of ​​the positive sample feature matrix and the convex area of ​​the negative sample feature matrix on the two-dimensional plane , execute S920;

[0096] S920, based on sensitivity parameters, intersection area Determine the probability ratio of flood events ;

[0097] S922, if the probability ratio of a flood event is less than the sensitivity parameter, it means that it is within the safety threshold and no flood event has occurred, and S904 is continued to be executed. Otherwise, it means that a flood event has occurred, and S924 is executed.

[0098] S924, reporting flood events, marking all point cloud data collected by subsequent target radars with the label "flood", and uploading them to the cloud server for use by other locations or sensors downstream of the river;

[0099] S926, end.

[0100] Through the embodiments of the present application, a target radar is used to detect the target area and determine multiple frames of point cloud data, wherein the detection direction of the target radar is set to be the same as the flow direction of the river in the target area; multiple frames of same-direction point cloud data whose radial velocity meets preset conditions are determined from the multiple frames of point cloud data, and the multiple frames of same-direction point cloud data are sampled at intervals to determine multiple groups of target point cloud data and multiple target feature vectors corresponding to the multiple groups of target point cloud data, wherein a target feature vector is determined by the radial velocity and radar reflection area corresponding to each frame of point cloud data in a group of target point cloud data; multiple positive sample weight values ​​are determined based on multiple target feature vectors and a predetermined positive sample feature matrix, and multiple negative sample weight values ​​are determined based on multiple target feature vectors and a predetermined negative sample feature matrix, wherein a positive sample weight value is used to indicate the similarity of a target feature vector to the positive sample feature matrix. The method comprises the following steps: a negative sample weight value is used to indicate the similarity between a target feature vector and a negative sample feature matrix; a method for determining whether a flood event has occurred in a target area is performed according to whether multiple positive sample weight values ​​satisfy a first preset condition and whether multiple negative sample weight values ​​satisfy a second preset condition, that is, using a radar to detect the target area along the river direction to obtain multiple frames of point cloud data; screening out the same-direction point cloud data whose radial velocity meets specific conditions, performing interval sampling, and forming multiple groups of target point cloud data and corresponding target feature vectors; then, using the target feature vector and the preset positive and negative sample feature matrices, calculating the corresponding weight value; accurately detecting whether a flood event has occurred in the target area based on the weight value, thereby achieving the technical effect of high-precision detection of flood events, and effectively improving the accuracy of detecting flood events, thereby solving the technical problem of being unable to accurately detect flood events.

[0101] As an optional scheme, the above-mentioned determining whether a flood event has occurred in the target area based on whether the weight values ​​of the above-mentioned multiple positive samples meet the first preset condition and whether the weight values ​​of the above-mentioned multiple negative samples meet the second preset condition includes: when the weight values ​​of the above-mentioned multiple positive samples meet the first preset condition and the weight values ​​of the above-mentioned multiple negative samples do not meet the second preset condition, determining that the flood event has occurred in the target area; when the weight values ​​of the above-mentioned multiple positive samples meet the first preset condition and the weight values ​​of the above-mentioned multiple negative samples meet the second preset condition, determining whether the flood event has occurred in the target area according to a pre-set sensitivity parameter.

[0102] Optionally, in the embodiment of the present application, the above sensitivity parameters can be preset manually, and the present application does not impose any limitation on this.

[0103] Exemplarily, the fact that the multiple positive sample weight values ​​satisfy the first preset condition indicates that the multiple positive sample weight values ​​have a high similarity with the positive sample feature matrix, while the fact that the multiple negative sample weight values ​​satisfy the second preset condition indicates that the multiple negative sample weight values ​​have a low similarity with the negative sample feature matrix. In this case, a group of target point cloud data corresponding to the target feature vector predicts a flood event.

[0104] Furthermore, determining the similarity between the plurality of positive sample weight values ​​and the positive sample feature matrix may include but is not limited to determining by the following methods:

[0105] S1, assuming that the weight values ​​of multiple positive samples are expressed as [α A , α B …];

[0106] S2, judge [α A , α B ...] whether all weight values ​​satisfy and =1, here refers to [α A , α B …] any weight value α= If both of them hold, then it means that α A satisfy and =1, then continue to judge α B Is it satisfied? and =1, until the weight values ​​of the above multiple positive samples all meet and =1;

[0107] S3, if the weight values ​​of multiple positive samples all satisfy and =1, it means that the similarity between the weight values ​​of the multiple positive samples and the positive sample feature matrix is ​​high, and the weight values ​​of the multiple positive samples meet the first preset condition;

[0108] S4, if at least one positive sample weight value among multiple positive sample weight values ​​does not satisfy and =1, it means that the similarity between the weight values ​​of the multiple positive samples and the above-mentioned positive sample feature matrix is ​​low, and the weight values ​​of the multiple positive samples do not meet the above-mentioned first preset condition.

[0109] Similarly, determining the similarity between the multiple negative sample weight values ​​and the negative sample feature matrix may include but is not limited to being achieved in the same manner as determining the similarity between the multiple positive sample weight values ​​and the positive sample feature matrix.

[0110] Furthermore, if the above-mentioned multiple positive sample weight values ​​meet the above-mentioned first preset condition, and the above-mentioned multiple negative sample weight values ​​meet the above-mentioned second preset condition, the above-mentioned sensitivity parameter can be obtained first, and then the sensitivity parameter can be used to determine whether the above-mentioned flood event has occurred in the above-mentioned target area. The sensitivity parameter determines the degree of strictness in judging a flood event. A high sensitivity parameter makes it easier to determine the occurrence of a flood event.

[0111] Through the embodiments of the present application, it is respectively determined whether the weight values ​​of the above-mentioned multiple positive samples meet the above-mentioned first preset condition, and whether the weight values ​​of the above-mentioned multiple negative samples meet the above-mentioned second preset condition, and the occurrence of a flood event is determined using the preset sensitivity parameters, thereby effectively improving the prediction accuracy of the flood event.

[0112] As an optional solution, when the weight values ​​of the plurality of positive samples satisfy the first preset condition and the weight values ​​of the plurality of negative samples satisfy the second preset condition, determining whether the flood event has occurred in the target area according to a preset sensitivity parameter includes: when the weight values ​​of the plurality of positive samples satisfy the first preset condition and the weight values ​​of the plurality of negative samples satisfy the second preset condition, mapping the positive sample feature matrix and the negative sample feature matrix to a target plane, determining a positive sample convex hull and a negative sample convex hull, wherein the positive sample convex hull represents a region surrounding each positive sample in the positive sample feature matrix. The minimum convex polygon of this feature point, the above-mentioned negative sample convex hull represents the minimum convex polygon surrounding each negative sample feature point in the above-mentioned negative sample feature matrix; determine the first convex hull centroid corresponding to the above-mentioned positive sample convex hull and the second convex hull centroid corresponding to the above-mentioned negative sample convex hull; determine the positive and negative sample two-dimensional dividing line according to the above-mentioned first convex hull centroid and the above-mentioned second convex hull centroid, wherein the above-mentioned positive and negative sample two-dimensional dividing line is used to divide the intersection area of ​​the above-mentioned positive sample convex hull and the above-mentioned negative sample convex hull; use the above-mentioned positive and negative sample two-dimensional dividing line to divide the above-mentioned intersection area, and determine whether the above-mentioned target area has caused the above-mentioned flood event based on the division result and the above-mentioned sensitivity parameter.

[0113] Optionally, in an embodiment of the present application, the above-mentioned target plane includes but is not limited to a two-dimensional plane, the above-mentioned positive sample convex hull refers to the smallest convex polygon in the target plane that contains point cloud data corresponding to the positive sample point cloud data, the negative sample convex hull refers to the smallest convex polygon in the target plane that contains point cloud data corresponding to the negative sample point cloud data, the centroid of the above-mentioned first convex hull represents the geometric center of the positive sample convex hull, and the above-mentioned negative sample convex hull represents the geometric center of the negative sample convex hull.

[0114] For example, Fig.10 is a schematic diagram of another optional flood event detection method according to an embodiment of the present application, such as Fig.10As shown, the target plane includes a positive sample convex hull and a negative sample convex hull. The arrow indicates the direction of the positive and negative samples. The arrow points to the right, indicating that the closer the feature point position is to the right, the more obvious the positive sample feature is, that is, the greater the probability of a flood event. There is an overlapping area between the positive and negative samples. The two-dimensional dividing line between the positive and negative samples can be understood as a threshold alarm line. The left side is the non-alarm area, and the right side of the two-dimensional dividing line between the positive and negative samples is the alarm area, that is, the feature point area corresponding to the flood event. The intersection area of ​​the positive sample convex hull and the negative sample convex hull can be divided by the two-dimensional dividing line between the positive and negative samples to obtain a division result. The division result refers to the possibility of a flood event. If the division result value is greater than the sensitivity parameter, it indicates that a flood event has occurred. Otherwise, it indicates that no flood event has occurred.

[0115] Through the embodiments of the present application, when the weight values ​​of the above-mentioned multiple positive samples satisfy the above-mentioned first preset condition, and the weight values ​​of the above-mentioned multiple negative samples satisfy the above-mentioned second preset condition, the positive sample convex hull and the negative sample convex hull on the target plane can be determined first, and then, based on the positive sample convex hull, the negative sample convex hull, and the sensitivity parameters, it can be judged whether a flood event has occurred. That is, by analyzing the distribution of positive and negative samples by geometric methods, a flood event can be judged more accurately.

[0116] As an optional scheme, the above-mentioned determination of the two-dimensional dividing line between positive and negative samples based on the first convex hull centroid and the second convex hull centroid includes: connecting the first convex hull centroid and the second convex hull centroid to obtain a target line segment; dividing the target line segment according to the above-mentioned sensitivity parameter to determine the target midpoint; determining the above-mentioned two-dimensional dividing line between positive and negative samples with the horizontal coordinate of the above-mentioned target midpoint, wherein the above-mentioned two-dimensional dividing line between positive and negative samples is parallel to the longitudinal axis of the above-mentioned target plane.

[0117] Optionally, in an embodiment of the present application, the target line segment represents a connecting line segment between the centroid of the first convex hull and the centroid of the second convex hull, and the target midpoint is a point on the two-dimensional dividing line of the positive and negative samples.

[0118] Exemplarily, assuming that the centroid of the first convex hull , the centroid of the second convex hull , sensitivity parameter , including but not limited to determining the above two-dimensional dividing line of positive and negative samples by the following methods:

[0119] S1, connect (2,3) and (6,3) to get the target line segment;

[0120] S2, divide the target line segment according to the sensitivity parameter value. The sensitivity parameter is 0.5, which means that the center point of the target line segment is taken as the target midpoint. The target midpoint is expressed as (4, 3).

[0121] S3, determine the two-dimensional dividing line between positive and negative samples as .

[0122] Furthermore, in the target plane, at the two-dimensional boundary between positive and negative samples The feature points on the left will be classified as negative samples (not related to the flood event), while the feature points on the right will be classified as positive samples (related to the flood event). The sensitivity parameters are also adjusted to change the position of the two-dimensional dividing line between positive and negative samples, thereby affecting the above division results.

[0123] As an optional scheme, the above-mentioned intersection area is divided by the above-mentioned two-dimensional dividing line of positive and negative samples, and whether the above-mentioned flood event has occurred in the above-mentioned target area is determined based on the division result and the above-mentioned sensitivity parameter, including: using the above-mentioned two-dimensional dividing line of positive and negative samples to divide the above-mentioned intersection area to obtain a first intersection area and a second intersection area; when the area ratio of the above-mentioned first intersection area and the above-mentioned intersection area is greater than the above-mentioned sensitivity parameter, it is determined that the above-mentioned flood event has occurred in the above-mentioned target area; when the area ratio of the above-mentioned first intersection area and the above-mentioned intersection area is less than or equal to the above-mentioned sensitivity parameter, it is determined that the above-mentioned flood event has not occurred in the above-mentioned target area.

[0124] Optionally, in an embodiment of the present application, the first intersection area refers to an area in the intersection area corresponding to a flood event, and the second intersection area refers to an area in the intersection area corresponding to a flood event that did not occur. For example, the area on the right side of the two-dimensional dividing line between positive and negative samples in the intersection area is the first intersection area, and the area on the left side of the two-dimensional dividing line between positive and negative samples is the second intersection area.

[0125] In an exemplary embodiment, assuming that the area of ​​the first intersection area is 55, the area of ​​the intersection area is 100, and the sensitivity parameter is 0.6, it means that the area ratio of the first intersection area to the intersection area is less than the sensitivity parameter, which means that the target area has not experienced the flood event. Conversely, if the sensitivity parameter is 0.5, it means that the area ratio of the first intersection area to the intersection area is greater than the sensitivity parameter, which means that the target area has experienced a flood event.

[0126] Through the embodiments of the present application, the intersection area is divided using a two-dimensional dividing line of positive and negative samples to obtain a first intersection area and a second intersection area; when the area ratio of the first intersection area and the intersection area is greater than a sensitivity parameter, it is determined that a flood event has occurred in the target area; when the area ratio of the first intersection area and the intersection area is less than or equal to the sensitivity parameter, it is determined that no flood event has occurred in the target area, thereby achieving the purpose of improving the prediction accuracy and sensitivity of flood events.

[0127] As an optional scheme, the above-mentioned determination of multi-frame same-direction point cloud data whose radial velocity satisfies preset conditions from the above-mentioned multi-frame point cloud data includes: obtaining the radial velocity of each point of each frame point cloud data in the above-mentioned multi-frame point cloud data; determining the points whose radial velocity satisfies the above-mentioned preset conditions as data points in the above-mentioned same-direction point cloud data, to obtain the above-mentioned multi-frame same-direction point cloud data; sampling the above-mentioned multi-frame same-direction point cloud data at intervals of frames, and determining multiple groups of target point cloud data and multiple target feature vectors corresponding to the above-mentioned multiple groups of target point cloud data.

[0128] Optionally, in an embodiment of the present application, the above-mentioned preset conditions may include but are not limited to the radial velocity of the above-mentioned data point being greater than or equal to a preset threshold. For example, if it is greater than 0, it means that the moving direction of the data point is the same as the river flow direction, and it is a data point in the above-mentioned same-direction point cloud data.

[0129] Exemplarily, after obtaining the above-mentioned multi-frame same-direction point cloud data, the multi-frame same-direction point cloud data can be sampled at intervals of frames to ensure the representativeness of each group of the above-mentioned target point cloud data. Thus, each target feature vector is obtained using each group of target point cloud data to achieve the purpose of extracting key feature data information from the multi-frame point cloud data, and further determine whether a flood event has occurred in the target area.

[0130] As an optional scheme, the above-mentioned multiple frames of same-direction point cloud data are sampled at intervals of frames to determine multiple groups of target point cloud data and multiple target feature vectors corresponding to the above-mentioned multiple groups of target point cloud data, including: obtaining a predetermined set of tail frame sequence numbers, wherein the tail frame sequence number is used to indicate the target point cloud data corresponding to the last frame in a group of the above-mentioned target point cloud data; sampling at intervals of frames according to the above-mentioned set of tail frame sequence numbers to determine the above-mentioned multiple groups of target point cloud data, wherein a group of the above-mentioned target point cloud data includes the above-mentioned same-direction point cloud data from different frames, and the above-mentioned different frames are not adjacent to each other, and any two data points in a group of the above-mentioned target point cloud data meet a preset distance condition, a preset speed difference condition, and a preset radar reflection area difference condition; the above-mentioned multiple target feature vectors are determined based on the radial velocity mean, radial velocity variance, radar reflection area mean, and radar reflection area variance of each group of the above-mentioned target point cloud data.

[0131] Optionally, in an embodiment of the present application, the above-mentioned group of tail frame sequence numbers includes several time frame sequence numbers with different values, and the above-mentioned preset distance condition refers to that the "ellipse" distance between a data point corresponding to a point cloud data in the above-mentioned group of the above-mentioned target point cloud data and another data point in the group of target point cloud data is less than or equal to a preset distance threshold. Similarly, the above-mentioned preset speed difference condition refers to that the speed difference between a data point corresponding to a point cloud data in the above-mentioned group of the above-mentioned target point cloud data and another data point in the group of target point cloud data is less than or equal to a preset speed difference threshold. The above-mentioned preset radar reflection area difference condition refers to that the radar reflection area difference between a data point corresponding to a point cloud data in the above-mentioned group of the above-mentioned target point cloud data and another data point in the group of target point cloud data is less than or equal to a preset radar reflection area difference threshold.

[0132] Furthermore, after determining the above-mentioned multiple groups of target point cloud data, the radial velocity mean, radial velocity variance, radar reflection area mean and radar reflection area variance can be obtained for each group of point cloud data, and the convex hull of the group of point cloud data can be determined to obtain the above-mentioned target feature vector, which includes the radial velocity mean, radial velocity variance, radar reflection area mean and radar reflection area variance, convex hull and other multi-dimensional data.

[0133] Through the embodiment of the present application, a predetermined set of tail frame numbers are first obtained; frame sampling is performed according to the above set of tail frame numbers to determine the above multiple groups of target point cloud data; based on the radial velocity mean, radial velocity variance, radar reflection area mean and radar reflection area variance of each group of the above target point cloud data, the above multiple target feature vectors are determined to ensure the accuracy and availability of the target feature vectors.

[0134] In an exemplary embodiment, the flood event detection method proposed in the present application can be applied to the application scenario of flood event detection based on millimeter wave radar sensor. Specifically, the millimeter wave radar obtains the measurement information of the moving target with high detection probability and high frame rate. The background target tracking algorithm is based on the millimeter wave radar measurement input information to quickly and accurately carry out business: such as detecting and tracking the real target trajectory, analyzing the target behavior and group target behavior, classifying the target characteristics, confirming the real moving target identity and motion information, timely filtering the false trajectory and terminating the target trajectory, and then outputting the accurate target identity and motion information to other links. The radar has the advantage of being all-weather and all-day, and can work normally regardless of wind and rain during the day and night. Flood events often occur in summer and autumn, and this flood period is generally called the summer flood and autumn flood. Since this kind of flood occurs every year with the arrival of the season, it has obvious periodicity. The flood event detection method proposed in the present application has a high degree of intelligence, does not require human intervention, has strong scene adaptability, high accuracy, and can work all-weather and all-day.

[0135] It should be noted that the measurement information of the target that can be obtained by millimeter-wave radar includes: distance, angle, radial speed (RadialSpeed), radar reflection area (RCS), etc. The millimeter-wave radar can set the measurement frequency / cycle as needed. For example, the common measurement cycle is 0.05 seconds, that is, the operating frequency is 20HZ. The millimeter-wave radar has a very strong ability to detect moving objects and can be used for the detection of slightly moving targets. Because millimeter-wave radar can detect moving targets with tiny speed changes all day and all weather, it is very suitable for flood event monitoring.

[0136] In an exemplary embodiment, Fig.11 is a schematic diagram of another optional flood event detection method according to an embodiment of the present application, such as Fig.11 As shown, including but not limited to:

[0137] S1102, Equipment Installation, The radar equipment is easy to install. It can be installed on a stable rack facing the area to be monitored;

[0138] S1104, parameter adjustment. Set system working time, detection sensitivity, reporting time delay, etc.;

[0139] S1106, data collection, the system operates normally and continuously collects point cloud data;

[0140] S1108, positive and negative sample labeling, pre-labeling data, for example, winter water flow data, the default labeling is "no flood event occurred", and S1110, S1116 are executed at the same time;

[0141] S1110, event detection. The algorithm determines whether a flood has occurred based on the collected data, and if so, proceeds to S1112; if not, proceeds to S1106.

[0142] S1112, notify the duty personnel, and notify the duty personnel in real time through network communication or the like;

[0143] S1114, manual review: on-duty personnel review whether a flood event has occurred through video or field surveys, and mark positive and negative samples. If a flood event has occurred, it is marked as positive sample data, otherwise, it is marked as negative sample data;

[0144] S1116, upload to the cloud, upload PTZ data for sharing with other systems or downstream systems of the river;

[0145] S1118, end.

[0146] Specifically, the principle of using millimeter-wave radar to detect whether a flood has occurred is as follows:

[0147] 1) Millimeter-wave radar has very high sensitivity to moving targets.

[0148] 2) Floods are mixed with a large amount of mud, sand, branches and other objects. As the flood moves from far to near (from near to far), the movement trends of all impurities are consistent, which is suitable for clustering.

[0149] 3) There are significant differences in the target motion characteristics in the corresponding areas before and after the flood event.

[0150] 4) Before and after the flood event, the target intensity information of the corresponding area was significantly different.

[0151] 5) The data features of floods are similar. The flood features in consecutive years are highly similar. The water flow feature vector radar data features corresponding to floods in different sections of the same river are also highly similar. For different rivers, when the flood runoff (the water flow in a river or water body during a flood, usually expressed in cubic meters per second (m³ / s)) is the same, the corresponding point cloud data features are also highly similar.

[0152] Based on this, the core of using millimeter-wave radar to detect flood occurrence is to filter out point cloud data that has a sufficiently long duration and meets the characteristics of flood data from the point cloud data, and then judge whether a flood event has occurred based on this part of the point cloud data, including but not limited to:

[0153] S1, construct water flow feature vector water flow data feature vector;

[0154] S2, water flow normality modeling: statistically calculate the fluctuation range of water flow characteristic vectors (the above positive sample characteristic matrix and negative sample characteristic matrix) and construct confidence intervals;

[0155] S3, collects point cloud data in the recent period through radar;

[0156] S4, calculating the water flow characteristic vector (the above-mentioned target characteristic vector);

[0157] S5, event detection, judging whether the water flow feature vector is within the confidence region;

[0158] S6, determining whether a flood event occurs based on the determination result of S5;

[0159] S7, report flood incidents.

[0160] It should be noted that you can choose to filter or not filter invalid point cloud data. Taking the river as an example, remove the point cloud data of people detected on both sides of the river.

[0161] Specifically, the data structure of the radar point cloud data is shown in Figure 3. The first row indicates the frame number, the number of measurements in the frame, and the measurement time of the frame; after the first row is the specific radar measurement point, the first column is the measurement ID, the second and third columns are the X and Y coordinates (in meters, the coordinate center is usually based on the radar position (radar deployment position) as the origin), the fourth column is RCS (in db, intensity unit, radar cross-section scattering, refers to the echo intensity of the target point to the radar wave), the fifth column is RadialSpeed ​​(in m / s, the speed of the target point in the radial direction (i.e., the direction of the line connecting the radar center)), then, the radar point data is plotted into a graph, Fig.12 It is a schematic diagram of another optional flood event detection method according to an embodiment of the present application, as shown in FIG12 , including data points close to the radar (whose radial velocity is less than zero), data points far away from the radar (whose radial velocity is greater than zero), and stationary data points. The point cloud data can also be grouped. Fig.13 This is a schematic diagram of another optional flood event detection method according to an embodiment of the present application. As shown in Figure 13, the static data points are directly removed and divided into the same-direction point cloud data of the destination (away from the radar) and the reverse point cloud data of the incoming direction (close to the radar).

[0162] In addition, the direction of the river is fixed. Pre-acquiring the river direction helps to accurately cluster and filter out the disturbance of weeds and trees on the water bank. Clustering the point cloud data in the same direction, including but not limited to clustering methods with variable thresholds, can construct a water flow feature vector, that is, construct a feature vector to describe the water flow and environmental characteristics.

[0163] Furthermore, a negative sample data set is constructed. This data set is composed of data when there is no flood. Point cloud data collected in spring, autumn and winter, or data that is not manually corrected most of the time, are all negative sample data. Since the water flow feature vector is high-dimensional spatial data, the water flow feature vector can be described by borrowing 2D spatial data. The set of negative sample points in the space constitutes a convex hull (which can be understood as the maximum area of ​​the horse racing enclosure); similarly, the positive sample data set is the data composed of floods. If a flood occurs in summer, or the data that has been manually annotated and corrected can be regarded as positive sample data, and the set of positive sample points in the space constitutes a convex hull.

[0164] Exemplarily, flood sensitivity can also be set. Different river sections, different areas, different regions, and different artificial dams have different resistance to floods. Therefore, the threshold of flood alarm needs to be set manually according to the actual situation. If the threshold is relatively small (more sensitive), it may trigger many false alarms (heavy rain, river water rises, but does not reach the real flood level); if the threshold is relatively large (relatively slow), when the flood occurs in the early stage, it will not be effectively warned, which may cause the downstream people to have insufficient response time. There are overlapping areas between positive and negative samples. The two-dimensional dividing line between positive and negative samples is the threshold / alarm line. The left side of the two-dimensional dividing line between positive and negative samples is the non-alarm area, and the right side is the alarm area.

[0165] Assume that the sample data feature collected by the radar is the sample data feature water flow feature vector V(k) at the current time k. It is necessary to determine whether it is in the positive sample convex hull or the negative sample convex hull. The judgment method is: the vertices (boundary points) of the negative sample convex hull constitute a point set. If a point V(k) is in the area determined by the negative sample convex hull, it satisfies the following constraints: V(k) = M α; ; ; ; ,in, Represents all data points of the convex hull, α is the positive sample weight value, M is the positive sample feature matrix, that is, only includes the convex hull vertices of the negative samples. Similarly, the convex hull vertices of the positive samples can be used to form a positive sample point set by the same method to determine whether the data point is in the area determined by the convex hull of the positive sample, and then determine whether the "flood event occurs" is true.

[0166] if There is a value Dissatisfied and , it means that the data point is outside the negative sample convex hull. If it is outside the convex hull, it means that it is not a negative sample data, and “no flood event occurred” is not true. The values ​​of and , it means that the data point is in the convex hull. If it is in the convex hull, it means that it is a negative sample data, and “no flood event occurred” is established.

[0167] In an exemplary embodiment, if the data point is within the convex hull of the positive and negative samples at the same time, the high-dimensional geometric center (center of gravity) of the convex hull of the positive and negative samples is calculated respectively, and the center of gravity is mapped on the two-dimensional plane. The data direction is determined by the line connecting the geometric centers of the center of gravity (the arrow indicates the positive and negative samples). Then, the intersection of the convex hulls of the positive and negative samples (the above-mentioned intersection area) is calculated, and based on the data direction and data points, the two-dimensional dividing line of the positive and negative samples is divided into the first intersection area and the second intersection area, and the positive sample area probability ratio (first intersection area / intersection area) and the negative sample area probability ratio (second intersection area / intersection area) are calculated respectively. If the positive sample area probability ratio is greater than the pre-set sensitivity parameter, it is judged that the "flood event" is established.

[0168] Through the embodiments of the present application, a high-accuracy and high-precision flood event prediction method is achieved.

[0169] It also meets the requirements of real-time, all-weather and all-day. On the other hand, the installation requirements of radar equipment are also relatively low.

[0170] It is understandable that in the specific implementation of this application, related data such as user information is involved. When the above embodiments of this application are applied to specific products or technologies, user permission or consent is required, and the collection, use and processing of relevant data need to comply with relevant laws, regulations and standards of relevant countries and regions.

[0171] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application.

[0172] According to another aspect of the embodiments of the present application, a flood event detection device for implementing the above-mentioned flood event detection method is also provided. Fig.14 As shown, the device comprises:

[0173] The detection module 1402 is used to detect the target area using a target radar to determine multiple frames of point cloud data, wherein the detection direction of the target radar is set to be the same as the flow direction of the river in the target area;

[0174] The processing module 1404 is used to determine, from the multiple frames of point cloud data, multiple frames of same-direction point cloud data whose radial velocity meets a preset condition, perform frame-by-frame sampling on the multiple frames of same-direction point cloud data, determine multiple groups of target point cloud data and multiple target feature vectors corresponding to the multiple groups of target point cloud data, wherein one target feature vector is determined by the radial velocity and radar reflection area corresponding to each frame of point cloud data in a group of target point cloud data;

[0175] An execution module 1406 is used to determine a plurality of positive sample weight values ​​according to a plurality of target feature vectors and a predetermined positive sample feature matrix, and to determine a plurality of negative sample weight values ​​according to a plurality of target feature vectors and a predetermined negative sample feature matrix, wherein a positive sample weight value is used to indicate a similarity between a target feature vector and a positive sample feature matrix, and a negative sample weight value is used to indicate a similarity between a target feature vector and a negative sample feature matrix;

[0176] The determination module 1408 is used to determine whether a flood event occurs in the target area according to whether the weight values ​​of the multiple positive samples meet the first preset condition and whether the weight values ​​of the multiple negative samples meet the second preset condition.

[0177] As an optional scheme, the above-mentioned device is used to determine whether the above-mentioned target area has caused a flood event according to whether the above-mentioned multiple positive sample weight values ​​meet the first preset condition and whether the above-mentioned multiple negative sample weight values ​​meet the second preset condition in the following manner: when the above-mentioned multiple positive sample weight values ​​meet the above-mentioned first preset condition and the above-mentioned multiple negative sample weight values ​​do not meet the above-mentioned second preset condition, it is determined that the above-mentioned target area has caused the above-mentioned flood event; when the above-mentioned multiple positive sample weight values ​​meet the above-mentioned first preset condition and the above-mentioned multiple negative sample weight values ​​meet the above-mentioned second preset condition, it is determined whether the above-mentioned target area has caused the above-mentioned flood event according to a pre-set sensitivity parameter.

[0178] As an optional solution, the above-mentioned device is used to determine whether the above-mentioned flood event has occurred in the above-mentioned target area according to a preset sensitivity parameter in the following manner when the above-mentioned multiple positive sample weight values ​​meet the above-mentioned first preset condition and the above-mentioned multiple negative sample weight values ​​meet the above-mentioned second preset condition: when the above-mentioned multiple positive sample weight values ​​meet the above-mentioned first preset condition and the above-mentioned multiple negative sample weight values ​​meet the above-mentioned second preset condition, the above-mentioned positive sample feature matrix and the above-mentioned negative sample feature matrix are mapped to the target plane to determine the positive sample convex hull and the negative sample convex hull, wherein the above-mentioned positive sample convex hull represents the area surrounding each of the above-mentioned positive sample feature matrices. The minimum convex polygon of each positive sample feature point, the negative sample convex hull represents the minimum convex polygon surrounding each negative sample feature point in the negative sample feature matrix; determining the first convex hull centroid corresponding to the positive sample convex hull and the second convex hull centroid corresponding to the negative sample convex hull; determining a positive-negative sample two-dimensional dividing line according to the first convex hull centroid and the second convex hull centroid, wherein the positive-negative sample two-dimensional dividing line is used to divide the intersection area of ​​the positive sample convex hull and the negative sample convex hull; using the positive-negative sample two-dimensional dividing line to divide the intersection area, and determining whether the target area has caused the flood event based on the division result and the sensitivity parameter.

[0179] As an optional scheme, the above-mentioned device is used to determine the two-dimensional dividing line between positive and negative samples according to the above-mentioned first convex hull centroid and the above-mentioned second convex hull centroid in the following manner: connect the above-mentioned first convex hull centroid and the above-mentioned second convex hull centroid to obtain the target line segment; divide the above-mentioned target line segment according to the above-mentioned sensitivity parameter to determine the target midpoint; determine the above-mentioned two-dimensional dividing line between positive and negative samples with the horizontal coordinate of the above-mentioned target midpoint, wherein the above-mentioned two-dimensional dividing line between positive and negative samples is parallel to the longitudinal axis of the above-mentioned target plane.

[0180] As an optional scheme, the above-mentioned device is used to divide the above-mentioned intersection area using the above-mentioned two-dimensional dividing line of positive and negative samples in the following manner, and determine whether the above-mentioned flood event has occurred in the above-mentioned target area based on the division result and the above-mentioned sensitivity parameter: use the above-mentioned two-dimensional dividing line of positive and negative samples to divide the above-mentioned intersection area to obtain a first intersection area and a second intersection area; when the area ratio of the above-mentioned first intersection area and the above-mentioned intersection area is greater than the above-mentioned sensitivity parameter, it is determined that the above-mentioned flood event has occurred in the above-mentioned target area; when the area ratio of the above-mentioned first intersection area and the above-mentioned intersection area is less than or equal to the above-mentioned sensitivity parameter, it is determined that the above-mentioned flood event has not occurred in the above-mentioned target area.

[0181] As an optional scheme, the above-mentioned device is used to determine multi-frame same-direction point cloud data whose radial velocity meets preset conditions from the above-mentioned multi-frame point cloud data in the following manner: obtain the radial velocity of each point of each frame point cloud data in the above-mentioned multi-frame point cloud data; determine the points whose radial velocity meets the above-mentioned preset conditions as data points in the above-mentioned same-direction point cloud data, and obtain the above-mentioned multi-frame same-direction point cloud data; sample the above-mentioned multi-frame same-direction point cloud data at intervals of frames, and determine multiple groups of target point cloud data and multiple target feature vectors corresponding to the above-mentioned multiple groups of target point cloud data.

[0182] As an optional scheme, the above-mentioned device is used to perform frame-by-frame sampling on the above-mentioned multiple frames of same-direction point cloud data in the following manner to determine multiple groups of target point cloud data and multiple target feature vectors corresponding to the above-mentioned multiple groups of target point cloud data: obtain a predetermined set of tail frame sequence numbers, wherein the above-mentioned tail frame sequence number is used to indicate the above-mentioned target point cloud data corresponding to the last frame in a group of the above-mentioned target point cloud data; perform frame-by-frame sampling according to the above-mentioned set of tail frame sequence numbers to determine the above-mentioned multiple groups of target point cloud data, wherein a group of the above-mentioned target point cloud data includes the above-mentioned same-direction point cloud data from different frames, and the above-mentioned different frames are not adjacent to each other, and any two data points in a group of the above-mentioned target point cloud data meet a preset distance condition, a preset speed difference condition, and a preset radar reflection area difference condition; the above-mentioned multiple target feature vectors are determined based on the radial velocity mean, radial velocity variance, radar reflection area mean, and radar reflection area variance of each group of the above-mentioned target point cloud data.

[0183] As an optional scheme, the above-mentioned device is used to determine whether a flood event has occurred in the above-mentioned target area according to whether the above-mentioned multiple positive sample weight values ​​meet the first preset condition and whether the above-mentioned multiple negative sample weight values ​​meet the second preset condition in the following manner: when the above-mentioned multiple positive sample weight values ​​do not meet the above-mentioned first preset condition, and the above-mentioned multiple negative sample weight values ​​meet the above-mentioned second preset condition, it is determined that the above-mentioned target area has not occurred in the above-mentioned flood event; when the above-mentioned multiple positive sample weight values ​​do not meet the above-mentioned first preset condition, and the above-mentioned multiple negative sample weight values ​​do not meet the above-mentioned second preset condition, it is determined that at least one corresponding sample data in the above-mentioned positive sample feature matrix and the above-mentioned negative sample feature matrix is ​​abnormal.

[0184] In the embodiments of the present application, the term "module" or "unit" refers to a computer program or a part of a computer program that has a predetermined function and works together with other related parts to achieve a predetermined goal, and can be implemented in whole or in part by using software, hardware (such as processing circuits or memories), or a combination thereof. Similarly, a processor (or multiple processors or memories) can be used to implement one or more modules or units. In addition, each module or unit can be part of an overall module or unit that includes the function of the module or unit.

[0185] Regarding the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0186] According to one aspect of the present application, a computer program product is provided. The computer program product includes a computer program.

[0187] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0188] Fig.15 The structure block diagram of a computer system for implementing an electronic device according to an embodiment of the present application is schematically shown.

[0189] It should be noted that Fig.15 The computer system 1500 of the electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0190] like Fig.15 As shown, the computer system 1500 includes a central processing unit 1501 (CPU), which can perform various appropriate actions and processes according to the program stored in the read-only memory 1502 (ROM) or the program loaded from the storage part 1508 to the random access memory 1503 (RAM). Various programs and data required for system operation are also stored in the random access memory 1503. The central processing unit 1501, the read-only memory 1502 and the random access memory 1503 are connected to each other through a bus 1504. The input / output interface 1505 (Input / Output interface, i.e., I / O interface) is also connected to the bus 1504.

[0191] The following components are connected to the input / output interface 1505: an input section 1506 including a keyboard, a mouse, etc.; an output section 1507 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 1508 including a hard disk, etc.; and a communication section 1509 including a network interface card such as a LAN card, a modem, etc. The communication section 1509 performs communication processing via a network such as the Internet. A drive 1510 is also connected to the input / output interface 1505 as needed. A removable medium 1511, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 1510 as needed so that a computer program read therefrom is installed into the storage section 1508 as needed.

[0192] In particular, according to an embodiment of the present application, the process described in each method flow chart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a computer readable medium, and the computer program contains a program code for executing the method shown in the flow chart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 1509, and / or installed from the removable medium 1511. When the computer program is executed by the central processor 1501, various functions defined in the system of the present application are executed.

[0193] In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 1509, and / or installed from the removable medium 1511. When the computer program is executed by the central processor 1501, various functions provided by the embodiment of the present application are performed.

[0194] According to another aspect of the embodiment of the present application, an electronic device for implementing the above-mentioned flood event detection method is also provided. The electronic device may be Figure 1 The terminal device or server shown in the figure. This embodiment is described by taking the electronic device as a terminal device as an example. Fig.16 As shown, the electronic device includes a memory 1602 and a processor 1604. The memory 1602 stores a computer program, and the processor 1604 is configured to execute the steps in any of the above method embodiments through the computer program.

[0195] Optionally, in this embodiment, the electronic device may be located in at least one network device among a plurality of network devices of a computer network.

[0196] Optionally, in this embodiment, the above-mentioned processor can be configured to execute the methods in each embodiment of the present application through a computer program.

[0197] Alternatively, a person skilled in the art may understand that: Fig.16 The structure shown is for illustration only. Fig.16 The structure of the electronic device is not limited. Fig.16 More or fewer components (such as network interfaces, etc.) as shown in, or with Fig.16 Different configurations are shown.

[0198] Among them, the memory 1602 can be used to store software programs and modules, such as the program instructions / modules corresponding to the flood event detection method and device in the embodiment of the present application. The processor 1604 executes various functional applications and data processing by running the software programs and modules stored in the memory 1602, that is, realizing the above-mentioned flood event detection method. The memory 1602 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 1602 may further include a memory remotely located relative to the processor 1604, and these remote memories may be connected to the terminal via a network. Examples of the above-mentioned networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof. Among them, the memory 1602 can be specifically, but not limited to, used to store information such as positive sample feature matrices and negative sample feature matrices. As an example, such as Fig.16 As shown, the memory 1602 may include, but is not limited to, the detection module 1402, the processing module 1404, the execution module 1406, and the determination module 1408 in the flood event detection device. In addition, other module units in the flood event detection device may also be included but are not limited to, which will not be repeated in this example.

[0199] Optionally, the transmission device 1606 is used to receive or send data via a network. Specific examples of the network may include a wired network and a wireless network. In one example, the transmission device 1606 includes a network adapter (Network Interface Controller, NIC), which can be connected to other network devices and routers via a network cable so as to communicate with the Internet or a local area network. In one example, the transmission device 1606 is a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0200] In addition, the electronic device further includes: a display 1608 for displaying the prediction results of flood events; and a connection bus 1610 for connecting various module components in the electronic device.

[0201] In other embodiments, the terminal device or server may be a node in a distributed system, wherein the distributed system may be a blockchain system, and the blockchain system may be a distributed system formed by connecting the multiple nodes through network communication. The nodes may form a peer-to-peer network, and any form of computing device, such as a server, terminal or other electronic device, may become a node in the blockchain system by joining the peer-to-peer network.

[0202] According to one aspect of the present application, a computer-readable storage medium is provided, and a processor of an electronic device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the electronic device executes the flood event detection method provided in various optional implementations of the above-mentioned flood event detection aspects.

[0203] Optionally, in this embodiment, the above-mentioned computer-readable storage medium can be configured to store data for executing the methods in various embodiments of the present application.

[0204] Optionally, in this embodiment, a person of ordinary skill in the art may understand that all or part of the steps in the various methods of the above embodiments may be completed by instructing hardware related to the terminal device through a program, and the program may be stored in a computer-readable storage medium, and the storage medium may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0205] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0206] If the integrated units in the above embodiments are implemented in the form of software functional units and sold or used as independent products, they can be stored in the above computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for one or more electronic devices to execute all or part of the steps of the methods described in each embodiment of the present application.

[0207] In the above embodiments of the present application, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0208] In the several embodiments provided in the present application, it should be understood that the disclosed application can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0209] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0210] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0211] The above is only a preferred implementation of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A method for detecting flood events, characterized in that: include: Using a target radar to detect a target area and determine multiple frames of point cloud data, wherein a detection direction of the target radar is set to be the same as a flow direction of a river in the target area; Determine multiple frames of same-direction point cloud data whose radial velocity meets preset conditions from the multiple frames of point cloud data, perform frame-by-frame sampling on the multiple frames of same-direction point cloud data, determine multiple groups of target point cloud data and multiple target feature vectors corresponding to the multiple groups of target point cloud data, wherein one target feature vector is determined by the radial velocity and radar reflection area corresponding to each frame of point cloud data in a group of the target point cloud data; Determine a plurality of positive sample weight values ​​according to the plurality of target feature vectors and a predetermined positive sample feature matrix, and determine a plurality of negative sample weight values ​​according to the plurality of target feature vectors and a predetermined negative sample feature matrix, wherein one of the positive sample weight values ​​is used to indicate the degree of similarity between one of the target feature vectors and the positive sample feature matrix, and one of the negative sample weight values ​​is used to indicate the degree of similarity between one of the target feature vectors and the negative sample feature matrix; Determining whether a flood event occurs in the target area according to whether the weight values ​​of the multiple positive samples meet a first preset condition and whether the weight values ​​of the multiple negative samples meet a second preset condition; The method of determining whether a flood event has occurred in the target area based on whether the multiple positive sample weight values ​​satisfy a first preset condition and whether the multiple negative sample weight values ​​satisfy a second preset condition includes: determining that the flood event has occurred in the target area when the multiple positive sample weight values ​​satisfy the first preset condition and the multiple negative sample weight values ​​do not satisfy the second preset condition; and determining whether the flood event has occurred in the target area based on a preset sensitivity parameter when the multiple positive sample weight values ​​satisfy the first preset condition and the multiple negative sample weight values ​​satisfy the second preset condition.

2. The method according to claim 1, characterized in that The method of determining whether the flood event has occurred in the target area according to a preset sensitivity parameter when the plurality of positive sample weight values ​​satisfy the first preset condition and the plurality of negative sample weight values ​​satisfy the second preset condition includes: In the case where the plurality of positive sample weight values ​​satisfy the first preset condition, and the plurality of negative sample weight values ​​satisfy the second preset condition, mapping the positive sample feature matrix and the negative sample feature matrix to a target plane, and determining a positive sample convex hull and a negative sample convex hull, wherein the positive sample convex hull represents a minimum convex polygon surrounding each positive sample feature point in the positive sample feature matrix, and the negative sample convex hull represents a minimum convex polygon surrounding each negative sample feature point in the negative sample feature matrix; Determine a first convex hull centroid corresponding to the positive sample convex hull and a second convex hull centroid corresponding to the negative sample convex hull; Determining a two-dimensional dividing line between positive and negative samples according to the first convex hull centroid and the second convex hull centroid, wherein the two-dimensional dividing line between positive and negative samples is used to divide the intersection area of ​​the positive sample convex hull and the negative sample convex hull; The intersection area is divided using the positive and negative sample two-dimensional dividing line, and whether the flood event occurs in the target area is determined based on the division result and the sensitivity parameter.

3. The method according to claim 2, characterized in that The determining of the two-dimensional dividing line between positive and negative samples according to the first convex hull centroid and the second convex hull centroid comprises: Connecting the first convex hull centroid and the second convex hull centroid to obtain a target line segment; Divide the target line segment according to the sensitivity parameter to determine the target midpoint; The two-dimensional dividing line between positive and negative samples is determined by the horizontal coordinate of the middle point of the target, wherein the two-dimensional dividing line between positive and negative samples is parallel to the longitudinal axis of the target plane.

4. The method according to claim 2, characterized in that The step of dividing the intersection area by using the positive and negative sample two-dimensional dividing line, and determining whether the flood event has occurred in the target area based on the division result and the sensitivity parameter, comprises: Dividing the intersection region using the positive and negative sample two-dimensional dividing line to obtain a first intersection region and a second intersection region; When the area ratio of the first intersection area to the intersection area is greater than the sensitivity parameter, determining that the flood event has occurred in the target area; When the area ratio of the first intersection area to the intersection area is less than or equal to the sensitivity parameter, it is determined that the flood event has not occurred in the target area.

5. The method according to claim 1, characterized in that: The step of determining, from the multiple frames of point cloud data, multiple frames of same-direction point cloud data whose radial velocity satisfies a preset condition comprises: Acquire the radial velocity of each point of each frame of point cloud data in the multiple frames of point cloud data; Determine the points whose radial velocities satisfy the preset conditions as data points in the same-direction point cloud data, and obtain the multiple frames of same-direction point cloud data; The multiple frames of same-direction point cloud data are sampled at alternate frames to determine multiple groups of target point cloud data and multiple target feature vectors corresponding to the multiple groups of target point cloud data.

6. The method according to claim 1, characterized in that The step of sampling the multiple frames of same-direction point cloud data at intervals to determine multiple groups of target point cloud data and multiple target feature vectors corresponding to the multiple groups of target point cloud data includes: Acquire a predetermined set of tail frame sequence numbers, wherein the tail frame sequence number is used to indicate the target point cloud data corresponding to the last frame in a set of the target point cloud data; Perform frame-by-frame sampling according to the group of tail frame numbers to determine the multiple groups of target point cloud data, wherein a group of the target point cloud data includes the same-direction point cloud data from different frames that are not adjacent to each other, and any two data points in a group of the target point cloud data meet a preset distance condition, a preset speed difference condition, and a preset radar reflection area difference condition; The plurality of target feature vectors are determined based on a radial velocity mean, a radial velocity variance, a radar reflection area mean, and a radar reflection area variance of each group of the target point cloud data.

7. A flood event detection device, characterized in that: include: A detection module, used to detect a target area using a target radar and determine multiple frames of point cloud data, wherein the detection direction of the target radar is set to be the same as the flow direction of a river in the target area; A processing module, used to determine from the multiple frames of point cloud data a plurality of frames of same-direction point cloud data whose radial velocity meets a preset condition, perform frame-by-frame sampling on the multiple frames of same-direction point cloud data, determine a plurality of groups of target point cloud data and a plurality of target feature vectors corresponding to the plurality of groups of target point cloud data, wherein one target feature vector is determined by a radial velocity and a radar reflection area corresponding to each frame of point cloud data in a group of the target point cloud data; An execution module, used to determine a plurality of positive sample weight values ​​according to the plurality of target feature vectors and a predetermined positive sample feature matrix, and to determine a plurality of negative sample weight values ​​according to the plurality of target feature vectors and a predetermined negative sample feature matrix, wherein one of the positive sample weight values ​​is used to indicate a degree of similarity between one of the target feature vectors and the positive sample feature matrix, and one of the negative sample weight values ​​is used to indicate a degree of similarity between one of the target feature vectors and the negative sample feature matrix; A determination module, configured to determine whether a flood event occurs in the target area according to whether the weight values ​​of the multiple positive samples meet a first preset condition and whether the weight values ​​of the multiple negative samples meet a second preset condition; The device is used to determine whether a flood event has occurred in the target area according to whether the multiple positive sample weight values ​​meet a first preset condition and whether the multiple negative sample weight values ​​meet a second preset condition in the following manner: when the multiple positive sample weight values ​​meet the first preset condition and the multiple negative sample weight values ​​do not meet the second preset condition, determine whether the flood event has occurred in the target area; when the multiple positive sample weight values ​​meet the first preset condition and the multiple negative sample weight values ​​meet the second preset condition, determine whether the flood event has occurred in the target area according to a pre-set sensitivity parameter.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored computer program, wherein the computer program can be executed by an electronic device to perform the method described in any one of claims 1 to 6.

9. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method described in any one of claims 1 to 6 are implemented.

10. An electronic device comprising a memory and a processor, characterized in that: A computer program is stored in the memory, and the processor is configured to execute the method according to any one of claims 1 to 6 through the computer program.

Images