Methods and apparatus for determining target data, storage media and electronic devices

Abstract

The present invention discloses a method and device for determining target data, a storage medium and an electronic device, wherein the method comprises: obtaining data to be analyzed, and determining the location information of the target object at multiple moments according to the data to be analyzed; when it is determined that the target object is at a first position at a first moment, determining the target state of the target object at a second position at a second moment according to a preset first threshold and a trajectory prediction model, wherein the first moment is earlier than the second moment; when the target state is an abnormal state, determining multiple target location information at the second moment based on a preset second threshold and a trajectory prediction model; determining the target area in the data to be analyzed according to the multiple target location information, and determining the target data in the target area according to a trained classification model. The above technical solution solves the problem that the traditional detection algorithm cannot detect data with low accuracy.

Description

Technical Field

[0001] The present invention relates to the field of communications, and in particular to a method and device for determining target data, a storage medium and an electronic device. Background Art

[0002] With the rapid development of society, more and more data are involved in social production. How to determine the location of different data from different networks is a key factor in improving data processing efficiency.

[0003] For example, in the field of traffic identification, due to some reasons, different types of scattered objects may appear on the road. In order to prevent the scattered objects from affecting the road traffic, it is necessary to identify the scattered objects on the road.

[0004] In the current existing technology, it can be summarized into three categories of methods. The first category is to give background modeling and foreground extraction, take the target data as the foreground, and then filter according to various filtering conditions. However, this method relies on background modeling and will fail for scenes with changes in data acquisition equipment; the second category is a detection-based solution, which uses various detectors to detect the target data, and then classifies and filters the wrong samples according to various conditions. However, due to the different attributes of the target data, errors are prone to occur and the effect is not good. The third category is a segmentation solution, which determines whether it is the target data one by one at the data coordinate point or pixel point, and then uses the classification model for secondary filtering to retain the real target data and reduce false detection. Although this method solves the problem of morphology, it is a simple and crude segmentation of two categories and then classification. Due to the large number of target data categories, the direct description of one category will not be very good, and the interpretability is also poor. In addition, the segmentation algorithm uses classification as the final filtering method. This method is also relatively simple and crude, and the interpretability is also relatively poor, and the essential characteristics of the target data are not captured.

[0005] Therefore, it is necessary to improve the related technology to overcome the above-mentioned defects in the related technology. Summary of the invention

[0006] The embodiments of the present invention provide a method and device for determining target data, a storage medium and an electronic device, so as to at least solve the problem of low data detection accuracy of traditional detection algorithms.

[0007] According to one aspect of an embodiment of the present invention, a method for determining target data is provided, comprising: acquiring data to be analyzed, and determining position information of a target object at multiple moments based on the data to be analyzed; when it is determined that the target object is at a first position at a first moment, determining a target state in which the target object is at a second position at a second moment based on a preset first threshold and a trajectory prediction model, wherein the first moment is earlier than the second moment; when the target state is an abnormal state, determining multiple target position information at the second moment based on a preset second threshold and the trajectory prediction model; determining a target area in the data to be analyzed based on the multiple target position information, and determining target data in the target area based on a trained classification model.

[0008] According to another aspect of an embodiment of the present invention, a device for determining target data is also provided, including: an acquisition module, used to acquire data to be analyzed, and determine position information of a target object at multiple moments based on the data to be analyzed; a first determination module, used to determine a target state in which the target object is located at a second position at a second moment based on a preset first threshold and a trajectory prediction model when it is determined that the target object is located at a first position at a first moment, wherein the first moment is earlier than the second moment; a second determination module, used to determine multiple target position information at the second moment based on a preset second threshold and the trajectory prediction model when the target state is an abnormal state; and a third determination module, used to determine a target area in the data to be analyzed based on the multiple target position information, and determine the target data in the target area based on a trained classification model.

[0009] According to another aspect of the embodiments of the present invention, 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 method for determining target data when running.

[0010] According to another aspect of an embodiment of the present invention, there is provided an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the method for determining the target data through the computer program.

[0011] Through the present invention, the position information of the target object at multiple moments is determined based on the data to be analyzed. When it is determined that the target object is at the first position at the first moment, the target state of the target object at the second position at the second moment is determined based on the preset first threshold and the trajectory prediction model. When the target state is an abnormal state, multiple target position information at the second moment is determined based on the preset second threshold and the trajectory prediction model. The target area in the data to be analyzed is determined based on the multiple target position information, and the target data in the target area is determined based on the trained classification model. The above technical solution solves the problem of low data detection accuracy of traditional detection algorithms. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0013] Figure 1 It is a hardware structure block diagram of a computer terminal of a method for determining target data according to an embodiment of the present invention;

[0014] Figure 2 is a flowchart of a method for determining target data according to an embodiment of the present invention;

[0015] Figure 3 is a framework diagram of a method for determining target data according to an embodiment of the present invention;

[0016] Figure 4 4 is a structural block diagram of a device for determining target data according to an embodiment of the present invention. DETAILED DESCRIPTION

[0017] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0018] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention 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 interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0019] The method embodiments provided in the embodiments of the present application can be executed in a computer terminal or a similar computing device. Taking running on a computer terminal as an example, Figure 1 FIG. 1 is a hardware structure block diagram of a computer terminal of a method for determining target data in an embodiment of the present invention. Figure 1 As shown, the computer terminal may include one or more ( Figure 1 Only one is shown in the figure) processor 102 (processor 102 may include but is not limited to a microprocessor (Microprocessor Unit, referred to as MPU) or a programmable logic device (Programmable logic device, referred to as PLD)) and a memory 104 for storing data. In an exemplary embodiment, the computer terminal may also include a transmission device 106 and an input and output device 108 for communication functions. It can be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the above-mentioned computer terminal. Figure 1 More or fewer components as shown, or with Figure 1 Equivalent functions or comparisons shown Figure 1 A different configuration with more features is shown.

[0020] The memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as the computer program corresponding to the method for determining target data in the embodiment of the present invention. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, that is, to implement the above method. The memory 104 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 examples, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories may be connected to the computer terminal via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0021] The transmission device 106 is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of a computer terminal. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, referred to as NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0022] In order to solve the above problem, a method for determining target data is provided in this embodiment. Figure 2 is a flow chart of a method for determining target data according to an embodiment of the present invention, the flow chart includes the following steps:

[0023] Step S202: acquiring data to be analyzed, and determining location information of the target object at multiple moments according to the data to be analyzed;

[0024] It should be noted that, in an exemplary embodiment, the data to be analyzed may be image data or other types of data, such as communication data. Specifically, the image data may be image data obtained by a camera recording a road within a preset time.

[0025] In an exemplary embodiment, the target object is a vehicle. The position information of the vehicle at multiple times can be determined from the image data by a vehicle detection algorithm. For example, the vehicle is located at a first position at a first time and at a second position at a second time.

[0026] Step S204: when it is determined that the target object is located at the first position at the first moment, determining a target state where the target object is located at the second position at the second moment according to a preset first threshold and a trajectory prediction model, wherein the first moment is earlier than the second moment;

[0027] In an optional embodiment, before executing step S204, it is necessary to obtain sample data, wherein the sample data contains location information of the target object at multiple moments; the neural network model is trained using the sample data to obtain a trajectory prediction model.

[0028] It should be noted that the sample data is the trajectory information of all vehicles on a road within 24 hours captured by a camera in the absence of spilled objects. The trajectory information of all vehicles can then be input into the trajectory prediction model for training to obtain a trained trajectory prediction model. The trained trajectory prediction model is a neural network model that can predict the probability of a vehicle at any position appearing at a certain position after time T.

[0029] Step S206: when the target state is an abnormal state, determining a plurality of target position information at the second moment based on a preset second threshold and the trajectory prediction model;

[0030] Step S208: determining a target area in the data to be analyzed according to the multiple target position information, and determining target data in the target area according to the trained classification model.

[0031] It should be noted that, in an exemplary embodiment, the target area is a storage area used for data storage, and may also be an image recognition area used for image recognition.

[0032] Through the above steps, the problem that traditional detection algorithms cannot detect serious spilled objects from the data is solved. Further, when the present invention determines that the state of the target object is abnormal, the trajectory prediction model can be used to predict the area where the target object should appear, and then the classification model can be used to determine whether there are spilled objects in the area. In this way, the spilled objects determined by the present invention are all serious spilled objects, and the dredging personnel can quickly clean them up and restore road traffic.

[0033] It should be noted that, in an exemplary embodiment, the above-mentioned target state of the target object being located at the second position at the second moment determined according to a preset first threshold and a trajectory prediction model can be achieved in the following manner: the first position is input into the trajectory prediction model to obtain a target probability set, wherein the target probability set includes position probability information of multiple positions, and the multiple positions include at least: the first position, the second position; the trajectory prediction model is used to predict the probability of the target object being at multiple positions after a preset time, and the preset time is the time difference between the second moment and the first moment; the first probability of the second position is determined according to the target probability set; when the first probability is less than the first threshold, the state of the target object being located at the second position at the second moment is determined to be an abnormal state.

[0034] For better understanding, the following is a specific explanation: Assume that the acquired image data is analyzed by the vehicle detection algorithm, and it is determined that vehicle A is located at position X (first position) at time T (first moment) and at position Y (second position) at time T+t (second moment). Then, the position X of vehicle A at time T is input into the trajectory prediction model, and then the position probability information of vehicle A at multiple positions at time T+t predicted by the trajectory prediction model can be obtained, such as the probability of being at position X is 10%, the probability of being at position Y is 10%, the probability of being at position M is 80%, the probability of being at position N is 70%, and the probability of being at position Z is 90%. Then, according to the position probability information predicted by the trajectory prediction model, it can be determined that the probability of vehicle A being at position Y at time T+t is 10%.

[0035] Since the trajectory prediction model predicts the possible positions of the vehicle under normal driving conditions, if the trajectory prediction model predicts that the probability of vehicle A being at position Y at time T+t is less than 50% (the first threshold), it means that the vehicle is not driving normally, and the vehicle can be determined to be an abnormal vehicle. Since 10% is less than 50%, the state of vehicle A at time T+t is determined to be an abnormal state.

[0036] In an exemplary embodiment, determining multiple target position information at the second moment based on a preset second threshold and the trajectory prediction model can be achieved in the following manner: determining multiple second probabilities in the target probability set whose probability values ​​are greater than or equal to the second threshold; determining the position corresponding to each second probability in the multiple second probabilities as the target position, and obtaining multiple target position information.

[0037] That is to say, if the state of vehicle A at time T+t is abnormal, it means that vehicle A encountered an emergency state from time T to time T+t, causing the vehicle to not travel along the normal route, and then a region where an emergency state exists can be predicted based on the multiple different positions where vehicle A may appear at time T+t predicted by the trajectory prediction model. Specifically, the position corresponding to the probability greater than the second threshold in the position probability information predicted by the trajectory prediction model can be determined as the position where vehicle A may appear at time T+t. Assuming that the target probability set predicted by the trajectory prediction model is: the probability at position X is 10%, the probability at position Y is 10%, the probability at position M is 80%, the probability at position N is 70%, and the probability at position Z is 90%. Assuming that the second threshold is 70%, positions M, N, and Z are determined as the positions where vehicle A may appear at time T+t, but in fact, vehicle A is at position Y at time T+t, which means that there may be obstacles at positions M, N, and Z, and then the spill area can be determined by positions M, N, and Z.

[0038] In an exemplary embodiment, determining the target area in the data to be analyzed based on the multiple target position information can be achieved in the following manner: determining the minimum circumscribed matrix corresponding to the multiple target position information through a target algorithm; and determining the target area in the data to be analyzed corresponding to the minimum circumscribed matrix.

[0039] That is to say, it is necessary to first determine the minimum circumscribed matrix of the positions M, N, and Z, and then determine the target area corresponding to the minimum circumscribed matrix from the data to be analyzed.

[0040] After the target area is determined, it is necessary to determine whether there is target data in the target area, that is, whether there are spilled objects. In an exemplary embodiment, the target data in the target area can be determined based on a trained classification model. Specifically: a trained classification model is obtained, wherein the trained classification model is trained based on multiple groups of data, and the classification model is used to determine the target type of data; the data located in the target area in the data to be analyzed is input into the trained classification model to determine the target data in the target area.

[0041] It should be noted that vehicles do not necessarily detour only when they encounter spilled objects. They may also encounter road construction, traffic accidents, etc., and thus need to use a classification model to determine whether this area is a spilled object area. It should be noted that the classification model is a neural network model that is trained using multiple sets of data.

[0042] It should be noted that the position in the above embodiment may be a pixel point or a set of pixel points in the image.

[0043] Obviously, the above-described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. In order to better understand the above-mentioned method for determining the target data, the above-mentioned process is described below in conjunction with the embodiments, but it is not intended to limit the technical solution of the embodiments of the present invention, specifically:

[0044] In an alternative embodiment, Figure 3 is a framework diagram of a method for determining target data according to an embodiment of the present invention, which can be specifically summarized into the following steps:

[0045] Step 1: Offline statistics of 24-hour traffic information are collected, and the trajectory of the vehicle is obtained using a detection and tracking algorithm (equivalent to the vehicle detection algorithm in the above embodiment). A statistical model of vehicle travel is established using the trajectory (equivalent to the trajectory prediction model in the above embodiment). The statistical model can predict the probability of a vehicle at any position appearing at a certain position after time T.

[0046] Step 2: Offline train a classification model based on a deep neural network. The classification categories include various possible factors that may affect traffic, including accidents, congestion, spilled objects, etc.

[0047] Step 3: In the real-time processing stage, images are obtained from the video stream (equivalent to the data to be analyzed in the above embodiment), and the real-time position information of the vehicle (equivalent to the position information at multiple moments in the above embodiment) is obtained by detection and tracking.

[0048] Step 4: Calculate the posterior probability of the vehicle reaching the location after time T (equivalent to the first probability in the above embodiment), set a threshold tr (equivalent to the first threshold in the above embodiment), and mark vehicles with a value less than the threshold as abnormal vehicles.

[0049] Step 5: Count the positions of the abnormal vehicles before time T where the predicted probability exceeds the threshold tr2 (equivalent to the second threshold in the above embodiment) in the statistical model, and use morphological means (dilation, erosion and connected domain analysis) to obtain the minimum circumscribed rectangle of the area.

[0050] Step 6: Use the minimum enclosing rectangle of the area to obtain the image, input the image into the pre-trained classification model, and output the category of the image.

[0051] Step 7: Determine the objects classified as spilled objects in step 6 as output of spilled objects.

[0052] In addition, the above technical solutions of the embodiments of the present invention solve the problem that the traditional method cannot distinguish whether the area of ​​the spilled objects has reached the point where it affects traffic. The spilled objects that will affect traffic are generally more serious, while the spilled objects that generally do not affect traffic are unimportant. This embodiment can distinguish between the above two types of spilled object events. At the same time, this embodiment utilizes the information of the trajectory of the previous 24 hours, is not easily affected by temporary environmental changes, has high stability, and uses abnormal driving as a trigger condition, which can improve the overall interpretability and improve the effectiveness of reported events. In addition, the performance consumption of using trajectory model sorting is very small, which helps to improve the overall processing efficiency.

[0053] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), including a number of instructions for a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods of various embodiments of the present invention.

[0054] In the present embodiment, a device for determining target data is also provided, and the device is used to implement the above-mentioned embodiment and preferred implementation mode, and the description has been made no further. As used below, the term "module" can implement a combination of software and / or hardware of a predetermined function. Although the equipment described in the following embodiments is preferably implemented in software, the implementation of hardware, or a combination of software and hardware is also possible and conceived.

[0055] Figure 4 is a structural block diagram of a device for determining target data according to an embodiment of the present invention, the device comprising:

[0056] An acquisition module 42 is used to acquire data to be analyzed, and determine the location information of the target object at multiple moments according to the data to be analyzed;

[0057] A first determination module 44 is used to determine, when it is determined that the target object is at a first position at a first moment, a target state where the target object is at a second position at a second moment according to a preset first threshold and a trajectory prediction model, wherein the first moment is earlier than the second moment;

[0058] A second determination module 46, configured to determine, when the target state is an abnormal state, a plurality of target position information at the second moment based on a preset second threshold and the trajectory prediction model;

[0059] The third determination module 48 is used to determine the target area in the data to be analyzed according to the multiple target position information, and determine the target data in the target area according to the trained classification model.

[0060] Through the above-mentioned module, the problem that traditional detection algorithms cannot detect serious spilled objects from the data is solved. Further, when the present invention determines that the state of the target object is abnormal, the trajectory prediction model can be used to predict the area where the target object should appear, and then the classification model can be used to determine whether there are spilled objects in the area. In this way, the spilled objects determined by the present invention are all serious spilled objects, and the dredging personnel can quickly clean them up and restore road traffic.

[0061] Optionally, the first determination module 44 is also used to input the first position into the trajectory prediction model to obtain a target probability set, wherein the target probability set includes position probability information of multiple positions, and the multiple positions include at least: the first position, the second position; the trajectory prediction model is used to predict the probability of the target object being at multiple positions after a preset time, and the preset time is the time difference between the second moment and the first moment; determine the first probability of the second position according to the target probability set; when the first probability is less than the first threshold, determine that the state of the target object being at the second position at the second moment is an abnormal state.

[0062] Optionally, the second determination module 46 is also used to determine multiple second probabilities in the target probability set whose probability values ​​are greater than or equal to the second threshold; determine the position corresponding to each second probability in the multiple second probabilities as the target position, and obtain multiple target position information.

[0063] Optionally, the third determination module 48 is further configured to determine a minimum circumscribed matrix corresponding to the plurality of target position information by a target algorithm; and determine a target region in the data to be analyzed that corresponds to the minimum circumscribed matrix.

[0064] Optionally, the third determination module 48 is also used to obtain a trained classification model, wherein the trained classification model is trained based on multiple groups of data, and the classification model is used to determine the target type of data; the data located in the target area in the data to be analyzed is input into the trained classification model to determine the target data in the target area.

[0065] Optionally, the first determination module 44 is further used to obtain sample data, wherein the sample data contains location information of the target object at multiple moments; and the neural network model is trained by using the sample data to obtain a trajectory prediction model.

[0066] An embodiment of the present invention further provides a computer-readable storage medium, in which a computer program is stored, wherein the computer program is configured to execute the steps of any of the above method embodiments when running.

[0067] Optionally, in this embodiment, the storage medium may be configured to store a computer program for performing the following steps:

[0068] S1, obtaining data to be analyzed, and determining location information of a target object at multiple times according to the data to be analyzed;

[0069] S2, when it is determined that the target object is located at a first position at a first moment, determining a target state where the target object is located at a second position at a second moment according to a preset first threshold and a trajectory prediction model, wherein the first moment is earlier than the second moment;

[0070] S3, when the target state is an abnormal state, determining multiple target position information at the second moment based on a preset second threshold and the trajectory prediction model;

[0071] S4, determining a target area in the data to be analyzed according to the multiple target position information, and determining target data in the target area according to the trained classification model.

[0072] In an exemplary embodiment, the computer-readable storage medium may include, but is not limited to, various media that can store computer programs, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk.

[0073] For specific examples in this embodiment, reference may be made to the examples described in the above embodiments and exemplary implementation modes, and this embodiment will not be described in detail herein.

[0074] An embodiment of the present invention further provides an electronic device, including a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0075] Optionally, in this embodiment, the processor may be configured to perform the following steps through a computer program:

[0076] S1, obtaining data to be analyzed, and determining location information of a target object at multiple times according to the data to be analyzed;

[0077] S2, when it is determined that the target object is located at a first position at a first moment, determining a target state where the target object is located at a second position at a second moment according to a preset first threshold and a trajectory prediction model, wherein the first moment is earlier than the second moment;

[0078] S3, when the target state is an abnormal state, determining multiple target position information at the second moment based on a preset second threshold and the trajectory prediction model;

[0079] S4, determining a target area in the data to be analyzed according to the multiple target position information, and determining target data in the target area according to the trained classification model.

[0080] In an exemplary embodiment, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor, and the input / output device is connected to the processor.

[0081] For specific examples in this embodiment, reference may be made to the examples described in the above embodiments and exemplary implementation modes, and this embodiment will not be described in detail herein.

[0082] Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order than here, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0083] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for determining target data, characterized in that: include: Acquire data to be analyzed, and determine location information of the target object at multiple times according to the data to be analyzed; When it is determined that the target object is located at a first position at a first moment, determining a target state in which the target object is located at a second position at a second moment according to a preset first threshold and a trajectory prediction model, wherein the first moment is earlier than the second moment; When the target state is an abnormal state, determining a plurality of target position information at the second moment based on a preset second threshold and the trajectory prediction model; Determine a target area in the data to be analyzed according to the multiple target position information, and determine target data in the target area according to the trained classification model; Wherein, determining the target state of the target object being located at the second position at the second moment according to a preset first threshold and a trajectory prediction model includes: inputting the first position into the trajectory prediction model to obtain a target probability set, wherein the target probability set includes position probability information of multiple positions, and the multiple positions include at least: the first position and the second position; the trajectory prediction model is used to predict the probability of the target object being at multiple positions after a preset time, and the preset time is the time difference between the second moment and the first moment; determining the first probability of the second position according to the target probability set; when the first probability is less than the first threshold, determining that the state of the target object being located at the second position at the second moment is an abnormal state; Among them, based on the preset second threshold and the trajectory prediction model, multiple target position information at the second moment is determined, including: determining multiple second probabilities in the target probability set whose probability values ​​are greater than or equal to the second threshold; determining the position corresponding to each second probability in the multiple second probabilities as the target position, and obtaining multiple target position information.

2. The method according to claim 1, characterized in that: Determining a target area in the data to be analyzed according to the plurality of target position information includes: Determine the minimum circumscribed matrix corresponding to the plurality of target position information by a target algorithm; A target region corresponding to the minimum circumscribed matrix in the data to be analyzed is determined.

3. The method according to claim 1, characterized in that Determining the target data in the target area according to the trained classification model includes: Obtaining a trained classification model, wherein the trained classification model is obtained by training based on multiple groups of data, and the classification model is used to determine data of a target type; The data located in the target area among the data to be analyzed is input into the trained classification model to determine the target data in the target area.

4. The method according to claim 1, characterized in that: Before determining, according to a preset first threshold and a trajectory prediction model, that the target object is in a target state at a second position at a second moment, the method further includes: Acquire sample data, wherein the sample data contains location information of the target object at multiple moments; The neural network model is trained using the sample data to obtain a trajectory prediction model.

5. A device for determining target data, characterized in that: include: An acquisition module, used to acquire data to be analyzed, and determine location information of a target object at multiple times according to the data to be analyzed; A first determination module is used to determine, when it is determined that the target object is at a first position at a first moment, a target state where the target object is at a second position at a second moment according to a preset first threshold and a trajectory prediction model, wherein the first moment is earlier than the second moment; A second determination module, configured to determine, when the target state is an abnormal state, a plurality of target position information at the second moment based on a preset second threshold and the trajectory prediction model; A third determination module is used to determine a target area in the data to be analyzed according to the multiple target position information, and determine target data in the target area according to the trained classification model; Wherein, the first determination module is further used to input the first position into the trajectory prediction model to obtain a target probability set, wherein the target probability set includes position probability information of multiple positions, and the multiple positions include at least: the first position and the second position; the trajectory prediction model is used to predict the probability of the target object being at multiple positions after a preset time, and the preset time is the time difference between the second moment and the first moment; determine the first probability of the second position according to the target probability set; when the first probability is less than the first threshold, determine that the state of the target object being at the second position at the second moment is an abnormal state; Among them, the second determination module is also used to determine multiple second probabilities in the target probability set whose probability values ​​are greater than or equal to the second threshold; determine the position corresponding to each second probability in the multiple second probabilities as the target position, and obtain multiple target position information.

6. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein the program executes the method described in any one of claims 1 to 4 when executed.

7. 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 4 through the computer program.

Images