Position monitoring method, device and equipment and computer readable storage medium

Abstract

The invention discloses a position monitoring method, device and equipment and a computer readable storage medium. Specifically, the current posture of a target object can be determined; determining a track state corresponding to the target object according to the current posture of the target object; and if the track state is a locking state, determining the latest historical position of the target object as the current position of the target object. Therefore, according to the scheme, whether the track state of the target object is locked or not can be determined by performing posture recognition on the target object, then the position of the target object can be accurately tracked, and the position information of the target object can be updated in real time. In this way, the situation that the target object is lost or deviated in position is avoided, the position of the target object can be rapidly and accurately obtained, and the user experience is improved.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a location monitoring method, device, equipment and computer-readable storage medium. Background Art

[0002] With the continuous development of smart home, smart home services have been widely used in people's homes. The implementation of smart home services requires tracking the user's location in the home space, and millimeter wave radar is a non-contact sensor that can well detect the user's location in the home space. Related technologies can combine millimeter wave radar with deep learning models to identify the coordinates of the user in the home space, thereby helping smart home services to track the user's location.

[0003] However, although the related technology can track the user's location, when the user is stationary or blocked by obstacles in the space, the related technology is likely to lose the user's movement trajectory and cannot locate the user, resulting in the smart home being unable to accurately provide services to the user. Summary of the invention

[0004] The embodiments of the present application provide a location monitoring method, apparatus, device, and computer-readable storage medium, which can accurately determine the location of a target object.

[0005] In a first aspect, an embodiment of the present application provides a location monitoring method, including:

[0006] Determine the current pose of the target object;

[0007] Determining a trajectory state corresponding to the target object according to the current posture of the target object;

[0008] If the track state is a locked state, the latest historical position of the target object is determined as the current position of the target object.

[0009] In a second aspect, an embodiment of the present application provides a position monitoring device, including:

[0010] A first determining unit, used to determine a current posture of the target object;

[0011] A second determining unit, configured to determine a trajectory state corresponding to the target object according to a current posture of the target object;

[0012] The third determining unit is configured to determine the latest historical position of the target object as the current position of the target object if the track state is a locked state.

[0013] In some implementations, the first determining unit is specifically configured to:

[0014] Determine a plurality of point cloud information corresponding to the target object from a plurality of point cloud data frames corresponding to the target space, wherein the plurality of point cloud data frames correspond to the plurality of point cloud information one by one;

[0015] The current posture of the target object is determined according to the plurality of point cloud information.

[0016] In some implementations, the plurality of point cloud information includes current frame point cloud information and a preset number of historical frame point cloud information, and the first determination unit determines the current posture of the target object according to the plurality of point cloud information, specifically for:

[0017] Constructing a posture feature vector of the target object according to the point cloud information of the current frame and the point clouds in the point cloud information of the historical frame;

[0018] The posture feature vector is classified using a preset classifier to obtain the current posture of the target object.

[0019] In some implementations, the first determining unit constructs a posture feature vector of the target object based on the point cloud information of the current frame and the point cloud in the historical frame point cloud information, specifically for:

[0020] Calculate the intra-frame feature vector of the current frame point cloud information and the intra-frame feature vector of the historical frame point cloud information according to the point clouds in the current frame point cloud information and the historical frame point cloud information;

[0021] According to the point clouds in the current frame point cloud information and the historical frame point cloud information, calculating a first inter-frame feature vector between the current frame point cloud information and the historical frame point cloud information, and a second inter-frame feature vector between each of the historical frame point cloud information;

[0022] The intra-frame feature vector, the first inter-frame feature vector, and the second inter-frame feature vector are concatenated to obtain a posture feature vector of the target object.

[0023] In some implementations, the posture includes a standing state and a non-standing state, and the second determining unit is specifically configured to:

[0024] If the current posture is a standing state, determining that the trajectory state corresponding to the target object is an unlocked state;

[0025] If the current posture is a non-standing state, it is determined that the trajectory state corresponding to the target object is a locked state.

[0026] In some embodiments, the position monitoring device further comprises:

[0027] The fourth determining unit is configured to determine the current position of the target object according to the point cloud information corresponding to the target object if the trajectory state is the unlocked state.

[0028] In some embodiments, the position monitoring device further comprises:

[0029] The sending unit is used to send indication information indicating that the track state is a locked state, so that the client continues to display the identifier of the target object at the current position.

[0030] In a third aspect, an embodiment of the present application provides a location monitoring method, including:

[0031] Display the monitoring page of the target space;

[0032] Receiving indication information for indicating that the state of the trajectory corresponding to the target object is a locked state;

[0033] The identification of the target object is continuously displayed at a position corresponding to the current position of the target object in the monitoring page according to the indication information.

[0034] In a fourth aspect, an embodiment of the present application provides a position monitoring device, including:

[0035] A display unit, used to display a monitoring page of the target space;

[0036] A receiving unit, used to receive indication information indicating that the state of the trajectory corresponding to the target object is a locked state;

[0037] The display unit is further configured to continuously display the identifier of the target object at a position corresponding to the current position of the target object in the monitoring page according to the indication information.

[0038] In a fifth aspect, an embodiment of the present application discloses a computer device, including a processor and a memory, wherein the memory stores a computer program, and the processor is used to run the computer program in the memory to execute the method disclosed in the first aspect or the third aspect above.

[0039] In a sixth aspect, an embodiment of the present application discloses a computer-readable storage medium, which stores a plurality of instructions suitable for loading by a processor to execute the method disclosed in the first or third aspect above.

[0040] The embodiment of the present application can determine the current posture of the target object; according to the current posture of the target object, determine the corresponding trajectory state of the target object; if the trajectory state is locked, the latest historical position of the target object is determined as the current position of the target object. It can be seen that this solution can determine whether the trajectory state is locked by performing posture recognition on the target object, and then accurately track the position of the target object and update the position information of the target object in real time. In this way, the loss or position shift of the target object is avoided, and the position of the target object can be obtained quickly and accurately, which improves the user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0042] Figure 1 is a schematic diagram of a scenario of a position monitoring system provided in an embodiment of the present application;

[0043] Figure 2 A flow chart of a location monitoring method provided in an embodiment of the present application;

[0044] Figure 3 A schematic diagram of a process for target object posture recognition provided in an embodiment of the present application;

[0045] Figure 4 A schematic diagram of a flow chart of another location monitoring method provided in an embodiment of the present application;

[0046] Figure 5 A schematic diagram of information transmission of a location monitoring method provided in an embodiment of the present application;

[0047] Figure 6 A flowchart of another location monitoring method provided in an embodiment of the present application;

[0048] Figure 7 is a structural schematic diagram of a position monitoring device provided in an embodiment of the present application;

[0049] Figure 8 is a schematic structural diagram of another position monitoring device provided in an embodiment of the present application;

[0050] Fig. 9 It is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0051] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions 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 those skilled in the art without creative work are within the scope of protection of this application.

[0052] The embodiments of the present application provide a location monitoring method, apparatus, device and computer-readable storage medium. Specifically, the embodiments of the present application will be described from the perspective of a location monitoring system, which may include a data processing terminal and a client 30. The data processing terminal may be as follows: Figure 1 The server 10 and / or detection device 20 shown. The server 10 may be an independent physical server, a server cluster or a distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. The client 30 may be a smart phone, a smart control panel, a tablet computer, a laptop computer, a desktop computer, a smart speaker, a smart watch, a smart home appliance, a vehicle terminal, an intelligent voice interaction device, an aircraft, etc., but is not limited thereto.

[0053] The location monitoring method provided in the embodiments of the present application can be applied to various application scenarios, such as, but not limited to, application scenarios of location monitoring of home spaces, which are specifically described by the following embodiments:

[0054] For example, see Figure 1 , is a schematic diagram of a location monitoring system provided in an embodiment of the present application. The system may include a client and a data processing end. Information can be transmitted between the client and the data processing end. The specific implementation scenario may be as follows:

[0055] The client can be used to: display a monitoring page of the target space; obtain the current position of the target object; and display the identifier of the target object at the current position in the monitoring page. Wherein, obtaining the current position of the target object includes: receiving indication information indicating that the corresponding trajectory state of the target object is a locked state; and displaying the identifier of the target object at a position corresponding to the current position in the monitoring page includes: continuously displaying the identifier of the target object at a position corresponding to the current position in the monitoring page according to the indication information.

[0056] The data processing end may be a detection device or a server, or a combination of a detection device and a server. The detection device may be a device having a function of position detection, such as a human body sensor, a millimeter wave radar, a camera, or the like.

[0057] Specifically, the data processing end can be used to: determine the current posture of the target object; determine the trajectory state corresponding to the target object according to the current posture of the target object; if the trajectory state is a locked state, determine the latest historical position of the target object as the current position of the target object.

[0058] It should be noted that the order of the following embodiments is not intended to limit the preferred order of the embodiments.

[0059] See also Figure 2 , Figure 2 A flow chart of a location monitoring method provided in an embodiment of the present application, which is applied to Figure 1 The specific process of data processing in is as follows:

[0060] 101. Determine the current posture of the target object.

[0061] The target object may be an object that can move in the target space, for example, a human body.

[0062] Posture can be a term that describes the posture or action state of an object or human body in space. It can include information about the position, direction, angle, posture, etc. of the body or object. In target posture recognition, posture usually refers to the physical state of the human body, such as standing, sitting, lying, etc.

[0063] Specifically, a camera or a depth sensor can be used to collect images or point cloud data of the target object. These data are used to capture the characteristics and posture changes of the target object. Secondly, the collected images or point cloud data are input into the posture recognition algorithm. The algorithm can learn the posture characteristics of the target object by using machine learning or deep learning technology, and compare it with the pre-stored posture model. Then, by comparing the characteristics and the model, the current posture of the target object can be determined. In addition, the posture can be assisted by the environmental information around the target object. For example, a sensor is used to obtain information such as light, temperature, gravity, etc. around the target object, and this information is fused with the image or point cloud data of the target object to improve the accuracy of posture recognition. In this way, by collecting the image or point cloud data of the target object and combining the posture recognition algorithm and environmental information, the current posture of the target object can be determined. This method can assist in target tracking and positioning, thereby ensuring its accuracy and preventing the target from being lost or being disturbed. It is worth noting that suitable sensors and algorithms can be selected according to specific needs, and the algorithm can be further optimized for the process of determining the current posture of the target object to ensure the accuracy and real-time performance of posture recognition.

[0064] In some implementations, the current posture of the target object may be determined based on a plurality of point cloud data frames corresponding to the target space. For example, step 101 may include:

[0065] (101.1) determining a plurality of point cloud information corresponding to the target object from a plurality of point cloud data frames corresponding to the target space, wherein the plurality of point cloud data frames correspond to the plurality of point cloud information in a one-to-one manner;

[0066] (101.2) Determine a current posture of the target object based on the multiple point cloud information.

[0067] The target space may refer to a specific area or scene used to identify, track, and monitor target objects.

[0068] A point cloud data frame refers to a frame of data collected in three-dimensional space that contains one or more point cloud data. Each point cloud data contains the three-dimensional coordinates of a point and other attribute information (such as reflection intensity, normal vector, signal-to-noise ratio, etc.), and a point cloud data frame is a collection of one or more point cloud data. By processing and analyzing the point cloud data frame, the posture, position and other information of the target object can be obtained.

[0069] Point cloud information may refer to point cloud data of a target object contained in a plurality of point cloud data frames acquired from a target space. Point cloud information may include current frame point cloud information of the target object and a preset number of historical frame point cloud information. Point cloud information is used to determine the current posture and position of the target object, and to construct a posture feature vector and an inter-frame feature vector of the target object.

[0070] Specifically, in order to determine the current posture of the target object, the collected multiple point cloud data frames can be matched with the target object in the target space to ensure that each point cloud data frame can accurately reflect the information of the target object. These point cloud data frames can be obtained through sensors or radar equipment, and each point cloud data frame can be matched with the target object through data processing and matching algorithms. In this way, multiple point cloud information of the target object can be obtained.

[0071] Exemplarily, the radar device can be placed at a suitable position in the target space so that it can perform effective position monitoring and point cloud data collection. After the radar device starts working, it sends out a series of microwave signals. After the microwave signal interacts with the target object, an echo signal is generated. The radar device receives and processes the echo signal, determines the point cloud data frame, and thus obtains information such as the position, distance, speed, and angle of the target object. The radar device can extract the point cloud information of the target object. The point cloud information of the target object may include relevant information such as its position, shape, and contour. The radar device processes and analyzes the point cloud information of the target object through data transmission. In this way, the corresponding point cloud information of the target object in the target space can be obtained through the radar device, and this information can be used for subsequent analysis and processing.

[0072] Next, the current posture of the target object is determined based on the multiple point cloud information. The posture of the target object can be constructed by analyzing the point cloud information and building an algorithm based on a preset posture. The point cloud data can also be classified and identified using a machine learning algorithm. By training the model, the point cloud information is associated with different postures, so that the current posture of the target object can be determined based on the point cloud information. In addition, the posture estimation algorithm can be used to infer the posture of the target object based on the point cloud information and the shape, position and other information of the target object. The posture estimation algorithm can analyze and deduce based on the geometric features and positional relationships of the point cloud information to derive the current posture of the target object. It is worth noting that different methods are suitable for different scenarios and needs, and the appropriate method can be selected for posture determination according to the specific situation.

[0073] In some embodiments, the multiple point cloud information includes current frame point cloud information and a preset number of historical frame point cloud information. The posture feature vector of the target object can be constructed based on the current frame point cloud information and the historical frame point cloud information to determine the current posture of the target object.

[0074] For example, step (101.2) may include:

[0075] (101.2.1) constructing a posture feature vector of the target object based on the point cloud information of the current frame and the point cloud information of the historical frame;

[0076] (101.2.2) Use the preset classifier to classify the posture feature vector to obtain the current posture of the target object

[0077] The current frame point cloud information may refer to the point cloud data acquired from the sensor at the current time point. The current frame point cloud information includes the three-dimensional spatial distribution information of the object or scene at the current time point, and can be used for tasks such as target recognition and posture prediction.

[0078] The historical frame point cloud information may refer to point cloud data acquired in the past period of time, including point cloud data of multiple historical frames. In gesture recognition, the historical frame point cloud information is used to construct a gesture feature vector of the target object to help determine the current gesture and trajectory state of the target.

[0079] Specifically, the point clouds in the current frame point cloud information and the historical frame point cloud information can be extracted and processed to obtain relevant information of the target object. For example, the point cloud data can be processed using a deep learning algorithm to extract features such as the coordinates and speed of the point cloud. Then, based on these extracted features, a posture feature vector of the target object is constructed. The point cloud information in the current frame point cloud information and the point cloud in the historical frame point cloud information can be compared, the difference between them can be calculated, and the difference information can be used as part of the posture feature vector. For example, the change in the number of point clouds, the change in coordinates, the change in speed, etc. can be calculated as a component of the posture feature vector.

[0080] After the posture feature vector of the target object is constructed, the constructed posture feature vector can be classified using a preset classifier to obtain the current posture of the target object. A machine learning algorithm, such as an adaptive boosting classifier (AdaptiveBoosting, adaboost classifier), can be used for classification. It is worth noting that the preset classifier needs to be trained with a training data set in advance so that it can accurately classify the posture feature vector. Then, the constructed posture feature vector is classified, and the current posture of the target object is judged according to the classification result. It should be noted that during the implementation process, it can be optimized and adjusted according to the needs of the specific application scenario. For example, the construction method of the feature vector can be adjusted, and certain features can be added or deleted to improve the accuracy and efficiency of the classification. At the same time, according to the specific real-time monitoring needs, a suitable classification threshold can be set to reduce misjudgment and improve the stability of the system. In summary, the posture feature vector of the target object is constructed according to the point cloud in the current frame point cloud information and the historical frame point cloud information, and the posture feature vector is classified using a preset classifier to obtain the current posture of the target object. This implementation method can effectively identify and monitor the posture changes of the target object, and provide important auxiliary information for position monitoring and target tracking.

[0081] In some implementations, the posture feature vector of the target object may be determined based on the intra-frame feature vector and the inter-frame feature vector between the current frame point cloud information and the historical frame point cloud information.

[0082] For example, step (101.2.1) may include:

[0083] (101.2.1.1) calculating, based on the point clouds in the current frame point cloud information and the historical frame point cloud information, an intra-frame feature vector of the current frame point cloud information and an intra-frame feature vector of the historical frame point cloud information;

[0084] (101.2.1.2) calculating, based on the point clouds in the current frame point cloud information and the historical frame point cloud information, a first inter-frame feature vector between the current frame point cloud information and the historical frame point cloud information, and a second inter-frame feature vector between each historical frame point cloud information;

[0085] (101.2.1.3) Concatenate the intra-frame feature vector, the first inter-frame feature vector, and the second inter-frame feature vector to obtain a posture feature vector of the target object.

[0086] The intra-frame feature vector can refer to the vector describing the internal features of the point cloud obtained by processing and calculating the point cloud data in the same frame of point cloud information. The intra-frame feature vector can be used to represent the posture state of the target object in the current frame, including information such as the number of point clouds, coordinates, and speed. By comparing the changes in the feature vectors in different frames, the posture of the target object can be identified and tracked.

[0087] The inter-frame feature vector refers to the feature vector obtained by calculating the point cloud information between different frames. In gesture recognition, the frame point cloud can be compared with the historical frame point cloud, and the difference between them can be calculated to obtain the inter-frame feature vector. The inter-frame feature vector can reflect the change of the target object's posture and is used to determine the state and position of the target object. Among them, the first inter-frame feature vector is the inter-frame feature vector between the current frame point cloud information and the historical frame point cloud information; the second inter-frame feature vector is the inter-frame feature vector between the historical frame point cloud information.

[0088] According to the point clouds in the current frame point cloud information and the historical frame point cloud information, the intra-frame feature vector of the current frame point cloud information and the intra-frame feature vector of the historical frame point cloud information are calculated. Specifically, the current frame point cloud data can be clustered first to obtain a clustered point cloud block. Then, the features of each point cloud can be extracted from each point cloud block, such as distance, shape, intensity and other indicators. According to the features of the point cloud, the extracted features can be combined into an intra-frame feature vector of the current frame point cloud information. For example, the point clouds in the point cloud block can be sorted from strong to weak according to the signal strength, and a preset number of point clouds can be subtracted from each other to obtain an intra-frame feature vector. Similarly, the point cloud in the historical frame point cloud information can be processed in the same way to obtain an intra-frame feature vector of the historical frame point cloud information.

[0089] Furthermore, based on the point clouds in the current frame point cloud information and the historical frame point cloud information, the first inter-frame feature vector between the current frame point cloud information and the historical frame point cloud information, and the second inter-frame feature vector between each historical frame point cloud information are calculated. Specifically, the point cloud data between the current frame point cloud information and the historical frame point cloud information can be subtracted from each other to obtain an inter-frame feature vector, i.e., a first inter-frame feature vector, and the point cloud data between different historical frame point cloud information can be subtracted from each other to obtain an inter-frame feature vector, i.e., a second inter-frame feature vector.

[0090] Finally, the intra-frame feature vector, the first inter-frame feature vector, and the second inter-frame feature vector are spliced ​​together to obtain the posture feature vector of the target object. Specifically, the intra-frame feature vector, the first inter-frame feature vector, and the second inter-frame feature vector can be arranged in order. Then, the arranged feature vectors can be spliced ​​together to form the posture feature vector of the target object. In this way, the posture feature vector of the target object can be calculated and spliced ​​according to the point cloud information of the current frame and the point cloud information of the historical frame, and the posture feature vector can be classified using a preset classifier, thereby realizing posture recognition and tracking and positioning of the target.

[0091] For ease of understanding, this solution provides a calculation process for a posture feature vector, the specific process is as follows:

[0092] 1. Use the density-based clustering method (Density-Based Spatial Clustering of Application with Noise, DBSCAN algorithm clustering) for the current frame point cloud data to obtain multiple clustered point cloud blocks n1, n2, n3, ..., n N1 ;

[0093] 2. There are N trajectories (one trajectory corresponds to one target object). For the kth (k=1,2,...N)th trajectory, find the point cloud block n that is closest to it and whose center is less than 1 meter from the trajectory coordinates from multiple point cloud blocks. i ;

[0094] 3. Point cloud block n i The structure features of the point cloud data pc1, pc2, pc3, pc4, pc5, pc6, pc7 associated with the first seven frames of trajectory k are:

[0095] a. Extract 10 point clouds from pc1 according to the signal strength from strong to weak (if there are no 10 point clouds in pc1, copy them to 10 according to the signal strength from strong to weak), and subtract the 10 point cloud data from each other to obtain the intra-frame feature vector f1;

[0096] b. Same as step a, from pc2, pc3, pc4, pc5, pc6, pc7, n iTake 10 point clouds respectively to obtain the intra-frame feature vectors f2, f3, f4, f5, f6, f7, f8;

[0097] c. Block point cloud n i The 10 point cloud data selected above are sequentially subtracted from the 10 point cloud data in pc4 to obtain the first inter-frame feature vector f9;

[0098] d. Subtract the 10 point cloud data in pc7 from the 10 point cloud data in pc3 in turn to obtain the second inter-frame feature vector f 10 ;

[0099] e. Subtract the 10 point cloud data in pc6 from the 10 point cloud data in pc2 in turn to obtain the second inter-frame feature vector f 11 ;

[0100] f. Subtract the 10 point cloud data in pc5 from the 10 point cloud data in pc1 in turn to obtain the second inter-frame feature vector f 12 ;

[0101] g. Substitute the feature vectors f1,f2,...,f 12 They are concatenated to form the posture feature vector.

[0102] It should be understood that the calculation process of the posture feature vector provided above is only an exemplary description of the calculation of the posture feature vector, and the specific calculation process of the posture feature vector may vary according to the specific situation and needs to be specifically designed and implemented according to actual needs. For example, 20 point cloud data are extracted from strong to weak according to the signal strength.

[0103] In this way, the current posture of the target object can be determined. In this way, the accuracy of the target tracking system in locating the target can be improved. The posture information can help the system to more accurately determine the target position and movement state, thereby reducing tracking errors and deviations, preventing the target from being lost or being carried away by other interference, and improving the stability of target positioning.

[0104] 102. Determine a trajectory state corresponding to the target object according to the current posture of the target object.

[0105] The track state can refer to the state of the target object at a specific point in time. The track state can be divided into a locked state and an unlocked state. When the target track is locked, the coordinates of the track remain unchanged, that is, the target is considered to be in the current position until the unlocked state; when the target track is in the unlocked state, the coordinates of the subsequent track can change according to the changes in the associated point cloud. The specific track state depends on the posture of the target object and other state changes.

[0106] After determining the current posture of the target object, the corresponding trajectory state can be determined according to the current posture of the target object. Specifically, the current posture of the target object can be classified using a pre-trained posture classifier, such as standing, sitting, lying, etc. The trajectory state of the target object can be determined by the classification result. It is also possible to extract posture-related features such as point cloud quantity, coordinates, speed changes, etc. according to the feature extraction algorithm of the current posture of the target object. The trajectory state of the target object can be determined by comparing the feature vector. The trajectory state corresponding to the target object can also be determined by observing the change law of the target object's posture. If the posture change of the target object is continuous and conforms to a certain pattern, the trajectory state of the target object can be inferred according to this law. For example, if the target changes from a standing state to a sitting state and then to a lying state, it can be determined that the trajectory of the target object is stationary or moving. In general, the specific implementation method of determining the trajectory state corresponding to the target object according to the current posture of the target object mainly involves posture recognition technology and logical judgment of determining the trajectory state according to the posture. Through reasonable posture recognition and state judgment, the tracking and positioning accuracy of the target object can be effectively improved, and the occurrence of target loss or offset can be avoided.

[0107] In some implementations, the posture includes a standing state and a non-standing state, and the trajectory state of the target object can be determined by determining whether the posture of the target object belongs to the standing state or the non-standing state.

[0108] For example, step (102) may include:

[0109] If the current posture is a standing state, it is determined that the trajectory state corresponding to the target object is an unlocked state;

[0110] If the current posture is a non-standing state, it is determined that the trajectory state corresponding to the target object is a locked state.

[0111] The standing state may refer to a state in which a person or object exists in an upright posture, that is, the upper part of the body is perpendicular to the ground and there is no other support. The non-standing state may refer to other posture states other than the standing state, such as sitting, lying, and other states.

[0112] Specifically, when the current posture is a standing state, the trajectory state corresponding to the target object is determined to be an unlocked state. In the unlocked state, it means that the target object can move freely, and the coordinates of the subsequent trajectory can be updated according to the changes in the associated point cloud. On the contrary, when the current posture of the target object is a non-standing state (such as sitting, lying, and other postures), it can be determined that the trajectory state corresponding to the target object is a locked state. The locked state means that the trajectory of the target object remains unchanged for a period of time until it is unlocked. This can prevent the trajectory of the target from changing due to posture changes and improve the accuracy of target tracking. It is worth noting that the unlocking condition can be judged based on the posture change of the target. For example, if a person changes from a sitting position to a lying position, the trajectory state of the target object can be locked to the state corresponding to the lying position until a new posture change occurs.

[0113] In this way, the corresponding trajectory state can be determined according to the current posture of the target object, which can help achieve accurate tracking and position monitoring of the target object. This method can be applied to various scenarios that require tracking the target posture and determining the trajectory state, such as security monitoring, human motion recognition and other fields.

[0114] Furthermore, the present application also provides a relationship table between posture and trajectory state, the specific table is as follows:

[0115] Status at T1 Status at time T2 T3 time status Track Status Standing Standing Standing Unlock Standing other sit locking Standing other other locking sit other Standing Unlock sit other lie locking sit sit sit locking lie other sit locking lie lie other locking lie lie lie locking other Standing Standing Unlock other other Standing Unlock other sit other locking other lie lie locking other lie other locking other Standing other locking other other other locking other sit sit locking

[0116] When the target object's trajectory state is locked, the coordinates of the trajectory remain unchanged, which means the target is considered to be at the current position until it is unlocked; when the target object's trajectory state is unlocked, the coordinates of the subsequent trajectory can be changed according to the associated point cloud. Whether the trajectory is locked is shown in the table above. The first three columns of each row in the table above represent the states in which the trajectory state of the target object appears in sequence (T1, T2, and T3 are not necessarily continuous in time, but are just a time relationship, but the states of the three columns appear in sequence, and there is no other state in between), and the fourth column corresponds to whether the target object's trajectory state is locked or unlocked.

[0117] It should be understood that the above-mentioned relationship table between posture and trajectory state is only an exemplary description of the relationship between posture and trajectory state. The specific relationship between posture and trajectory state may vary according to specific circumstances and needs to be specifically designed and implemented according to actual needs.

[0118] Through the above method, the trajectory state corresponding to the target object can be determined according to the current posture of the target object, which improves the accuracy and robustness of target tracking and position monitoring, and takes into account the requirements of real-time and low memory usage.

[0119] 103. If the track state is a locked state, the latest historical position of the target object is determined as the current position of the target object.

[0120] The latest historical location may refer to the latest location information recorded by the target object in the past period of time, that is, the location information closest to the current time.

[0121] Specifically, if the trajectory state of the target object is a locked state, the latest historical position of the target object is determined as the current position of the target object, which means that when the target object is in a locked state, its current position will be fixed at its latest historical position. Since the locked state is a state indicating that the target trajectory is fixed, that is, the subsequent coordinate position of the target object will not change, when the target trajectory is locked, the current position of the target object will always remain at the position before locking, that is, the subsequent position information will use the latest historical position, which greatly saves computing resources. At the same time, in the case where the corresponding point cloud information cannot be detected because the target object is stationary, the technical solution of the embodiment of the present application can also avoid the loss and offset of the position of the target object.

[0122] Exemplarily, assume that there is a monitoring system for tracking the position of a human target. During the process of the target being tracked by the system, the system determines the trajectory state based on the changes in the target's posture. If the current posture of the target object is a non-standing state (such as sitting or lying), the system will set the trajectory state corresponding to the target object to a locked state. In the locked state, the latest historical position of the target object will be determined as its current position, and subsequent point cloud changes will not affect its position. For example, if the target object is a person sitting on a sofa, when the system detects that the target is in a sitting state, the system will lock the trajectory state of the target object. In the next update, even if the point cloud on the sofa changes, the position of the target object will remain unchanged and will always be displayed on the sofa.

[0123] Through the above methods, the stability and accuracy of the target position can be ensured when the target object is not standing. This is very important for tracking and positioning the target, especially in some application scenarios that require stable tracking of the target position, such as intelligent monitoring, behavior analysis, etc.

[0124] By implementing any one implementation or combination of implementations in the embodiments of the present application, the application scenario of the position monitoring process can be realized.

[0125] As can be seen from the above, the embodiments of the present application can determine the current posture of the target object; according to the current posture of the target object, determine the trajectory state corresponding to the target object; if the trajectory state is locked, the latest historical position of the target object is determined as the current position of the target object. It can be seen that this solution can determine whether the trajectory state is locked by performing posture recognition on the target object, and then accurately track the position of the target object and update the position information of the target object in real time. In this way, the loss or position displacement of the target object is avoided, and the position of the target object can be obtained quickly and accurately, which improves the user experience.

[0126] In order to facilitate the understanding of the embodiments of the present application, the embodiments of the present application will be described with specific application scenario examples. Figure 3 , Figure 3 It is a flow chart of target object posture recognition, such as Figure 3 :

[0127] The point cloud data of the current frame is clustered to obtain the clustered point cloud blocks, and then the motion trajectories in the target space are tracked and associated to determine the point cloud information corresponding to the target objects of each motion trajectory. After determining the posture feature vector corresponding to the target object according to the point cloud information, the first preset classifier is first used to perform the first posture recognition to determine whether the posture of the target object belongs to one of the three postures of sitting, standing, and lying. If so, the second preset classifier is used to perform the second posture recognition to determine whether the posture of the target object belongs to the standing state or the non-standing state. If it is the standing state, the current posture of the target object is directly determined. If it is the non-standing state, the third preset classifier is used to perform the third posture recognition to determine whether the posture of the target object belongs to sitting or lying. In this way, the posture of the target object can be determined.

[0128] See also Figure 4 , Figure 4 : is a flow chart of another location monitoring method provided in an embodiment of the present application, and the specific flow is as follows:

[0129] 201. Determine a current posture of the target object.

[0130] The specific content of step 201 may refer to the content of step 101 .

[0131] 202. Determine a trajectory state corresponding to the target object according to the current posture of the target object.

[0132] The specific content of step 202 may refer to the content of step 102 .

[0133] 203. If the trajectory state is unlocked, determine the current position of the target object according to the point cloud information corresponding to the target object.

[0134] Specifically, if the trajectory state of the target object is unlocked, the current position of the target object can be determined according to the point cloud information currently corresponding to the target object. Because the point cloud information is a series of point data in three-dimensional space, it can provide the shape, position and motion information of the target object. Therefore, the point cloud information corresponding to the target object can be used to determine its current position. For example, suppose a pedestrian moves in an indoor environment. After judgment, the trajectory state of the pedestrian is unlocked, so according to the corresponding point cloud information, his current position can be determined in real time. When the pedestrian moves, the point cloud data in the surrounding environment can be continuously obtained, and the associated motion trajectory can be determined according to the point cloud data obtained in real time, and then the latest position corresponding to the pedestrian can be determined in real time according to the point cloud data and its corresponding motion trajectory. For example, when a pedestrian walks from a corridor to a room, the point cloud data corresponding to the pedestrian can include the coordinate position change of the pedestrian entering the room from the corridor. According to this coordinate position change, the pedestrian's identification can be continuously displayed on the monitoring page to track the pedestrian's position in real time.

[0135] Through the above methods, the current position of the target object can be determined based on the point cloud information corresponding to the target object to provide accurate position monitoring and tracking. Regardless of whether the position of the target object changes, as long as the track state is unlocked, the position information of the target object can be updated based on the latest point cloud data, thereby achieving real-time target positioning and monitoring.

[0136] See also Figure 5 , Figure 5 This is a schematic diagram of information transmission of a location monitoring method provided by an embodiment of the present application. The specific process is as follows:

[0137] The data processing end sends indication information for indicating that the track state is a locked state, so that the client continues to display the identifier of the target object at the current position.

[0138] The indication information may be information used to indicate, guide or command a certain behavior or operation. The indication information may be used to indicate whether the trajectory state of the target object is locked and to allow the client to continuously display the target object's identification.

[0139] Specifically, the data processing end informs the client that the target object's trajectory state is locked by sending an indication message to ensure that the client can continue to display the target object's logo at the current location. The purpose of this is to enable the client to continuously track and display the target object's location information, enhancing the accuracy and real-time nature of monitoring. For example, when monitoring a human body indoors, when the data processing end detects that the target object's trajectory state is locked, it can send an indication message to the client. On the client's monitoring page, its logo will continue to be displayed at the current location.

[0140] This can help users track and display their location information in real time on the monitoring page, improving the reliability and accuracy of monitoring.

[0141] See also Figure 6 , Figure 6 is a flow chart of another location monitoring method provided in an embodiment of the present application, which is applied to Figure 1 The specific process for the client is as follows:

[0142] 301. Display the monitoring page of the target space.

[0143] In the embodiment of the present application, in order to obtain the position information of the target object in the target space in real time, the position information of the target object in the target space can be displayed to the user through the monitoring page on the terminal interface in the client.

[0144] The monitoring page may be a content screen including position marks corresponding to the target objects. The monitoring page corresponds to the monitoring range of the target space and is used to indicate the positions of the target objects in the target space in real time.

[0145] Specifically, a location monitoring-related application may be pre-installed on the client, or a corresponding monitoring page may be displayed on the client via a web page, so that the user can view the location information of the target object through the monitoring page to achieve location monitoring of the target object.

[0146] Through the above method, the monitoring page can be displayed through the client, so that the user can view the location information of the target object through the monitoring page.

[0147] 302. Receive indication information indicating that the state of the trajectory corresponding to the target object is a locked state.

[0148] Specifically, after the user opens the monitoring page on the user side, the user obtains the location information of the target object sent by the data processing side through the information transmission channel established with the data processing side. It is worth noting that the location information may include indication information for indicating that the corresponding trajectory state of the target object is a locked state. According to the indication information, the current coordinate position of each target object in the target space can be obtained.

[0149] In the above manner, the current position of the target object in the target space can be obtained, so that the position of the target object can be displayed on the monitoring page later.

[0150] 303. Continuously display the target object's identifier at a position corresponding to the current position in the monitoring page according to the indication information.

[0151] In order to visualize the position of the target object, after obtaining the current position of the target object, the identifier corresponding to the target object can be displayed in real time on the monitoring page of the user end, so as to visualize the position of the target object in the target space to the user.

[0152] It should be understood that there is a corresponding relationship between the monitoring page and the position of the target space, that is, each position in the monitoring page can correspond to a specific position in the target space. Specifically, at each position on the monitoring page, the position of a specific target object in the target space can be displayed by a corresponding mark. When the position of the target object changes, the position of the corresponding mark of the target object in the monitoring page will also change accordingly, so that the latest location information of the target object can always be displayed. Through this correspondence, the user can understand the position of the target object in real time, which is convenient for tracking and monitoring. Such a design can provide accurate location information and keep it updated in real time, providing users with a more convenient and intuitive target monitoring experience.

[0153] Specifically, after the user end obtains the location information of the target object sent by the data processing end, the user end displays the corresponding identifier of the target object in real time on the monitoring page. It is worth noting that the user end can continuously display the identifier of the target object at the position corresponding to the current position on the monitoring page according to the indication information sent by the data processing end.

[0154] Through the above method, the mark corresponding to the target object can be displayed in real time on the monitoring page, so as to visually display the position of the target object in the target space to the user.

[0155] In order to better implement the above method, the present application embodiment also provides a position monitoring device. Figure 7 As shown, the position monitoring device may include a first determining unit 401 , a second determining unit 402 and a third determining unit 403 .

[0156] A first determining unit 401 is used to determine the current posture of the target object;

[0157] A second determining unit 402 is used to determine a trajectory state corresponding to the target object according to the current posture of the target object;

[0158] The third determining unit 403 is configured to determine the latest historical position of the target object as the current position of the target object if the track state is the locked state.

[0159] In some implementations, the first determining unit 401 is specifically configured to:

[0160] Determine a plurality of point cloud information corresponding to the target object from a plurality of point cloud data frames corresponding to the target space, wherein the plurality of point cloud data frames correspond to the plurality of point cloud information one by one;

[0161] Based on multiple point cloud information, the current posture of the target object is determined.

[0162] In some implementations, the plurality of point cloud information includes current frame point cloud information and a preset number of historical frame point cloud information, and the first determination unit 401 determines the current posture of the target object according to the plurality of point cloud information, specifically for:

[0163] Construct the posture feature vector of the target object according to the point cloud information of the current frame and the point cloud information of the historical frame;

[0164] The posture feature vector is classified using a preset classifier to obtain the current posture of the target object.

[0165] In some implementations, the first determining unit 401 constructs a posture feature vector of the target object based on the point cloud information of the current frame and the point cloud information of the historical frame, specifically for:

[0166] According to the point clouds in the current frame point cloud information and the historical frame point cloud information, calculate the intra-frame feature vector of the current frame point cloud information and the intra-frame feature vector of the historical frame point cloud information;

[0167] According to the point clouds in the current frame point cloud information and the historical frame point cloud information, a first inter-frame feature vector between the current frame point cloud information and the historical frame point cloud information and a second inter-frame feature vector between each historical frame point cloud information are calculated;

[0168] The intra-frame feature vector, the first inter-frame feature vector, and the second inter-frame feature vector are concatenated to obtain a posture feature vector of the target object.

[0169] In some implementations, the posture includes a standing state and a non-standing state, and the second determining unit 402 is specifically configured to:

[0170] If the current posture is a standing state, it is determined that the trajectory state corresponding to the target object is an unlocked state;

[0171] If the current posture is a non-standing state, it is determined that the trajectory state corresponding to the target object is a locked state.

[0172] In some implementations, the position monitoring device further includes a fourth determining unit 404, which is specifically configured to:

[0173] If the track state is unlocked, the current position of the target object is determined according to the point cloud information corresponding to the target object.

[0174] In some implementations, the location monitoring device further includes a sending unit 405, which is specifically configured to:

[0175] The indication information for indicating that the track state is in a locked state is sent so that the client continues to display the identifier of the target object at the current position.

[0176] The present application also provides another position monitoring device. Figure 8 As shown, the position monitoring device may include a display unit 501 and a receiving unit 502 .

[0177] Display unit 501, used to display the monitoring page of the target space;

[0178] The receiving unit 502 is used to receive indication information indicating that the state of the trajectory corresponding to the target object is a locked state;

[0179] The display unit 501 is further configured to continuously display the identifier of the target object at a position corresponding to the current position of the target object in the monitoring page according to the indication information.

[0180] As can be seen from the above, the embodiment of the present application can determine whether the trajectory state is locked by performing posture recognition on the target object, and then accurately track the position of the target object and update the position information of the target object in real time. In this way, the target object is avoided from being lost or shifted, and the position of the target object can be quickly and accurately obtained, thereby improving the user experience.

[0181] The present application also provides a computer device, such as Fig. 9 As shown, it shows a schematic diagram of the structure of the computer device involved in the embodiment of the present application, specifically:

[0182] The computer device may include components such as a processor 601 with one or more processing cores, a memory 602 with one or more computer-readable storage media, a power supply 603, and an input unit 604. Those skilled in the art will appreciate that Fig. 9 The computer device structure shown in the figure does not constitute a limitation on the computer device, and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently. Among them:

[0183] The processor 601 is the control center of the computer device, and uses various interfaces and lines to connect various parts of the entire computer device. By running or executing software programs and / or modules stored in the memory 602, and calling data stored in the memory 602, the processor 601 performs various functions of the computer device and processes data. Optionally, the processor 601 may include one or more processing cores; preferably, the processor 601 may integrate an application processor and a modem processor, wherein the application processor mainly processes the operating system, user interface, and application programs, and the modem processor mainly processes wireless communications. It is understandable that the above-mentioned modem processor may not be integrated into the processor 601.

[0184] The memory 602 can be used to store software programs and modules. The processor 601 executes various functional applications and location monitoring by running the software programs and modules stored in the memory 602. The memory 602 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area may store data created according to the use of the computer device, etc. In addition, the memory 602 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage devices. Accordingly, the memory 602 may also include a memory controller to provide the processor 601 with access to the memory 602.

[0185] The computer device also includes a power supply 603 for supplying power to various components. Preferably, the power supply 603 can be logically connected to the processor 601 through a power management system, so as to manage charging, discharging, and power consumption through the power management system. The power supply 603 can also include any components such as one or more DC or AC power supplies, recharging systems, power failure detection circuits, power converters or inverters, and power status indicators.

[0186] The computer device may further include an input unit 604, which may be used to receive input digital or character information and generate keyboard, mouse, joystick, optical or trackball signal input related to user settings and function control.

[0187] Although not shown, the computer device may also include a display unit, etc., which will not be described in detail here. Specifically, in the embodiment of the present application, the processor 601 in the computer device will load the executable file corresponding to the process of one or more application programs into the memory 602 according to the following instructions, and the processor 601 will run the application program stored in the memory 602, thereby realizing various functions, as follows:

[0188] Determine the current posture of the target object; determine the trajectory state corresponding to the target object according to the current posture of the target object; if the trajectory state is a locked state, determine the latest historical position of the target object as the current position of the target object.

[0189] The specific implementation of the above operations can be found in the previous embodiments and will not be described in detail here.

[0190] It can be concluded that this solution can determine whether the trajectory state is locked by performing posture recognition on the target object, and then accurately track the position of the target object and update the position information of the target object in real time. In this way, the target object is avoided from being lost or offset, and the position of the target object can be quickly and accurately obtained, which improves the user experience.

[0191] A person of ordinary skill in the art will appreciate that all or part of the steps in the various methods of the above embodiments may be completed by instructions, or by controlling related hardware through instructions. The instructions may be stored in a computer-readable storage medium and loaded and executed by a processor.

[0192] To this end, an embodiment of the present application provides a computer-readable storage medium, in which a plurality of instructions are stored, and the instructions can be loaded by a processor to execute the steps in any one of the location monitoring methods provided in the embodiments of the present application. For example, the instructions can execute the following steps:

[0193] Determine the current posture of the target object; determine the trajectory state corresponding to the target object according to the current posture of the target object; if the trajectory state is a locked state, determine the latest historical position of the target object as the current position of the target object.

[0194] The specific implementation of the above operations can be found in the previous embodiments, which will not be described in detail here.

[0195] The computer-readable storage medium may include: a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0196] Since the instructions stored in the computer-readable storage medium can execute the steps in any data processing method provided in the embodiments of the present application, the beneficial effects that can be achieved by any data processing method provided in the embodiments of the present application can be achieved. Please refer to the previous embodiments for details and will not be repeated here.

[0197] According to one aspect 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 executes the method provided in various optional implementations provided in the above embodiments.

[0198] The above is a detailed introduction to a position monitoring method, device and computer-readable storage medium provided in the embodiments of the present application. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea; at the same time, for technical personnel in this field, according to the ideas of the present application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A location monitoring method, characterized in that: include: Determine the current pose of the target object; Determining a trajectory state corresponding to the target object according to the current posture of the target object; If the track state is a locked state, the latest historical position of the target object is determined as the current position of the target object.

2. The method according to claim 1, characterized in that The posture includes a standing state and a non-standing state, and determining the trajectory state corresponding to the target object according to the current posture of the target object includes: If the current posture is a standing state, determining that the trajectory state corresponding to the target object is an unlocked state; If the current posture is a non-standing state, it is determined that the trajectory state corresponding to the target object is a locked state.

3. The method according to claim 1, characterized in that The method comprises: If the track state is an unlocked state, the current position of the target object is determined according to the point cloud information corresponding to the target object.

4. The method according to any one of claims 1 to 3, characterized in that: The determining the current posture of the target object comprises: Determine a plurality of point cloud information corresponding to the target object from a plurality of point cloud data frames corresponding to the target space, wherein the plurality of point cloud data frames correspond to the plurality of point cloud information one by one; The current posture of the target object is determined according to the plurality of point cloud information.

5. The method according to claim 4, characterized in that The plurality of point cloud information includes current frame point cloud information and a preset number of historical frame point cloud information, and determining the current posture of the target object according to the plurality of point cloud information includes: Constructing a posture feature vector of the target object according to the point cloud information of the current frame and the point clouds in the point cloud information of the historical frame; The posture feature vector is classified using a preset classifier to obtain the current posture of the target object.

6. The method according to claim 5, characterized in that The step of constructing a posture feature vector of the target object according to the point cloud information of the current frame and the point cloud in the historical frame point cloud information includes: Calculate the intra-frame feature vector of the current frame point cloud information and the intra-frame feature vector of the historical frame point cloud information according to the point clouds in the current frame point cloud information and the historical frame point cloud information; According to the point clouds in the current frame point cloud information and the historical frame point cloud information, calculating a first inter-frame feature vector between the current frame point cloud information and the historical frame point cloud information, and a second inter-frame feature vector between each of the historical frame point cloud information; The intra-frame feature vector, the first inter-frame feature vector, and the second inter-frame feature vector are concatenated to obtain a posture feature vector of the target object.

7. The method according to any one of claims 1 to 3, characterized in that: The method further comprises: Indication information for indicating that the track state is a locked state is sent, so that the client continues to display the identifier of the target object at the current position.

8. A location monitoring method, characterized in that: include: Display the monitoring page of the target space; Receiving indication information for indicating that the state of the trajectory corresponding to the target object is a locked state; The identification of the target object is continuously displayed at a position corresponding to the current position of the target object in the monitoring page according to the indication information.

9. A position monitoring device, characterized in that: include: A first determining unit, used to determine a current posture of the target object; A second determining unit, configured to determine a trajectory state corresponding to the target object according to a current posture of the target object; The third determining unit is configured to determine the latest historical position of the target object as the current position of the target object if the track state is a locked state.

10. A location monitoring method, characterized in that: include: A display unit, used to display a monitoring page of the target space; A receiving unit, used to receive indication information indicating that the state of the trajectory corresponding to the target object is a locked state; The display unit is further configured to continuously display the identifier of the target object at a position corresponding to the current position of the target object in the monitoring page according to the indication information.

11. A computer device, characterized in that: The method comprises a processor and a memory, wherein the memory stores a computer program, and the processor is used to execute the computer program in the memory to implement the method according to any one of claims 1 to 8.

12. A computer-readable storage medium, characterized in that: The computer-readable storage medium is computer-readable and stores a plurality of instructions, wherein the instructions are suitable for being loaded by a processor to execute the method according to any one of claims 1 to 8.

Images