Seat status management method, device, equipment and storage medium

Abstract

The present application discloses a seat status management method, apparatus, device and storage medium. When the mark corresponding to a target seat is vacant, a first monitoring video of the target seat is obtained; based on the first monitoring video, it is determined that a first person exists in the target seat area and the first person is sitting on the target seat; a first person feature and at least one first object feature corresponding to the first person are obtained from the first monitoring video; the target seat is marked as occupied, and the first person feature and the at least one first object feature are bound to the current occupancy mark of the target seat as feature information corresponding to the current occupancy mark of the target seat, so that the study seats can be effectively monitored and managed.

Description

Technical Field

[0001] The present disclosure generally relates to the field of image recognition technology, and more particularly to a seat status management method, apparatus, device, and storage medium. Background Art

[0002] The vacant seats in places such as libraries and study rooms are usually very tight. In real life, students usually walk back and forth in the study area to observe and look for vacant study seats. They are prone to overlooking the actually vacant study seats, which greatly reduces the occupancy rate of the study seats and affects the experience of students. Summary of the Invention

[0003] In view of the above-mentioned defects or deficiencies in the prior art, it is desired to provide a seat status management method, device, equipment and storage medium that can effectively monitor and manage self-study seats.

[0004] In a first aspect, an embodiment of the present application provides a seat status management method, comprising:

[0005] When the mark corresponding to the target seat is vacant, obtaining a first monitoring video of the target seat;

[0006] Based on the first surveillance video, determining that a first person exists in the target seat area and that the first person is sitting in the target seat;

[0007] Acquire a first person feature and at least one first object feature corresponding to the first person from the first surveillance video;

[0008] The target seat is marked as occupied, and the first person feature and the at least one first object feature are bound to the current occupancy mark of the target seat as feature information corresponding to the current occupancy mark of the target seat.

[0009] In some embodiments, determining, based on the first surveillance video, that a first person exists in the target seating area and is seated in the target seat includes:

[0010] Inputting the first surveillance video into a behavior detection model to obtain pedestrian feature information, the pedestrian feature information including a bounding box for framing pedestrians in the surveillance video;

[0011] When the pedestrian characteristic information is within the reference area of ​​the target seat, determining that a first person exists in the reference area, and the pedestrian included in the pedestrian characteristic information is the first person;

[0012] Inputting the first surveillance video into a behavior posture analysis model to obtain posture information of the first person;

[0013] When the posture information is a sitting posture, it is determined that the first person is sitting on the target seat.

[0014] In some embodiments, obtaining a first object feature corresponding to the first person from the first surveillance video includes:

[0015] extracting at least one object feature from the first surveillance video;

[0016] For each of the object features, obtaining a first distance between the object feature and the pedestrian feature information corresponding to the first person;

[0017] The feature of the object whose first distance is less than or equal to the first preset distance is used as the first feature of the object corresponding to the first person.

[0018] In some embodiments, after marking the target seat as occupied, the method further includes:

[0019] Continue to obtain a first surveillance video corresponding to the target seat, and when it is determined based on the first surveillance video that the target seat area changes from the presence of the first person to the absence of the first person, obtain a second surveillance video corresponding to the target seat;

[0020] Identifying whether the at least one first object feature exists in the second surveillance video;

[0021] When the at least one first object feature does not exist in the second surveillance video, the target seat is marked as vacant.

[0022] In some embodiments, when the first item feature exists in the second surveillance video, the cumulative duration of the second surveillance video is obtained by counting;

[0023] When the accumulated time is greater than or equal to a first preset time length, an alarm message is sent according to a first preset rule.

[0024] In some embodiments, when the at least one first item feature is present in the second surveillance video, a first value corresponding to the first item feature is identified based on the at least one first item feature;

[0025] When the first value is less than or equal to a first preset value threshold, an alarm message is sent according to a second preset rule.

[0026] In some embodiments, based on the second surveillance video, it is determined that a second person exists in the target seat area and the second person is sitting in the target seat, and an alarm message is sent according to a third preset rule;

[0027] The second character feature corresponding to the second character is different from the first character feature corresponding to the first character.

[0028] In some embodiments, the alarm time is counted, and when the alarm condition is met next time, it is determined whether the alarm time reaches a second preset time length;

[0029] When the alarm time does not reach the second preset time length, the current alarm is canceled.

[0030] In a second aspect, an embodiment of the present application provides a seat status management device, comprising:

[0031] an acquisition module, configured to acquire a first monitoring video of a target seat when the mark corresponding to the target seat is vacant;

[0032] an analysis module, configured to determine, based on the first surveillance video, that a first person exists in the target seat area and that the first person is sitting in the target seat;

[0033] an extraction module, configured to obtain a first person feature and a first object feature corresponding to the first person from the first surveillance video;

[0034] The marking module is used to mark the target seat as occupied, and bind the first person feature and the first object feature to the target seat as feature information corresponding to the current occupancy mark of the target.

[0035] In some embodiments, the analysis module is further configured to:

[0036] Inputting the first surveillance video into a behavior detection model to obtain pedestrian feature information, the pedestrian feature information including a bounding box for framing pedestrians in the surveillance video;

[0037] When the pedestrian characteristic information is within the reference area of ​​the target seat, determining that a first person exists in the reference area, and the pedestrian included in the pedestrian characteristic information is the first person;

[0038] Inputting the first surveillance video into a behavior posture analysis model to obtain posture information of the first person;

[0039] When the posture information is a sitting posture, it is determined that the first person is sitting on the target seat.

[0040] In some embodiments, the extraction module is further configured to:

[0041] extracting at least one object feature from the first surveillance video;

[0042] For each of the object features, obtaining a first distance between the object feature and the pedestrian feature information corresponding to the first person;

[0043] The feature of the object whose first distance is less than or equal to the first preset distance is used as the first feature of the object corresponding to the first person.

[0044] In some embodiments, the acquisition module is further configured to: continue to acquire a first surveillance video corresponding to the target seat, and when it is determined based on the first surveillance video that the target seat area changes from the presence of the first person to the absence of the first person, acquire a second surveillance video corresponding to the target seat;

[0045] The analysis module is further configured to identify whether the at least one first object feature exists in the second surveillance video;

[0046] The marking module is further configured to mark the target seat as vacant when the at least one first object feature does not exist in the second surveillance video.

[0047] In some embodiments, the seat status management device further includes an alarm module configured to:

[0048] When the first object feature exists in the second surveillance video, obtaining a cumulative duration of the second surveillance video;

[0049] When the accumulated time is greater than or equal to a first preset time length, an alarm message is sent according to a first preset rule.

[0050] In some embodiments, the alarm module is further configured to:

[0051] When the at least one first item feature is present in the second surveillance video, identifying a first value corresponding to the first item feature based on the at least one first item feature;

[0052] When the first value is less than or equal to a first preset value threshold, an alarm message is sent according to a second preset rule.

[0053] In some embodiments, the alarm module is further configured to:

[0054] determining, based on the second surveillance video, that a second person exists in the target seat area and that the second person is sitting in the target seat, and sending an alarm message according to a third preset rule;

[0055] The second character feature corresponding to the second character is different from the first character feature corresponding to the first character.

[0056] In some embodiments, the alarm module is further configured to:

[0057] Counting the alarm time, and determining whether the alarm time reaches a second preset time length when the alarm condition is met next time;

[0058] When the alarm time does not reach the second preset time length, the current alarm is canceled.

[0059] In a third aspect, an embodiment of the present application provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method described in the embodiment of the present application when executing the program.

[0060] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium on which a computer program is stored, which, when executed by a processor, implements the method described in the embodiment of the present application.

[0061] The seat status management method proposed in the embodiment of the present application analyzes the first surveillance video when the mark corresponding to the target seat is idle, and marks the target seat as occupied when there is a first person in the target seat area and the first person sits on the target seat, and uses the first person feature and at least one object feature corresponding to the first person to bind to the current occupancy mark, thereby realizing automatic monitoring and marking management of target seats in environments such as study rooms or libraries, without the need for additional supervisors, and effectively improving the efficiency and reliability of seat status management.

[0062] Additional aspects and advantages of the present invention will be set forth in part in the description which follows and, in part, will be obvious from the description which follows, or may be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Other features, objects and advantages of the present application will become more apparent upon reading the detailed description of non-limiting embodiments made with reference to the following drawings:

[0064] Figure 1 The following is a diagram showing an implementation environment architecture of the seat status management method provided in an embodiment of the present application;

[0065] Figure 2 A flowchart of a seat status management method provided by an embodiment of the present application is shown;

[0066] Figure 3 A flowchart of a seat status management method provided by another embodiment of the present application is shown;

[0067] Figure 4 A diagram showing the relationship between pedestrian feature information and reference areas provided by an embodiment of the present application is shown;

[0068] Figure 5A flowchart of a seat status management method provided by another embodiment of the present application is shown;

[0069] Figure 6 A flowchart of a seat status management method provided in another embodiment of the present application is shown;

[0070] Figure 7 A flowchart of a seat status management method provided by a specific embodiment of the present application is shown;

[0071] Figure 8 An exemplary structural block diagram of a seat status management device provided by an embodiment of the present application is shown;

[0072] Figure 9 shows an exemplary structural block diagram of a seat status management device provided by another embodiment of the present application;

[0073] Figure 10 A schematic diagram of the structure of a computer system of an electronic device or server suitable for implementing an embodiment of the present application is shown. DETAILED DESCRIPTION

[0074] The present application will be further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are merely for the purpose of explaining the relevant invention and are not intended to limit the invention. It should also be noted that, for ease of description, only portions relevant to the invention are shown in the accompanying drawings.

[0075] It should be noted that, in the absence of conflict, the embodiments and features of the embodiments in this application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0076] The specific implementation environment of the seat status management method proposed in this application can be found in Figure 1 . Figure 1 The following diagram shows an implementation environment architecture diagram of the seat status management method provided in an embodiment of the present application.

[0077] like Figure 1 As shown, the implementation environment architecture includes: video acquisition equipment 1, server 2, monitoring terminal 3 and client 4.

[0078] Among them, the implementation environment may include at least one video acquisition device 1, which is used to capture video of the target seat to obtain a surveillance video. Optionally, one video acquisition device 1 can correspond to at least one target seat, that is, a one-to-one corresponding video acquisition device 1 can be set for each target seat, or one video acquisition device 1 can simultaneously capture surveillance videos of multiple target seats. Among them, when one video acquisition device 1 simultaneously captures surveillance videos of multiple target seats, multiple target seats should be within the visual range of the video acquisition device 1.

[0079] The server 2 is used to analyze the surveillance video collected by the video acquisition device 1 to obtain the seat status management result and feed it back to the supervisory terminal 3 and the client terminal 4. Figure 1 As shown, it may further include an algorithm server 21, a data server 22 and a business server 23, wherein the algorithm server 21 may be connected to at least one video acquisition device 1 to receive surveillance videos and perform algorithmic analysis on the surveillance videos to obtain, for example, pedestrian feature information, person feature information, object feature information and behavior posture information, etc. It should be understood that the algorithm server 21 may be provided with a trained or self-supervisory pedestrian detection model, a person feature extraction model, an object feature extraction model and a behavior posture analysis model.

[0080] Among them, the pedestrian detection model is used to detect whether there is a person in the surveillance video corresponding to the target seat, the person feature extraction model is used to extract the feature information of the person in the surveillance video corresponding to the target seat, the object feature extraction model is used to continuously extract the feature information of the object in the surveillance video after there is a person in the surveillance video corresponding to the target seat or the seat is marked as occupied, and the behavior posture analysis model is used to determine the person's behavior posture when there is a person in the surveillance video corresponding to the target seat, so as to determine whether the seat is occupied.

[0081] Then, the algorithm server 21 sends the pedestrian feature information, person feature information, item feature information and behavior posture information to the data server 22 connected to it. The data server 22 performs data statistical analysis based on the pedestrian feature information, person feature information, item feature information and behavior posture information to obtain an analysis result of whether the target seat is occupied. The data server 22 sends the analysis result of whether the target seat is occupied to the business server 23. The business server 23 sends the received occupancy status of the target seat to the supervision terminal 3 for occupancy rate monitoring, and sends a reminder to the supervision terminal 3 when the seat is vacant for a long time, and sends a reminder to the original occupant when the seat is occupied by others.

[0082] Among them, server 2 can be an independent physical server, or a server cluster or 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, as well as big data and artificial intelligence platforms.

[0083] The supervisory terminal 3 and the client terminal 4 are respectively connected to the server 2 in communication, and in particular to the business server 23 in the server 3. The supervisory terminal 3 can actively query the seat status information. For example, the supervisor sends a query request to the business server 23 through the supervisory terminal 3 to query the status information of at least one target seat. The business server 23 sends the status information of at least one target seat contained in the query request to the supervisory terminal 3 according to the query request. The supervisory terminal 3 can also passively receive alarm information and / or status information of at least one target seat sent by the business server 23. For example, the business server 23 can periodically send the status information of at least one target seat to the supervisory terminal 3 so that the supervisor can monitor the seat status through the supervisory terminal 3. The business server 23 can also send an alarm message to the supervisory terminal 3 when the seat is occupied for a long time. The client 4 also actively queries the seat status information. For example, the user can use the client 4 to check whether there are any vacant seats in the study room or library, and then decide whether to go to the study room or library to study. The client 4 can also passively receive alarm information sent by the business server 23, such as when the user leaves the seat for a long time and occupies the seat for a long time with items, or when the seat occupied by the user is occupied by other users.

[0084] The video capture device 1, algorithm server 21, data server 22, service server 23, monitoring terminal 3, and client 4 are directly or indirectly connected via wired or wireless communication. Optionally, the wireless or wired network described above utilizes standard communication technologies and / or protocols. The network is typically the Internet, but can also be any network, including but not limited to a local area network (LAN), a metropolitan area network (MAN), a wide area network (WAN), a mobile, wired or wireless network, a private network, or any combination of a virtual private network.

[0085] The seat status management method proposed in this application can be implemented by a seat status management device, and the seat status management device can be installed on a server.

[0086] In order to further illustrate the technical solutions provided by the embodiments of the present application, this is described in detail below with reference to the accompanying drawings and specific embodiments. Although the embodiments of the present application provide the method operation instruction steps shown in the following embodiments or drawings, more or fewer operation instruction steps may be included in the method based on conventional or no creative labor. In the steps where there is no necessary causal relationship logically, the execution order of these steps is not limited to the execution order provided by the embodiments of the present application. The method may be executed in the order of the methods shown in the embodiments or drawings or in parallel during the actual processing process or when the device is executed.

[0087] Please refer to Figure 2 , Figure 2 FIG. 1 shows a flow chart of a seat status management method provided by an embodiment of the present application. Figure 2 As shown, the method includes:

[0088] Step 101: When the mark corresponding to the target seat is vacant, obtain a first monitoring video of the target seat.

[0089] Among them, the first monitoring video can be obtained by a video acquisition device through video acquisition of the target seat. The video acquisition device can correspond to the target seat one by one, that is, one video acquisition device corresponds to collecting the first monitoring video of one target seat. The video acquisition device can also correspond to multiple target seats at the same time, that is, one video acquisition device corresponds to collecting the first monitoring videos of multiple target seats.

[0090] Optionally, the first surveillance video may be a segmented video or a spliced ​​video. For example, when a video acquisition device captures the first surveillance videos of multiple target seats, the original video captured by the video acquisition device may be segmented according to the position information or identification information of the target seats to obtain a first surveillance video corresponding to each target seat. Alternatively, if the video acquisition device is a video acquisition system composed of multiple cameras, the original videos captured by the multiple cameras may be spliced ​​according to the spatial position relationship to obtain a spatial first surveillance video, and then segmented based on the position information or identification information of each target seat to obtain a first surveillance video corresponding to each target seat. Alternatively, the target seat may be captured by at least two cameras for partial area video, and then the videos of the multiple partial areas may be spliced ​​according to the spatial position relationship to obtain the first surveillance video corresponding to the target seat.

[0091] Step 102: Based on the first surveillance video, determine that a first person exists in the target seat area and sits in the target seat.

[0092] The first person may be a pedestrian in the target seating area.

[0093] Optionally, it is possible to first determine whether the first person exists in the target seat area based on the first surveillance video. If the first person exists in the first surveillance video, it is further determined whether the first person is sitting in the target seat. If the first person does not exist in the first video, there is no need to determine whether the first person is sitting in the target seat.

[0094] Optionally, the first surveillance video can be analyzed using a trained convolutional neural network or deep learning model to obtain information on whether the first person exists in the target seating area and whether the first person sits in the target seat when the first person exists in the target seating area.

[0095] Step 103: Acquire a first person feature and at least one first object feature corresponding to the first person from the first surveillance video.

[0096] The first person feature can be a partial image of the first person, such as a facial image, or abstract feature data obtained by performing convolution image processing on the first person. The first object feature can be a partial or full image of the first object, such as a front view of the first person's backpack, or abstract feature data obtained by performing convolution image processing on the first object.

[0097] Step 104 : Mark the target seat as occupied, and bind the first person feature and the at least one first object feature to the current occupancy mark of the target seat as feature information corresponding to the current occupancy mark of the target seat.

[0098] The current occupancy mark of the target seat may be bound to the first person feature and the at least one object feature by establishing a mapping relationship between the current occupancy mark of the target seat and the first person feature and the at least one object feature.

[0099] Therefore, the seat status management method proposed in the embodiment of the present application analyzes the first surveillance video when the mark corresponding to the target seat is idle, and when there is a first person in the target seat area and the first person sits on the target seat, marks the target seat as occupied, and uses the first person feature and at least one object feature corresponding to the first person to bind to the current occupancy mark, thereby realizing automatic monitoring and marking management of target seats in environments such as study rooms or libraries, without the need for additional supervisors, and effectively improving the efficiency and reliability of seat status management.

[0100] In one or more embodiments, Figure 3 As shown, step 102, based on the first monitoring video, determines that a first person exists in the target seat area and sits in the target seat, including:

[0101] Step 1021: Input the first surveillance video into a behavior detection model to obtain pedestrian feature information, where the pedestrian feature information includes a bounding box for framing pedestrians in the surveillance video.

[0102] Among them, the pedestrian detection model can be a target recognition model trained using pedestrian label information. The pedestrian detection model can identify people in video images and mark them with bounding boxes, and output the video images with bounding box marks as pedestrian feature information.

[0103] It should be understood that the embodiments of the present application do not improve the target recognition model structure used as a pedestrian detection model. That is, during the application process, the embodiments of the present application can directly select a relatively mature target recognition model that can analyze video data, and then use a training set with pedestrian labels for training until the expected effect is achieved.

[0104] Optionally, a pedestrian detection model is set in the server. Specifically, the pedestrian detection model is set in the algorithm server, wherein a pedestrian detection model may be set in the algorithm server, and surveillance videos corresponding to multiple target seats may be input into the pedestrian detection model at the same time. The pedestrian detection model outputs the pedestrian feature information obtained by analysis in sequence or simultaneously according to the analysis completion time. In order to improve the analysis efficiency, multiple pedestrian detection models may also be set in the algorithm server at the same time, and the surveillance videos may be input into multiple pedestrian detection models according to strategies such as the grouping information of the target seats or the acquisition time, so as to reduce the amount of surveillance videos analyzed by each pedestrian detection model and improve the efficiency of pedestrian feature analysis. Optionally, a pedestrian detection model corresponding to the number of surveillance videos may be set in the algorithm server, so that each surveillance video can be input into the corresponding pedestrian detection model in a timely manner to ensure the timeliness of pedestrian feature analysis.

[0105] Step 1022: When the pedestrian characteristic information is within the reference area of ​​the target seat, it is determined that a first person exists in the reference area, and the pedestrian included in the pedestrian characteristic information is the first person.

[0106] It should be noted that since the pedestrian feature information includes a bounding box for framing the pedestrian in the image, the target position of the pedestrian in the image can be determined based on the bounding box. Therefore, based on the target position of the pedestrian in the image and the reference area of ​​the target seat set in advance, it can be determined whether there is someone at the target seat, thereby determining the occupancy status of the target seat.

[0107] In one or more embodiments, the surveillance video of the target seat is a surveillance video collected from a fixed area, that is, the position of the video acquisition device remains unchanged or the position is the same when collecting the same target seat. Therefore, a reference area in the surveillance video can be set according to attribute information such as the position and size of the target seat, and then the occupancy status of the target seat can be determined based on the positional relationship between the target position and the reference area.

[0108] In one or more embodiments, the surveillance video can be parsed to obtain the spatial coordinates of each position point in the surveillance video, and then the spatial coordinates corresponding to the target seat are determined based on the position of the attribute characteristics of the target seat in the surveillance video, and the spatial coordinates corresponding to the target seat are used as the reference area. After the pedestrian detection model analyzes the surveillance video to obtain pedestrian feature information, the spatial coordinates corresponding to the pedestrian feature information are parsed and the spatial coordinates corresponding to the pedestrian feature information are used as the target position, and then the occupancy status of the target seat is determined based on the positional relationship between the target position and the reference area.

[0109] It should be understood that if the target position is within the reference area, it is determined that the first person exists in the reference area; if the target position is not within the reference area, it is determined that the first person does not exist in the reference area.

[0110] Optionally, the positional relationship between the target position and the reference area can be determined by comparing their coordinate values ​​in the same coordinate system. A reference area includes at least two coordinates: a minimum coordinate value and a maximum coordinate value. Therefore, the target position needs to be compared with at least two reference areas to determine the positional relationship between the target position and the reference area.

[0111] For example, when both the target position and the reference area are expressed by pixel coordinates, the pixel coordinate value corresponding to the target position is compared with the minimum coordinate value and the maximum coordinate value of the area range corresponding to the reference area respectively. If the target position is greater than or equal to the minimum coordinate value of the area range and less than or equal to the maximum coordinate value of the area range, it is determined that the target position is within the reference area, indicating that there are pedestrians in the reference area, and it is determined that the first person exists in the target seating area. If the target position is less than the minimum coordinate value of the area range or the target position is greater than the maximum coordinate value of the area range, indicating that there are no pedestrians in the reference area, it is determined that the target position is not within the reference area, and it is obtained that the first person does not exist in the target seating area.

[0112] Furthermore, the target position is the two lower corner points of the bounding box in the pedestrian feature information. According to the target position corresponding to the pedestrian feature information and the reference area corresponding to the target seat, it is determined whether there is a first person in the reference area corresponding to the target seat, including: if both lower corner points are within the reference area, it is determined that there is a first person in the reference area corresponding to the target seat.

[0113] It should be understood that the image of a person has a certain width. In order to avoid the problem of misjudging part of the body area of ​​an employee in an adjacent seat as a pedestrian in the current target seat due to the close distance to the target seat, this application proposes to use the two lower corner points of the bounding box in the pedestrian feature information as redundant judgment conditions, that is, the target coordinates corresponding to the two lower corner points expressing the width of the pedestrian must be within the reference area before it is determined that there is a first person in the target seat, that is, the first person is within the reference area corresponding to the target seat.

[0114] Specifically, if Figure 4 As shown, the first and second lower corners of the bounding box in the pedestrian feature information are obtained respectively, and it is determined whether the first lower corner is within the reference area. If the first lower corner is within the reference area, it is further determined whether the second lower corner is within the reference area. If the second lower corner is also within the reference area, it means that the entire pedestrian is within the reference area corresponding to the target seat, and the first person is determined to be present in the target seat. If the second lower corner is not within the reference area, it means that the pedestrian is a pedestrian in the nearby area, which is a misjudgment. If the first lower corner is not within the reference area, it means that the pedestrian does not belong to the seat. The first lower corner can be the lower left corner of the pedestrian feature information, and the second lower corner can be the lower right corner of the pedestrian feature information.

[0115] Optionally, the two upper corner points of the bounding box in the pedestrian feature information can also be used as redundant judgment conditions. However, since the heights of pedestrians may vary greatly, it is not as accurate as the determined ground area range judgment. Therefore, it is preferred to use the two lower corner points as redundant judgment conditions.

[0116] Optional, such as Figure 4 As shown, Box 1 can represent the spatial range covered by the target seat. To further reduce judgment errors, the spatial range of the target seat can be reduced, and Box 2 can be used as the reference area corresponding to the target seat. The size of Box 2 can be the spatial range of the pedestrian's self-study state, determined based on historical data within the spatial range covered by the target seat. Boxes 1 and 2 are located in the pedestrian's foot area, with Box 1 nested outside of Box 2.

[0117] Step 1023: Input the first surveillance video into a behavior posture analysis model to obtain posture information of the first person.

[0118] It should be understood that, similar to the behavior detection model, the posture analysis model can be a target classification model trained using pedestrian posture information labels. The posture analysis model can identify the behavior posture of people in video images, and output the mapping relationship between the behavior posture information and the video segment (frame), or output the mapping relationship between the behavior posture information and the people in the video segment (frame).

[0119] It should be understood that the embodiments of the present application do not improve the target classification model structure used as a posture analysis model. That is, during the application process, the embodiments of the present application can directly select the current more mature target classification model that can analyze video data, and then use the training set with behavioral posture labels for training until the expected effect is achieved.

[0120] Optionally, the behavior posture analysis model can also be an image analysis model, that is, the extracted pedestrian feature information can be directly input into the behavior posture analysis model, so that the posture analysis model can directly perform posture analysis on the pedestrians who have been determined to be in the target seat reference area, ensuring that the analysis result is the posture of the first person, effectively reducing the data processing volume of the posture analysis model, and improving the efficiency of the target seat status analysis.

[0121] Step 1024: When the posture information is a sitting posture, determine that the first person is sitting in the target seat.

[0122] Furthermore, if Figure 5 As shown, step 103, obtaining a first object feature corresponding to the first person from the first surveillance video, includes:

[0123] Step 1031: extract at least one object feature from the first surveillance video.

[0124] The object features extracted from the first surveillance video may be newly added object features, that is, newly added object features obtained by deleting environmental objects (such as paintings, lamps, etc.).

[0125] Step 1032: For each object feature, obtain a first distance between the object feature and pedestrian feature information corresponding to the first person.

[0126] Step 1033 : Using the object feature with the first distance less than or equal to the first preset distance as the first object feature corresponding to the first person.

[0127] It will be appreciated that, since the purpose of this embodiment is to obtain items carried by a first person, the distance information between the item characteristics and the first person is used to determine the closeness between the item and the first person, thereby determining whether the item characteristics are the first item characteristics corresponding to the first person. Specifically, when the first distance is less than or equal to a first preset distance, the item is determined to be the first item characteristics corresponding to the first person; when the first distance is greater than the first preset distance, the item is determined not to be the first item characteristics corresponding to the first person.

[0128] The first preset distance may be obtained based on a limited number of experiments. Specifically, the first preset distance is usually no larger than the size of the target as the corresponding reference area.

[0129] Therefore, the embodiment of the present application can effectively determine the characteristics of the item related to the first person through distance, effectively improve the accuracy of item binding, and provide more accurate data support for subsequent occupancy judgment.

[0130] In one or more embodiments, Figure 6 As shown, after marking the target as occupied, it also includes:

[0131] Step 105 , continue to obtain the first surveillance video corresponding to the target seat, and when it is determined based on the first surveillance video that the target seat area changes from the presence of the first person to the absence of the first person, obtain the second surveillance video corresponding to the target seat.

[0132] It is understandable that the first surveillance video obtained is a video of the first person in the target seat, and the second surveillance video is a video of the first person not in the target seat. When the target seat changes from the first person in the seat to the first person in the seat, it means that the first person associated with the current occupancy mark has left the seat. At this time, the first person may have left to finish self-study or temporarily leave. Therefore, we need to use the second surveillance video to make further judgments.

[0133] That is to say, this application needs to continue to monitor the target seat after determining that the target seat is occupied, so as to identify and remind the seat-occupying behavior or seat-grabbing behavior.

[0134] Step 106: Identify whether the first object feature exists in the second surveillance video.

[0135] Step 107: When the first object feature does not exist in the second surveillance video, mark the target seat as vacant.

[0136] Specifically, the first surveillance video of the target seat is continuously obtained, and the first person is judged on the first surveillance video. If the first person exists in the first surveillance video, the first surveillance video of the target video is continued to be obtained. If the first person does not exist in the first surveillance video, the second surveillance video of the target seat is obtained. After obtaining the second surveillance video, the object features in the second surveillance video are extracted, and it is determined whether the multiple extracted object features are the first object features bound to the current occupancy mark of the target seat and the first person. If so, it is determined that there is at least one first object feature in the second surveillance video. At this time, it is determined that the user occupying the target seat (the first person) has left the target seat and left an object seat. If at least one object feature extracted from the second financial control video is traversed and does not match the first object feature, it is determined that there is no first object feature in the second surveillance video. At this time, it is determined that the user occupying the target seat (the first person) has left the target seat and ended self-study, and the target seat is marked as vacant.

[0137] Furthermore, when the first object feature exists in the second surveillance video, the cumulative duration of the second surveillance video is obtained by counting, and when the cumulative duration is greater than or equal to the first preset time length, an alarm message is sent according to the first preset rule.

[0138] Among them, the first preset rule includes sending an alarm message to the user occupying the target seat.

[0139] That is to say, when the first object feature is identified in the second surveillance video, the target seat is determined to be in a state of being temporarily left but occupied by the first person. At this time, the length of time the first person was away is determined by counting the duration of the second surveillance video. If the cumulative duration is greater than or equal to the first estimated time length, it means that the first person was away for too long, which affected the utilization efficiency of the study room or library. At this time, an alarm message is sent according to the first preset rule to remind the user occupying the target seat (the first person) to return or end the study in time.

[0140] Among them, the first preset rule may also include sending a reminder message to the supervision end to remind the supervisor to pay attention to the occupancy status of the target seat, so that when the target seat is forcibly marked as free from the occupied state, the supervisor can help to store the first item to avoid the loss of user items.

[0141] It should also be understood that in the embodiment of the present application, the surveillance video is determined to be the second surveillance video only when the first person is absent in the surveillance video. That is to say, if the acquired surveillance video changes from not having the first person to having the first person present, then it is determined that the acquired surveillance video changes from the second surveillance video to the first surveillance video. At this time, it is determined that the first person returns to the target seat, the timing is no longer continued, and the step of marking the target seat as occupied is returned to.

[0142] In one or more embodiments, when a first object feature exists in the second surveillance video, a first value corresponding to the first object feature is identified based on the first object feature, and when the first value is less than or equal to a first preset value, an alarm message is sent according to a second preset rule.

[0143] Among them, the second preset rule is to send an alarm message to the supervision end, and the first value is the total value of at least one first item feature.

[0144] That is to say, the present application further assigns a value to the items used to occupy a seat, and then calculates whether the total value (first value) of at least one first item feature used to occupy a seat is less than or equal to a first preset value threshold. If the first value is less than or equal to the first preset value threshold, an alarm message is sent according to a second preset rule to remind the supervisory end to verify the value of the first item feature, and determine whether to manually mark the target seat as vacant based on the actual value of the first item, so as to avoid the system misjudging the value of high-value items and causing the user to lose valuable items. If the first value is greater than the first preset value, it is determined that the first item is indeed the item used by the first person to occupy the seat, and the current occupancy mark of the target seat is maintained.

[0145] It should be understood that the value of the first item used to reserve a seat can be the value of an item used for internships or self-study. For example, while the market value of items such as books, notebooks, pencil cases, and pencil cases may be relatively low, they will have a higher value when assigned a seat-reserving item. Items such as bottled water and grocery bags will have a lower value when assigned a seat-reserving item. The value assigned to various items can be determined based on statistical analysis of the historical frequency of seat-reserving item use, and this application does not impose any specific limitations.

[0146] Therefore, the present application can determine whether the user is occupying the seat based on the value of the first item left on the target seat, thereby avoiding mistaking low-value forgotten items for seat-occupying items, which affects the efficiency of using the study room or library.

[0147] In one or more embodiments, based on the second surveillance video, it is determined that there is a second person in the target seat area and the second person is sitting in the target seat, and an alarm message is sent according to a third preset rule, wherein the second person feature corresponding to the second person is different from the first person feature corresponding to the first person.

[0148] Among them, the third preset rule is to send an alarm message to the first person and send an alarm message to the supervisory end at the same time.

[0149] That is to say, this application further judges the second surveillance video, wherein the judgment of whether there is a second person in the target seat area and whether the second person is sitting in the target seat can refer to the judgment of the first person, that is, whether there is someone else occupying the first person's seat when the first person temporarily leaves.

[0150] Specifically, the second surveillance video can be first judged whether there is a person. When the pedestrian feature information is identified, the reference person feature is extracted, and then it is judged whether the extracted reference person feature is the same as the first person feature. If the reference person feature is the same as the first person feature, it is determined that the first person has returned to the target seat, and the step of obtaining the first surveillance video is returned. If the reference person feature is different from the reference feature of the first person, it is determined that the reference person feature is the second person feature and corresponds to the second person, and then it is further judged whether the second person is sitting in the target seat. If the second person is sitting in the target seat, it means that the second person has occupied the target seat occupied by the first person. At this time, an alarm message is sent to the first person and the supervision end at the same time, so that the first person can return to the target seat in time. At the same time, the supervision end intervenes in the behavior of occupying other people's seats. If the second person is not sitting in the target seat, the step of timing the second surveillance video is returned.

[0151] It should be understood that the alarm information can be sent to the first person by using a client or applet installed on the user's mobile terminal. After the first person occupies the target seat, he or she can bind the location by scanning the QR code corresponding to the target seat so as to receive the alarm message when the alarm condition is triggered.

[0152] It should also be understood that the first person feature and the first object feature can be directly obtained through the mapping relationship of the current occupancy mark, without the need to repeatedly retrieve and compare from the feature database, effectively improving the efficiency of feature data comparison.

[0153] Furthermore, it also includes: counting the alarm time, and when the alarm condition is met next time, judging whether the alarm time reaches a preset time threshold, and canceling the current alarm if the alarm time does not reach the preset time threshold.

[0154] Among them, the alarm time is the cumulative length of time since the alarm was issued. That is to say, two consecutive alarms need to be issued at an interval of a preset time threshold, thereby effectively reducing the frequency of alarm issuance and avoiding repeated alarms.

[0155] As a specific embodiment, Figure 7 As shown, the seat status management method includes:

[0156] Step 201: A video capture device obtains a surveillance video of a target seat.

[0157] Step 202: Determine whether the first person exists in the surveillance video.

[0158] If yes, it is determined that the surveillance video is the first surveillance video and step 203 is executed; if not, the process returns to step 201.

[0159] Step 203: Determine whether the first person is sitting in the target seat.

[0160] If yes, execute step 204 ; if no, return to step 201 .

[0161] Step 204 : Acquire a first person feature and at least one first object feature corresponding to the first person from the first surveillance video.

[0162] Step 205: Mark the target seat as occupied, and bind the first person feature and at least one first object feature to the current occupancy mark of the target seat, and the feature information corresponding to the current occupancy mark of the target seat.

[0163] Step 206: Continue to obtain the first surveillance video corresponding to the target seat.

[0164] Step 207 , determining whether the target seat area in the first surveillance video changes from the presence of the first person to the absence of the first person.

[0165] If yes, go to step 208 , if no, go back to step 206 .

[0166] Step 208: Acquire a second surveillance video corresponding to the target seat.

[0167] Step 209: Determine whether there is at least one first object feature in the second video.

[0168] If yes, step 210 is executed, and if no, the target seat is marked as vacant.

[0169] Step 210: Count and obtain the cumulative duration of the second surveillance video.

[0170] Step 211 : Identify a first value corresponding to the first item feature based on at least one first item feature.

[0171] Step 212: Determine whether the first value is less than or equal to a first preset value threshold.

[0172] If yes, send an alarm message according to the second preset rule; if no, execute step 213.

[0173] Step 213: Determine whether a second person exists in the target seat area and sits in the target seat in the second surveillance video.

[0174] If yes, send an alarm message according to the third preset rule; if no, execute step 214.

[0175] Step 214: Determine whether the accumulated time is greater than or equal to a first preset time length.

[0176] If yes, send an alarm message according to the first preset rule; if no, return to step 208.

[0177] Among them, steps 211 to 213 can be executed only in the first cycle, that is, the value judgment of the item characteristics does not need to be repeated, avoiding unnecessary increase in calculation amount.

[0178] To sum up, the seat status management method proposed in the embodiment of the present application analyzes the first surveillance video when the mark corresponding to the target seat is idle, and when there is a first person in the target seat area and the first person sits on the target seat, marks the target seat as occupied, and uses the first person feature and at least one object feature corresponding to the first person to bind to the current occupancy mark, thereby realizing automatic monitoring and marking management of target seats in environments such as study rooms or libraries, without the need for additional supervisors, and effectively improving the efficiency and reliability of seat status management.

[0179] It should be noted that although the operations of the present method are described in a particular order in the drawings, this does not require or imply that the operations must be performed in this particular order, or that all illustrated operations must be performed to achieve desirable results.

[0180] Figure 8 An exemplary structural block diagram of a seat status management device provided in an embodiment of the present application is shown.

[0181] like Figure 8 As shown, the seat status management device 10 includes:

[0182] An acquisition module 11 is configured to acquire a first surveillance video of a target seat when the mark corresponding to the target seat is vacant;

[0183] an analyzing module 12, configured to determine, based on the first surveillance video, that a first person exists in the target seat area and that the first person is sitting in the target seat;

[0184] An extraction module 13 is configured to obtain a first person feature and a first object feature corresponding to the first person from the first surveillance video;

[0185] The marking module 14 is configured to mark the target seat as occupied, and bind the first person feature and the first object feature to the target seat as feature information corresponding to the current occupancy mark of the target seat.

[0186] In some embodiments, the analysis module 12 is further configured to:

[0187] Inputting the first surveillance video into a behavior detection model to obtain pedestrian feature information, the pedestrian feature information including a bounding box for framing pedestrians in the surveillance video;

[0188] When the pedestrian characteristic information is within the reference area of ​​the target seat, determining that a first person exists in the reference area, and the pedestrian included in the pedestrian characteristic information is the first person;

[0189] Inputting the first surveillance video into a behavior posture analysis model to obtain posture information of the first person;

[0190] When the posture information is a sitting posture, it is determined that the first person is sitting on the target seat.

[0191] In some embodiments, the extraction module 13 is further configured to:

[0192] extracting at least one object feature from the first surveillance video;

[0193] For each of the object features, obtaining a first distance between the object feature and the pedestrian feature information corresponding to the first person;

[0194] The feature of the object whose first distance is less than or equal to the first preset distance is used as the first feature of the object corresponding to the first person.

[0195] In some embodiments, the acquisition module 11 is further configured to: continue to acquire a first surveillance video corresponding to the target seat, and when it is determined based on the first surveillance video that the target seat area changes from the presence of the first person to the absence of the first person, acquire a second surveillance video corresponding to the target seat;

[0196] The analysis module 12 is further configured to identify whether the at least one first object feature exists in the second surveillance video;

[0197] The marking module 14 is further configured to mark the target seat as vacant when the at least one first object feature does not exist in the second surveillance video.

[0198] In some embodiments, as Figure 9 As shown, the seat status management device 10 further includes an alarm module 15 for:

[0199] When the first object feature exists in the second surveillance video, obtaining a cumulative duration of the second surveillance video;

[0200] When the accumulated time is greater than or equal to a first preset time length, an alarm message is sent according to a first preset rule.

[0201] In some embodiments, the alarm module 15 is further configured to:

[0202] When the at least one first item feature is present in the second surveillance video, identifying a first value corresponding to the first item feature based on the at least one first item feature;

[0203] When the first value is less than or equal to a first preset value threshold, an alarm message is sent according to a second preset rule.

[0204] In some embodiments, the alarm module 15 is further configured to:

[0205] determining, based on the second surveillance video, that a second person exists in the target seat area and that the second person is sitting in the target seat, and sending an alarm message according to a third preset rule;

[0206] The second character feature corresponding to the second character is different from the first character feature corresponding to the first character.

[0207] In some embodiments, the alarm module 15 is further configured to:

[0208] Counting the alarm time, and determining whether the alarm time reaches a second preset time length when the alarm condition is met next time;

[0209] When the alarm time does not reach the second preset time length, the current alarm is canceled.

[0210] To sum up, the seat status management device proposed in the embodiment of the present application analyzes the first surveillance video when the mark corresponding to the target seat is idle, and marks the target seat as occupied when there is a first person in the target seat area and the first person sits on the target seat, and uses the first person feature and at least one object feature corresponding to the first person to bind to the current occupancy mark, thereby realizing automatic monitoring and marking management of target seats in environments such as study rooms or libraries, without the need for additional supervisors, and effectively improving the efficiency and reliability of seat status management.

[0211] It should be understood that the units or modules described in the apparatus 10 are similar to those described in the reference Figure 2 The various steps in the described method correspond to each other. Therefore, the operations and features described above for the method are also applicable to the device 10 and the units contained therein, and will not be repeated here. The device 10 can be pre-implemented in the browser or other security application of the electronic device, or loaded into the browser or its security application of the electronic device by downloading or other means. The corresponding units in the device 10 can cooperate with the units in the electronic device to implement the solution of the embodiment of the present application.

[0212] The several modules or units mentioned in the detailed description above are not necessarily divided into one module or unit. In fact, according to the embodiments of the present disclosure, the features and functions of two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided into multiple modules or units to be embodied.

[0213] Reference below Figure 10 , Figure 10 A schematic diagram of the structure of a computer system of an electronic device or server suitable for implementing the embodiments of the present application is shown.

[0214] like Figure 10 As shown, the computer system includes a central processing unit (CPU) 1001, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 1002 or the program loaded from the storage part 1008 into the random access memory (RAM) 1003. Various programs and data required for the operation instructions of the system are also stored in the RAM 1003. The CPU 1001, ROM 1002 and RAM 1003 are connected to each other via a bus 1004. An input / output (I / O) interface 1005 is also connected to the bus 1004.

[0215] The following components are connected to the I / O interface 1005: an input section 1006 including a keyboard, a mouse, and the like; an output section 1007 including devices such as a cathode ray tube (CRT), a liquid crystal display (LCD), and a speaker; a storage section 1008 including a hard disk; and a communication section 1009 including a network interface card such as a LAN card or a modem. The communication section 1009 performs communication processing via a network such as the Internet. A drive 1010 is also connected to the I / O interface 1005 as needed. A removable medium 1011, such as a magnetic disk, an optical disk, a magneto-optical disk, or a semiconductor memory, is installed in the drive 1010 as needed, so that computer programs read therefrom can be installed into the storage section 1008 as needed.

[0216] In particular, according to the embodiment of the present application, the above reference flow chart Figure 2The described process can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes program code for executing the method shown in the flowchart. In such an embodiment, the computer program includes program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network via the communication section 1009, and / or installed from a removable medium 1011. When the computer program is executed by the central processing unit (CPU) 1001, the above-mentioned functions defined in the system of the present application are executed.

[0217] It should be noted that the computer-readable medium shown in this application can be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or device, or any combination of the above. More specific examples of computer-readable storage media can include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this application, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, device, or device. In this application, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, which carries computer-readable program code. This propagated data signal can take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transfer a program for use by or in conjunction with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium may be transmitted using any suitable medium, including but not limited to wireless, wire, optical cable, RF, or any suitable combination thereof.

[0218] The flowcharts and block diagrams in the accompanying drawings illustrate the possible implementation architecture, functions and operating instructions of the systems, methods and computer program products according to various embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, program segment, or a part of code, and the aforementioned module, program segment, or a part of code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order than the order marked in the accompanying drawings. For example, the boxes represented by two connections can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of the boxes in the block diagram and / or flowchart, can be implemented using a dedicated hardware-based system that performs the specified function or operating instruction, or can be implemented using a combination of dedicated hardware and computer instructions.

[0219] The units or modules involved in the embodiments described in the present application can be implemented by software or hardware. The units or modules described can also be set in a processor. For example, they can be described as: a processor includes an acquisition module, an analysis module, an extraction module and a marking module. Among them, the names of these units or modules do not constitute a limitation of the units or modules themselves under certain circumstances. For example, the first acquisition module can also be described as "acquiring the first monitoring video of the target seat when the mark corresponding to the target seat is idle."

[0220] As another aspect, the present application further provides a computer-readable storage medium, which may be included in the electronic device described in the above embodiments, or may exist independently and not incorporated into the electronic device. The computer-readable storage medium stores one or more programs, which, when used by one or more processors, execute the seat status management method described in the present application.

[0221] The above description is merely a preferred embodiment of the present application and an illustration of the technical principles employed. Those skilled in the art should understand that the scope of disclosure in this application is not limited to the technical solutions formed by a specific combination of the above-mentioned technical features, but also encompasses other technical solutions formed by any combination of the above-mentioned technical features or their equivalents without departing from the aforementioned disclosed concepts. For example, a technical solution formed by replacing the above-mentioned features with (but not limited to) technical features with similar functions disclosed in this application.

Claims

1. A seat status management method, characterized in that: include: When the mark corresponding to the target seat is vacant, obtaining a first monitoring video of the target seat; Based on the first surveillance video, determining that a first person exists in the target seat area and that the first person is sitting in the target seat; Obtaining a first person feature and at least one first object feature corresponding to the first person from the first surveillance video; the first person feature is a partial image of the first person and abstract feature data obtained by performing convolution image processing on the first person; the first object feature is a partial or full image of the first object and abstract feature data obtained by performing convolution image processing on the first object; Marking the target seat as occupied, and binding the first person feature and the at least one first object feature to the current occupancy mark of the target seat as feature information corresponding to the current occupancy mark of the target seat; Wherein, after marking the target seat as occupied, the method further includes: Continue to obtain a first surveillance video corresponding to the target seat, and when it is determined based on the first surveillance video that the target seat area changes from the presence of the first person to the absence of the first person, obtain a second surveillance video corresponding to the target seat; Identifying whether the at least one first object feature exists in the second surveillance video; When the at least one first object feature does not exist in the second surveillance video, the target seat is marked as vacant.

2. The method according to claim 1, characterized in that The determining, based on the first surveillance video, that a first person exists in the target seat area and that the first person is sitting in the target seat includes: Inputting the first surveillance video into a behavior detection model to obtain pedestrian feature information, the pedestrian feature information including a bounding box for framing pedestrians in the surveillance video; When the pedestrian characteristic information is within the reference area of ​​the target seat, determining that a first person exists in the reference area, and the pedestrian included in the pedestrian characteristic information is the first person; Inputting the first surveillance video into a behavior posture analysis model to obtain posture information of the first person; When the posture information is a sitting posture, it is determined that the first person is sitting on the target seat.

3. The method according to claim 1, characterized in that The obtaining, from the first surveillance video, a first object feature corresponding to the first person, includes: extracting at least one object feature from the first surveillance video; For each of the object features, obtaining a first distance between the object feature and the pedestrian feature information corresponding to the first person; The feature of the object whose first distance is less than or equal to the first preset distance is used as the first feature of the object corresponding to the first person.

4. The method according to claim 1, wherein Also includes: When the first object feature exists in the second surveillance video, obtaining a cumulative duration of the second surveillance video; When the accumulated time is greater than or equal to a first preset time length, an alarm message is sent according to a first preset rule.

5. The method according to claim 1, wherein Also includes: When the at least one first item feature is present in the second surveillance video, identifying a first value corresponding to the first item feature based on the at least one first item feature; When the first value is less than or equal to a first preset value threshold, an alarm message is sent according to a second preset rule.

6. The method according to claim 4, characterized in that Also includes: determining, based on the second surveillance video, that a second person exists in the target seat area and that the second person is sitting in the target seat, and sending an alarm message according to a third preset rule; The second character feature corresponding to the second character is different from the first character feature corresponding to the first character.

7. The method according to any one of claims 4 to 6, characterized in that: Also includes: Counting the alarm time, and determining whether the alarm time reaches a second preset time length when the alarm condition is met next time; When the alarm time does not reach the second preset time length, the current alarm is canceled.

8. A seat status management device, characterized in that: include: an acquisition module, configured to acquire a first monitoring video of a target seat when the mark corresponding to the target seat is vacant; an analysis module, configured to determine, based on the first surveillance video, that a first person exists in the target seat area and that the first person is sitting in the target seat; an extraction module, configured to obtain, from the first surveillance video, a first person feature and a first object feature corresponding to the first person; the first person feature being a partial image of the first person and abstract feature data obtained by performing convolution image processing on the first person; and the first object feature being a partial or full image of the first object and abstract feature data obtained by performing convolution image processing on the first object; a marking module, configured to mark the target seat as occupied, and bind the first person feature and the first object feature to the target seat as feature information corresponding to the target seat being marked as currently occupied; Wherein, after marking the target seat as occupied, the method further includes: Continue to obtain a first surveillance video corresponding to the target seat, and when it is determined based on the first surveillance video that the target seat area changes from the presence of the first person to the absence of the first person, obtain a second surveillance video corresponding to the target seat; Identifying whether the at least one first object feature exists in the second surveillance video; When the at least one first object feature does not exist in the second surveillance video, the target seat is marked as vacant.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the seat status management method according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the seat status management method according to any one of claims 1 to 7 is implemented.

Images