Monitoring prompting method and device, and storage medium

Abstract

The embodiment of the invention discloses a monitoring prompting method and device, and a storage medium, and the method comprises the steps: obtaining a target monitoring video, the target monitoring video comprises at least two sub-regions, and the at least two sub-regions correspond to different image positions in the monitoring video; information of an event in the target monitoring video is obtained, and the information of the event comprises a target sub-region where the event occurs in the at least two sub-regions and at least one of the following information: event occurrence time, a target object and an event type; and sending the information of the event to the terminal equipment, so that the terminal equipment provides an event prompt for a user according to the information of the event. And the prompt information with more comprehensive content can be obtained.

Description

Technical Field

[0001] The embodiments of the present application relate to video processing technology, and relate to a monitoring prompt method and device, and a storage medium. Background Art

[0002] The video alarm function is usually used in the fields of security, monitoring, etc. It can determine whether to generate prompt information based on the monitoring content in the video, thereby realizing the alarm of the monitoring content.

[0003] In the related art, the monitoring prompt method generally adopts to determine whether the target object appears in the monitoring image, and when the target object appears, the corresponding prompt information is generated to realize the alarm.

[0004] However, in actual applications, generating prompt information only when the target object appears will result in the prompt content being too vague and too one-sided, making it impossible to obtain prompt information with more comprehensive content. Summary of the invention

[0005] In view of this, the monitoring prompt method, device, and storage medium provided in the embodiments of the present application can obtain prompt information with more comprehensive content. The monitoring prompt method, device, and storage medium provided in the embodiments of the present application are implemented as follows:

[0006] In one aspect of an embodiment of the present application, a monitoring prompt method is provided, which is applied to a shooting device, the shooting device is communicatively connected with a terminal device, and the method includes:

[0007] Acquire a target surveillance video, where the target surveillance video includes at least two sub-areas, and the at least two sub-areas correspond to different image positions in the surveillance video respectively;

[0008] Acquire information about an event in a target surveillance video, the event information including: a target sub-region where the event occurs in at least two sub-regions, and at least one of the following information: event occurrence time, target object, and event type;

[0009] The event information is sent to the terminal device so that the terminal device provides the user with an event prompt according to the event information.

[0010] In the monitoring prompt method provided in the embodiment of the present application, event information in the target monitoring video can be obtained, and the information includes the target sub-area where the event occurs, and at least one of the event occurrence time, target object and event type, so as to provide users with more specific and comprehensive event prompts. The information of the sub-area can further make the prompt content of the event more accurate, thereby achieving more precise event monitoring.

[0011] The target monitoring video includes at least two sub-areas, which means that the image corresponding to the target monitoring video includes at least two sub-areas.

[0012] In one embodiment, obtaining information about an event in a target surveillance video includes: performing event detection in at least two sub-areas in the target surveillance video, determining the target sub-area where the event occurred, and obtaining at least one of the following information: the time of event occurrence, the target object, and the event type, and the union of at least two sub-areas is smaller than the image corresponding to the surveillance video. In the monitoring prompt method provided in the embodiment of the present application, event detection is only performed on the sub-area that the user is concerned about, which can avoid interference from invalid information on the one hand, and reduce the amount of device calculation corresponding to event detection on the other hand.

[0013] It should be noted that, in one implementation, the union of at least two sub-regions is the same as the image area of ​​the surveillance video. Exemplarily, the at least two sub-regions are obtained by image segmentation of the surveillance video; or, in another implementation, the union of at least two sub-regions is smaller than the image area of ​​the surveillance video. Exemplarily, the union of at least two sub-regions is part of the image area of ​​the surveillance video, that is, a partial area of ​​interest selected from the surveillance image.

[0014] In one embodiment, obtaining information about events in a target surveillance video includes: performing target detection on the target surveillance video to determine a position of a target detection frame in an image, where the target detection frame is used to indicate an object that appears in the image; if the position of the target detection frame in the image overlaps with the position of a target sub-region in the image, obtaining information about events in the target sub-region in the target surveillance video.

[0015] In this embodiment, by determining that the detection frame overlaps with the image position of the sub-region, the target sub-region where the event occurs can be accurately and quickly determined, that is, the information of the target sub-region in the event information is obtained.

[0016] In one embodiment, obtaining information about events in a target sub-area in a target surveillance video includes at least one of the following: identifying the identity of an object to determine the target object in the target surveillance video; identifying the action of the object to determine the event type based on the identified action; and obtaining the event occurrence time when the object performs the action.

[0017] In this embodiment, the identity and action of the object appearing in the target surveillance video can be recognized, and the time of the event can be determined, so that the information of the event in the target surveillance video can be obtained more comprehensively and accurately.

[0018] In one embodiment, before sending the event information to the terminal device, the method further includes: inputting the target sub-region, and at least one of the target object, event occurrence time, and event type into a video understanding model for analysis, and outputting the programmed event information, wherein the video understanding model is trained based on the sample event information and the programmed sample event information, wherein the sample event information includes: a sample sub-region, and at least one of the sample object, sample event occurrence time, and sample event type; sending the event information to the terminal device includes: sending the programmed event information to the terminal device.

[0019] In one embodiment, the information of the organized event includes event information expressed in the form of sentences;

[0020] In one embodiment, the organized event information includes summary information of the event information obtained by statistically analyzing information of multiple events according to preset rules.

[0021] In one embodiment, the arranged events include information of the events sorted in sequence.

[0022] In this embodiment, the analysis performed through the video understanding model can realize the arrangement of event information, so that the information of the arranged event can be obtained more quickly through the video understanding model. It can be understood that the arranged event information includes richer and easier to understand event content, which can enable users to accurately obtain more comprehensive event information. In some possible implementation methods, the shooting device only sends the arranged event information to the terminal device, but does not send the video corresponding to the event. In this implementation method, the traffic consumed by the shooting device for information transmission can be greatly reduced, and it is highly applicable in some scenarios with poor network environment.

[0023] In one embodiment, before acquiring the target surveillance video, the method also includes: sending image information of the surveillance scene corresponding to the shooting device to the terminal device; receiving a first area division instruction determined based on the image information of the surveillance scene sent by the terminal device, the first area division instruction being used to indicate the position of each sub-area in the target surveillance video in the image.

[0024] In this embodiment, by obtaining the first area division instruction, the shooting device can determine the position of each sub-area in the target monitoring video in the image, accurately monitor the area of ​​interest to the user, avoid invalid alarm events caused by irrelevant areas, and improve the accuracy of event prompt information.

[0025] In one embodiment, before acquiring the target surveillance video, the method also includes: determining multiple initial sub-areas from an image of the surveillance scene corresponding to the shooting device; sending image information of the surveillance scene carrying multiple initial sub-areas to the terminal device; receiving a second area division instruction sent by the terminal device that is determined based on the image information of the surveillance scene carrying multiple initial sub-areas, the second area division instruction being used to indicate the position of each sub-area in the target surveillance video in the image, each sub-area being at least two of the multiple initial sub-areas.

[0026] Determine multiple initial sub-areas from the image of the monitoring scene corresponding to the shooting device, including: dividing the image of the monitoring scene corresponding to the shooting device into regions to obtain multiple initial sub-areas; or performing intelligent region recognition on the image of the monitoring scene corresponding to the shooting device to obtain multiple initial sub-areas.

[0027] In this embodiment, the shooting device can intelligently identify the monitoring scene, determine multiple initial sub-areas and then send them to the terminal device. By obtaining the second area division instruction, the position of each sub-area in the target monitoring video in the image is determined, so that the target sub-area where the event occurs can be accurately determined.

[0028] In one embodiment, the method further includes: sending a video corresponding to the event to a terminal device. In a possible implementation, the video only includes an image of a sub-region where the event occurs.

[0029] In this embodiment, by sending the video corresponding to the event to the terminal device, the terminal device can generate more comprehensive and accurate prompt information in the process of prompting the user, thereby increasing the accuracy and completeness of the prompt information.

[0030] In a possible implementation manner, the video corresponding to the event sent to the terminal device only includes an image of the sub-region where the event occurs.

[0031] In this embodiment, since only the image of the sub-area corresponding to the event is sent to the terminal device, the amount of data transmission can be reduced, and it is also helpful for users to clearly and directly watch the video content of interest, effectively reducing the difficulty of users repeatedly observing and searching for target events in large images.

[0032] In another aspect of the embodiment of the present application, a monitoring prompt method is provided, which is applied to a terminal device, the terminal device is communicatively connected with a shooting device, and the method includes:

[0033] Sending a region division instruction to the photographing device, where the region division instruction is used to indicate the position of each sub-region in the image taken by the photographing device, and any two sub-regions in each sub-region correspond to different positions in the image;

[0034] Acquire information about an event in a target surveillance video sent by a shooting device, the target surveillance video including at least two sub-areas determined according to an area division instruction, and the event information including: a target sub-area in which the event occurs in the at least two sub-areas, and at least one of the following information: event occurrence time, target object, and event type;

[0035] Prompt information is generated according to the event information, and the prompt information includes a target sub-region where the event occurs.

[0036] In the monitoring prompt method provided in the embodiment of the present application, event information based on the target sub-area in the target monitoring video can be obtained, and the information can also include at least one of the event occurrence time, target object and event type. The sub-area information can further make the event prompt content more accurate, thereby achieving more precise event monitoring.

[0037] In one embodiment, prompt information is generated based on event information, including: determining prompt text based on event information and a sentence organization model, wherein the sentence organization model is trained based on sample event information and sample prompt text; generating prompt information, the prompt information including the prompt text and a video clip of the target video in a time interval corresponding to the event.

[0038] In this embodiment, prompt information can be clearly and unambiguously output based on the sentence organization model, so that users can obtain the content of the prompt more accurately. Even if they do not watch the video corresponding to the event, they can accurately obtain relatively accurate event content, which can improve the efficiency of users in obtaining event information.

[0039] In one embodiment, after generating prompt information according to event information, the method further includes: generating first integrated information of the target object according to prompt information for the same target object within a preset time period, the first integrated information being used to characterize various events executed by the target object.

[0040] In this embodiment, first integrated information for the target object can be generated based on prompt information for the same target object, and analysis of behavioral events for the target object is further provided, thereby obtaining information such as the behavioral habits of the target object.

[0041] In one embodiment, after generating prompt information according to event information, the method further includes: generating second integrated information of the target sub-area according to prompt information for the same sub-area within a preset time period, the second integrated information being used to characterize each event executed in the target sub-area.

[0042] In this embodiment, second integrated information for the target object can be generated based on the prompt information for the same sub-area, further providing users with statistical information with reference value. For example, in a store monitoring scenario, different shelf areas correspond to different sub-areas, and the popularity of different shelf areas can be obtained through the second integrated information.

[0043] In one embodiment, the method also includes: generating recommendation information for the target object and the target event when the frequency of executing the target event in the first integrated information meets the preset frequency requirement; and / or generating recommendation information for the target sub-area and the target event when the frequency of executing the target event in the second integrated information meets the preset frequency requirement.

[0044] In this embodiment, recommendation information for a certain target object or a certain target sub-region can be implemented based on the first integrated information for the same target object or the second integrated information for the same sub-region, thereby improving the accuracy and rationality of the recommendation.

[0045] In one embodiment, before sending a region division instruction to the shooting device, the method also includes: acquiring an image captured by the shooting device; segmenting the image to obtain a plurality of blocks; and determining a sub-region based on a user's selection from the plurality of blocks, wherein the sub-region is a collection of target blocks selected by the user from the plurality of blocks.

[0046] In this embodiment, the terminal device can segment the image information of the monitoring scene into multiple blocks, and then determine the image position of each sub-area in the target monitoring video based on the user's selection, thereby improving the efficiency and accuracy of determining the sub-area position.

[0047] In a possible implementation, the terminal device divides the monitoring image into a plurality of equal-proportion blocks, and determines a set of the plurality of equal-proportion blocks as a sub-region according to user selection. Optionally, the user determines a set of blocks included in the plurality of sub-regions respectively.

[0048] In this embodiment, a convenient implementation method for determining a sub-region is provided, which can reduce the difficulty of setting a sub-region for a user.

[0049] In one embodiment, before sending a region division instruction to the shooting device, the method also includes: receiving an image carrying multiple initial sub-regions sent by the shooting device, the initial sub-regions being obtained by region division of the image shot by the shooting device; and determining, in response to a selection operation for the initial sub-regions, a position of each sub-region in the target surveillance video in the image, each sub-region in the target surveillance video being at least two of the multiple initial sub-regions.

[0050] In this embodiment, an image carrying multiple initial sub-regions may be received by the terminal device, and a user may select each sub-region in the target surveillance video from the multiple initial sub-regions, thereby improving the efficiency of determining the sub-region position.

[0051] In another aspect of the embodiment of the present application, a monitoring prompt method is provided, which is applied to a shooting device, the shooting device is communicatively connected with a terminal device, and the method includes:

[0052] Acquire a target surveillance video, where the target surveillance video includes at least two sub-areas, and the at least two sub-areas correspond to different image positions in the surveillance video respectively;

[0053] The target surveillance video is sent to the terminal device so that the terminal device can determine the information of the event in the target surveillance video and provide the user with an event prompt based on the event information. The event information includes: the target sub-area where the event occurs in at least two sub-areas, and at least one of the following information: the time of event occurrence, the target object and the event type.

[0054] In the monitoring prompt method provided in the embodiment of the present application, the shooting device can acquire the target monitoring video, and the terminal device can determine the event information and provide event prompts. In addition, the information can include the target sub-area where the event occurs, the time of the event, the target object and the event type, so that event prompts can be provided to users based on more comprehensive event information, and the event prompt content can be further made more accurate, thereby achieving more precise event monitoring.

[0055] In another aspect of the embodiment of the present application, a monitoring prompt method is provided, which is applied to a terminal device, the terminal device is communicatively connected with a shooting device, and the method includes:

[0056] Acquire a target surveillance video sent by a shooting device, where the target surveillance video includes at least two sub-areas, and the at least two sub-areas correspond to different image positions in the surveillance video respectively;

[0057] Acquire information about an event in a target surveillance video, the event information including: a target sub-region where the event occurs in at least two sub-regions, and at least one of the following information: event occurrence time, target object, and event type;

[0058] Prompt information is generated according to the event information, and the prompt information includes a target sub-region where the event occurs.

[0059] In the monitoring prompt method provided in the embodiment of the present application, the target monitoring video can be obtained, and then the event information in the target monitoring video can be determined, and the information can include the target sub-area where the event occurs, the time when the event occurs, the target object and the event type, so that event prompts can be provided to users based on more comprehensive event information, and the prompt content of the event can be further made more accurate, thereby achieving more precise event monitoring.

[0060] The computer device provided in the embodiment of the present application includes a memory and a processor. The memory stores a computer program that can be run on the processor. When the processor executes the program, the method of the embodiment of the present application is implemented.

[0061] The computer-readable storage medium provided in the embodiment of the present application has a computer program stored thereon, and when the computer program is executed by a processor, the method provided in the embodiment of the present application is implemented.

[0062] In the monitoring prompt method, device, and storage medium provided in the embodiments of the present application, a target monitoring video can be obtained, the target monitoring video includes at least two sub-areas, and the at least two sub-areas correspond to different image positions in the monitoring video; information about events in the target monitoring video is obtained; and the event information is sent to a terminal device, so that the terminal device provides event prompts to the user based on the event information. Among them, event information in the target monitoring video can be obtained, and the information can include the target sub-area where the event occurs, the time of event occurrence, the target object, and the event type, so that event prompts can be provided to the user based on more comprehensive event information, which can further make the event prompt content more accurate and achieve more accurate event monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0064] Figure 1 A schematic diagram of an application scenario of the monitoring prompt method provided in an embodiment of the present application;

[0065] Figure 2 A schematic diagram of implementing a monitoring prompt method for different electronic devices provided in an embodiment of the present application;

[0066] Figure 3 A flowchart of a monitoring prompt method provided in an embodiment of the present application;

[0067] Figure 4A schematic diagram of the structure of a video understanding model provided in an embodiment of the present application;

[0068] Figure 5 A schematic diagram of sub-region division provided in an embodiment of the present application;

[0069] Figure 6 Another schematic diagram of the monitoring prompt method provided in an embodiment of the present application;

[0070] Figure 7 A schematic diagram of the structure of a sentence arrangement model provided in an embodiment of the present application;

[0071] Figure 8 A schematic diagram of the generated prompt information provided in an embodiment of the present application;

[0072] Fig. 9 This is a schematic diagram of one of the processes for generating recommendation information provided in an embodiment of the present application;

[0073] Fig.10 Another schematic diagram of a process for generating recommendation information provided in an embodiment of the present application;

[0074] Fig.11 A schematic diagram of the interaction between a photographing device and a terminal device provided in an embodiment of the present application;

[0075] Fig.12 Another schematic diagram of interaction between a photographing device and a terminal device provided in an embodiment of the present application;

[0076] Fig.13 This is a schematic diagram of the structure of the monitoring prompt device provided in an embodiment of the present application;

[0077] Fig.14 Another structural schematic diagram of the monitoring prompt device provided in an embodiment of the present application;

[0078] Fig.15 A schematic diagram of the structure of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0079] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the specific technical solution of the present application will be further described in detail below in conjunction with the drawings in the embodiments of the present application. The following embodiments are used to illustrate the present application, but are not used to limit the scope of the present application.

[0080] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0081] In the following description, reference is made to “some embodiments”, which describe a subset of all possible embodiments, but it will be understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0082] It should be pointed out that the terms "first\second\third" involved in the embodiments of the present application are used to distinguish similar or different objects, and do not represent a specific ordering of the objects. It can be understood that "first\second\third" can be interchanged in a specific order or sequence where permitted, so that the embodiments of the present application described here can be implemented in an order other than that illustrated or described here.

[0083] Figure 1 This is a schematic diagram of an application scenario of the monitoring prompt method provided in the embodiment of the present application. Please refer to Figure 1 The scene may include a shooting device 110 and a terminal device 120 that can be communicatively connected to the shooting device 110.

[0084] It should be noted that the monitoring prompt method provided in the embodiment of the present application can be implemented by the shooting device 110 and the terminal device 120 as the execution subjects respectively, and the steps that the shooting device 110 needs to execute and the steps that the terminal device 120 needs to execute can be set according to actual needs.

[0085] Optionally, the method can be applied to any electronic device of the shooting device 110 or the terminal device 120, wherein the shooting device 110 can be, for example, a surveillance camera set up in a public or private place, a camera shooting at a fixed position, or a movable image acquisition device, etc., without specific limitation herein. The terminal device 120 can include, for example, but is not limited to, a mobile phone, a wearable device (such as a smart watch, a smart bracelet, smart glasses, etc.), a tablet computer, a laptop computer, a vehicle-mounted terminal, a PC (Personal Computer), etc.

[0086] The functions implemented by the method can be implemented by calling program codes through the processor of any of the two types of electronic devices mentioned above. Of course, the program codes can be stored in computer storage media. It can be seen that the shooting device 110 and the terminal device 120 at least include a processor and a storage medium.

[0087] Figure 1 The scene shown in the figure is a scene used by the shooting device 110 and the terminal device 120. In the actual implementation process, the shooting device 110 with a data processing function or the terminal device 120 with a shooting capability can also be completed by itself, and no specific limitation is made here. Figure 1 The implementation process shown in the figure is a scenario where the shooting device 110 and the terminal device 120 are used together.

[0088] Optionally, in a scenario where the shooting device 110 and the terminal device 120 are used in combination, the shooting device 110 can be set in a preset monitoring area to achieve shooting in the preset monitoring area, and then the monitoring video in the preset monitoring area can be obtained; the terminal device 120 can maintain communication with the shooting device 110, for example: it can be communication between a mobile phone and a camera.

[0089] Please refer to Figure 1 In this scene, multiple areas may be included: kitchen, bar, front door, dining room, living room, public bathroom, bedroom door, etc., and surveillance videos of these areas in the scene may be obtained through the shooting device 110. Among them, different surveillance videos may be obtained by adjusting the shooting angle of the shooting device 110, and the content captured by the same surveillance video may include multiple of the above areas.

[0090] It should be noted that, in the case where there are multiple areas, the area division can be implemented by the shooting device 110, or the area division can also be implemented by the terminal device, and no specific limitation is made here.

[0091] Optionally, the above-mentioned multiple areas all have corresponding image positions in the surveillance video, each image position can serve as a sub-area in the surveillance video, and the content in the image position corresponding to the sub-area can be one of the above-mentioned multiple areas. For example, the image content corresponding to the restaurant area in the surveillance video can be used as the sub-area of ​​the restaurant.

[0092] The monitoring prompt method provided in the embodiment of the present application is explained below using the shooting device to implement area division and the terminal device to implement area division respectively.

[0093] Figure 2 For a schematic diagram of a monitoring prompt method for different electronic devices provided in the embodiments of the present application, please refer to Figure 2 , Figure 2 The above process is an implementation process of performing video acquisition and image analysis through the shooting device 110 and generating prompts through the terminal device 120. Figure 2 The process below is an implementation process of acquiring video through the shooting device 110, performing image analysis through the terminal device 120, and generating prompts.

[0094] The video acquisition refers to the process of acquiring the surveillance video, for example, the shooting device 110 acquires the target surveillance video by shooting the video in the scene.

[0095] Image analysis refers to the process of analyzing events based on captured target surveillance videos, for example, determining information about events in target surveillance videos.

[0096] Generating a prompt refers to the process of generating a monitoring prompt based on the information of the event, for example, issuing an alarm based on the information of the event.

[0097] It should be noted that, in the actual implementation process, any one of the above two monitoring prompt methods can be adopted, and there is no specific limitation here. The corresponding setting method can be selected according to the actual function of the electronic device.

[0098] In the art, the monitoring prompt method generally adopts determining whether a target object appears in a monitoring image, and generating corresponding prompt information when the target object appears, thereby realizing an alarm.

[0099] However, in actual applications, generating prompt information only when the target object appears will result in the prompt content being too vague and too one-sided, making it impossible to obtain prompt information with more comprehensive content.

[0100] In order to solve the above-mentioned drawbacks existing in the related art, a monitoring prompt method is provided in an embodiment of the present application, and the implementation process of the monitoring prompt method is explained below.

[0101] Figure 3 Please refer to the flowchart of the monitoring prompt method provided in the embodiment of this application. Figure 3 , the method comprising:

[0102] S310: Obtain target surveillance video.

[0103] The target surveillance video includes at least two sub-areas, and the at least two sub-areas correspond to different image positions in the surveillance video respectively.

[0104] It should be noted that the target surveillance video can be a video shot in a preset surveillance area, and the preset surveillance area can be any place that needs to be monitored, such as: private residences, supermarkets, hotels, villas, art exhibition halls, shops, etc., without specific restrictions here.

[0105] Optionally, the target surveillance video may include at least two sub-areas, each of which may correspond to an area in the actual scene, for example Figure 1 various areas in .

[0106] In one embodiment, the target monitoring video can be determined according to the angle at which the shooting device is set in a preset monitoring area. For example, in a private residence scenario, the shooting device can capture videos in three rooms: the living room, the kitchen, and the balcony. Then the sub-areas of the target monitoring video may include the three sub-areas of the living room, the kitchen, and the balcony.

[0107] It should be noted that the shooting device in the present application can be a single-lens shooting device, i.e. a single-camera device, or a dual-lens or multi-lens shooting device, i.e. a dual-camera device or a multi-camera device. It can be understood that for a shooting device that expands the shooting field of view through two or more lenses, the images captured are often larger in size, and the positioning of monitoring events in the image will be relatively more difficult.

[0108] It should be noted that the shooting device in the present application can be a shooting device with a fixed shooting angle, such as a shooting device fixedly installed on the wall of a room; it can also be a shooting device with a changeable shooting angle, such as a pan-tilt camera. The shooting device changes the shooting angle during the shooting process by means of methods such as pan-tilt rotation, and the area corresponding to the acquired image in the physical space also changes accordingly. In one possible implementation method, it can be set to maintain the selection of the original physical space area after the camera angle is adjusted, that is, the position of the sub-area in the image changes according to the rotation of the camera angle, for example, spatial calibration is performed through image analysis and recognition. In this implementation method, the rotation of the camera can also be set to be limited to an angle range that does not lose the image of the physical area of ​​interest. Within the range, for example, the physical space corresponding to the sub-area set by the user is the living room area, and the camera device can be kept fixed and not rotated, or rotate but can always capture the living room area; in another possible implementation method, the user can obtain the panoramic image that the shooting device can capture and set the sub-area of ​​interest. The actual shooting range may be a part of the panoramic image. The sub-area in the target monitoring video can be changed by rotation. For example: when the shooting device is at a first angle, the living room and kitchen can be captured, and when the shooting device is at a second angle, the living room and bedroom can be captured. In the case of the first angle, it can be determined that the sub-area includes the living room and kitchen, and in the case of the second angle, it can be determined that the sub-area includes the living room and bedroom.

[0109] S320: Obtain information about events in the target surveillance video.

[0110] The event information includes: a target sub-region where the event occurs in at least two sub-regions, and at least one of the following information: event occurrence time, target object, and event type.

[0111] It should be noted that image analysis may be performed on the target surveillance video to determine information about events in the surveillance video.

[0112] Among them, an event can be a behavior event of an object, or an event occurring in an area, wherein a behavior event can be an event generated by a behavior object, wherein the behavior object can be a person, animal, plant, etc., without specific limitation. The generated event can be an activity performed by the behavior object, for example: user A is cooking, then cooking can be a behavior event; user B is exercising, then exercising can be a behavior event. Events occurring in an area can be events triggered in an area, for example, user A appears in a certain area, and user B leaves a certain area, then both appearance and leaving can be events occurring in a certain area.

[0113] The event information may be relevant information of the event that occurred, wherein the target sub-area may be the area where the event occurred, for example: if an event occurred in sub-area A, then for the event, sub-area A is the target sub-area, and the target sub-area may be one of at least two sub-areas of the target surveillance video.

[0114] It should be noted that the event occurrence time can be the time point when the event is started, or the time interval during which the event is continuously executed. For example, if the event is opening a door, and the duration of this action is short, the event occurrence time can be the time point when the door opening action is started, such as opening the door at 8:00 in the morning. Correspondingly, if the event is eating, and the duration of this action is long, the event occurrence time can be the time interval during which the eating action is continuously executed, such as eating from 11:30 to 12:00 noon.

[0115] It should be noted that the time of occurrence of an event can be represented by either a time point or a time interval, and there is no specific limitation here. A threshold can be set in advance. When the duration is greater than the threshold, the time interval of continuous execution of the event can be used to represent the time of occurrence of the event. When the duration is less than or equal to the threshold, the time point when the event is started can be used to represent the time of occurrence of the event.

[0116] The target object may be the behavior object that executes the event. For example, if the event is fitness and user B is exercising, user B is the target object.

[0117] The event type may be a specific classification of the event, for example, different events such as cooking, eating, and fitness can all be used as different event types.

[0118] Optionally, information about events in the target surveillance video can be obtained through image analysis, such as recognition and processing of image content. For example, if it is determined that a user is cooking in sub-area A in the target video, the cooking event occurring in the sub-area A can be obtained through image recognition, and the corresponding event occurrence time, target object, event type and other information can be determined.

[0119] S330: Sending the event information to the terminal device, so that the terminal device provides the user with an event prompt according to the event information.

[0120] After the photographing device obtains the event information, it can send the event information to the terminal device.

[0121] It should be noted that after obtaining the event information, the terminal device can generate prompt information according to the event information to provide the user with an event prompt.

[0122] It should be noted that the prompt information can be a warning information, and the warning prompt can be made in a pre-configured format, for example: it can prompt a specific behavior event generated by a user within a certain period of time, or it can also prompt a specific event occurring in a certain sub-area within a certain period of time.

[0123] For example, if prompt information is set for user A, the corresponding prompt information can be generated after obtaining that user A performs a certain behavior event; if prompt information is set for sub-area C, the corresponding prompt information can be generated after obtaining that a certain behavior event occurs in sub-area C.

[0124] In one embodiment, in addition to using the above-mentioned prompt method, a prompt can also be generated after any event occurs. For example, if user A is cooking in the kitchen at 8:00 in the morning, a prompt message for the event can be generated. In addition to prompting that an event that requires a prompt has occurred, corresponding prompt information can also be generated based on the information of the event, such as: prompting a specific time, prompting a specific object, prompting a sub-area where the event occurred, or prompting a specific event type, etc., without specific limitations here.

[0125] In the monitoring prompt method provided in the embodiment of the present application, a target monitoring video can be obtained, the target monitoring video includes at least two sub-areas, and the at least two sub-areas correspond to different image positions in the monitoring video; information about events in the target monitoring video is obtained; and the event information is sent to a terminal device, so that the terminal device provides event prompts to the user based on the event information. Among them, event information in the target monitoring video can be obtained, and the information includes the target sub-area where the event occurs, as well as at least one of the event occurrence time, target object, and event type, so that a more specific and comprehensive event prompt can be provided to the user, and the information of the sub-area can further make the event prompt content more accurate, so as to achieve more accurate event monitoring.

[0126] It should be noted that the target monitoring video includes at least two sub-areas, which means that the image corresponding to the target monitoring video includes at least two sub-areas.

[0127] In one embodiment, obtaining information about an event in a target surveillance video includes: performing event detection in at least two sub-areas in the target surveillance video, determining the target sub-area where the event occurred, obtaining at least one of the following information: event occurrence time, target object, and event type, and the union of at least two sub-areas is smaller than an image corresponding to the surveillance video.

[0128] For example, the target surveillance video may include three areas, and event detection may be performed on all three areas, or on at least two sub-areas, without specific limitation herein.

[0129] Optionally, if the target surveillance video needs to be fully monitored, event detection can be performed on all sub-areas in the target surveillance video; if only part of the target surveillance video needs to be monitored, event detection can be performed on at least two corresponding sub-areas, without specific limitation here.

[0130] In the monitoring prompt method provided in the embodiment of the present application, event detection is performed only on the sub-area that the user is concerned about. On the one hand, it can avoid interference from invalid information, and on the other hand, it can reduce the amount of device calculation corresponding to event detection.

[0131] It should be noted that, in one implementation, the union of at least two sub-regions is the same as the image area of ​​the surveillance video. Exemplarily, the at least two sub-regions are obtained by image segmentation of the surveillance video; or, in another implementation, the union of at least two sub-regions is smaller than the image area of ​​the surveillance video. Exemplarily, the union of at least two sub-regions is part of the image area of ​​the surveillance video, that is, a partial area of ​​interest selected from the surveillance image.

[0132] The image may be selected actively by the user, or by a terminal device or a photographing device, and is not specifically limited here.

[0133] In one embodiment, obtaining information about events in a target surveillance video includes: performing target detection on the target surveillance video to determine a position of a target detection frame in an image, where the target detection frame is used to indicate an object that appears in the image; if the position of the target detection frame in the image overlaps with the position of a target sub-region in the image, obtaining information about events in the target sub-region in the target surveillance video.

[0134] It should be noted that what is to be detected by the target detection may be the position of the target detection frame, and the target detection frame may be used to indicate an object appearing in the image.

[0135] For example, if an object such as a person, an animal or a plant appears in an image, the object can be framed by a target detection frame. During the target detection process, the position of the target detection frame in the image can be determined.

[0136] Optionally, the corresponding target sub-region can be determined according to the position of the target detection frame in the image. If the position of the target detection frame in the image overlaps with the position of any sub-region in the image, it can be determined that the sub-region can be used as the target sub-region, and the information of the event of the target sub-region in the target monitoring video can be obtained.

[0137] Among them, the corresponding object can be framed by the target detection frame according to the dynamic changes of the target monitoring video.

[0138] It should be noted that if a dynamic object appears in a static video, the object can be framed out using a target detection frame. For example, if a walking person appears in the video, the person can be framed out using a target detection frame and the person can be regarded as an object; or, if a running pet appears in the video, the pet can be framed out using a target detection frame and the pet can also be regarded as an object.

[0139] In one embodiment, the target detection frame may be a square detection frame or a circular detection frame, and no specific limitation is given here, as long as the object can be framed.

[0140] In the monitoring prompt method provided in the embodiment of the present application, target detection can be performed on the target monitoring video to determine the position of the target detection frame in the image; if the position of the target detection frame in the image overlaps with the position of the target sub-region in the image, the information of the event of the target sub-region in the target monitoring video is obtained. Among them, by determining that the image position of the detection frame overlaps with the sub-region, the target sub-region where the event occurred can be accurately and quickly determined, that is, the information of the target sub-region in the event information is obtained.

[0141] In one embodiment, obtaining information about events in a target sub-area in a target surveillance video includes at least one of the following: identifying the identity of an object to determine the target object in the target surveillance video; identifying the action of the object to determine the event type based on the identified action; and obtaining the event occurrence time when the object performs the action.

[0142] The identity recognition may be, for example, face recognition.

[0143] It should be noted that the target object can be a preset object. For example, after a user appears in the sub-area, facial recognition and other tasks can be performed on the user to determine the user's identity. If the user is the preset object, then the user can be used as the target object. If the user is not the preset object, then it is determined that the user is not the target object.

[0144] For example, if user A needs to be monitored, user A can be the target object. After a user appears in a sub-area, face recognition processing can be performed on the user. If it is user A, it can be determined that the user is the target object. If it is not user A, it can be determined that the user is not the target object.

[0145] It should be noted that multiple facial portraits can be pre-set in the electronic device, and the facial portrait of the user that appears can be obtained through facial recognition. The degree of match between the facial portrait of the user that appears and the pre-set facial message can be compared to determine whether the user that appears is the target object.

[0146] Among them, action recognition can be to determine the action of an object through the movement of the object over a period of time. For example, if it is recognized that the object moves repeatedly with a kitchen knife in its hand over a period of time, it can be determined that the action of the object is cutting vegetables. Correspondingly, it can be determined that the event corresponding to the action is cooking.

[0147] Optionally, after the action of the object is determined, the actual event corresponding to the action can be further determined through a preset video understanding model, or a preset mapping relationship between the action and the event.

[0148] The event time of the object performing the action in the target surveillance video may be the time point of the event or the duration of the event, which is not specifically limited here and one of the time representation methods may be selected according to the actual event.

[0149] In a monitoring prompt method provided in an embodiment of the present application, the identity of the object appearing in the target monitoring video can be identified to determine the target object in the target monitoring video; the action of the object appearing in the target monitoring video can be identified to determine the action of the object, and the action of the object is used to determine the type of event; and the time of the event when the object appearing in the target monitoring video performs the action is obtained. Among them, by identifying the identity of the object appearing in the target monitoring video, identifying the action, and determining the time of the event, the information of the event in the target monitoring video can be obtained more comprehensively and accurately.

[0150] It should be noted that after the event information is determined, the event information can be input into the video understanding model to obtain the edited event information.

[0151] For example, the event information can be input into the video understanding model, so that the video understanding model outputs a result, which is the information of an edited event.

[0152] It should be noted that the arrangement may be the organization of the event information, for example, the arrangement may be performed according to at least one of the input target sub-area, the target object, the event occurrence time, and the event type.

[0153] In one embodiment, the input of the video understanding model may be a plurality of scattered information of an event, and the output may be integrated information obtained based on the scattered information.

[0154] In one embodiment, the arranged events include information of the events sorted in sequence.

[0155] For example, multiple scattered information can be input into a table, as shown in Table 1:

[0156] Table 1

[0157] Event number Target sub-area Target audience Time of the incident Event Type 1 kitchen User A 12 noon Cook 2 living room Pet B 3 p.m. sleep … … … … … N balcony User B 8 p.m. walk

[0158] As shown in Table 1, the content of Table 1 is the information of the events after arrangement. The information of these events can be arranged and sorted to obtain the information of the arranged events.

[0159] It should be noted that the above-explained organized event information is only an example, and there may be other forms in the actual implementation process.

[0160] In one embodiment, the organized event information may include event information expressed in the form of sentences.

[0161] For example: User A is cooking in the kitchen at 12 noon. This sentence can be used as the information of the choreographed event.

[0162] In one embodiment, the information of the organized events may include summary information of the event information obtained by statistically analyzing information of multiple events according to preset rules.

[0163] For example, all events performed by user A may be summarized, or all events occurring in the kitchen area may be summarized, so as to obtain corresponding summary information.

[0164] In one embodiment, before sending the event information to the terminal device, the method also includes: inputting the target sub-area, and at least one of the target object, event occurrence time, and event type into a video understanding model for analysis, and outputting the organized event information; sending the event information to the terminal device includes: sending the organized event information to the terminal device.

[0165] Figure 4 This is a schematic diagram of the structure of the video understanding model provided in the embodiment of the present application. Please refer to Figure 4 , Figure 4 The input of the video understanding model may include: a target sub-region, and at least one of a target object, an event occurrence time, and an event type. The output of the video understanding model may be information about the programmed event.

[0166] In one embodiment, the video understanding model may be a neural network model, or a pre-trained convergent model, wherein the video understanding model is trained based on information about sample events and information about arranged sample events, wherein the information about sample events includes: a sample sub-area, and at least one of a sample object, a sample event occurrence time, and a sample event type.

[0167] For example: the sample sub-area is the kitchen, the sample object is user A, the sample event occurs at 11:00 noon, the sample event type is cutting vegetables, and the information of the sample event after arrangement is that user A is cooking in the kitchen at 11:00 noon; the sample sub-area is the living room, the sample object is user B, the sample event occurs at 3:00 p.m., the sample event type is squatting, and the information of the sample event after arrangement is that user B is exercising in the living room at 3:00 p.m.

[0168] It should be noted that a large amount of sample data may be pre-set for model training, and the converged model may be used as the above-mentioned video understanding model.

[0169] In a monitoring prompt method provided in an embodiment of the present application, the information of the organized event can be obtained by inputting the target sub-area, and at least one of the target object, event occurrence time, and event type into the video understanding model for analysis. Among them, the analysis performed by the video understanding model can realize the arrangement of event information, so that the information of the organized event can be obtained more quickly through the video understanding model. It can be understood that the information of the organized event includes richer and easier to understand event content, which can enable users to accurately obtain more comprehensive event information. In some possible implementations, the shooting device only sends the information of the organized event to the terminal device, but does not send the video corresponding to the event. In this implementation, the traffic consumed by the shooting device for information transmission can be greatly reduced, and it is highly applicable in some scenarios with poor network environment.

[0170] The above process is a process of arrangement using a neural network model. In the actual implementation process, other methods may also be used, such as a preset mapping relationship, etc., which are not specifically limited here.

[0171] In one embodiment, before acquiring the target surveillance video, the method also includes: sending image information of the surveillance scene corresponding to the shooting device to the terminal device; receiving a first area division instruction determined based on the image information of the surveillance scene sent by the terminal device, the first area division instruction being used to indicate the position of each sub-area in the target surveillance video in the image.

[0172] It should be noted that the image information of the monitoring scene corresponding to the shooting device can be a static image, for example, an image of the monitoring scene captured by the shooting device at any time can be sent to the terminal device.

[0173] Optionally, the terminal device can divide the area according to the image information of the monitoring scene and generate a first area division instruction. The shooting device can receive the first area division instruction sent by the terminal device and determine the position of each sub-area in the target monitoring video in the image according to the first area division instruction.

[0174] For example: the first area division instruction can indicate the location of each sub-area in the target surveillance video in the image, for example: the upper left corner is the living room, the upper right corner is the balcony, the middle is the bedroom, etc., and the first area division instruction can indicate the specific range size of each sub-area.

[0175] After receiving the first area division instruction, the shooting device can divide the target monitoring video into various sub-areas according to the content indicated by the first area division instruction, thereby determining the position of each sub-area in the target monitoring video in the image.

[0176] In the monitoring prompt method provided in the embodiment of the present application, by obtaining the first area division instruction, the shooting device can determine the position of each sub-area in the target monitoring video in the image, so as to accurately monitor the area of ​​interest to the user, avoid invalid alarm events caused by irrelevant areas, and improve the accuracy of event prompt information.

[0177] It should be noted that the above is only one way to implement area division. In the actual implementation process, the area division can also be performed by a shooting device. The specific implementation method is as follows:

[0178] In one embodiment, before acquiring the target surveillance video, the method also includes: determining multiple initial sub-areas from an image of the surveillance scene corresponding to the shooting device; sending image information of the surveillance scene carrying multiple initial sub-areas to the terminal device; receiving a second area division instruction sent by the terminal device, the second area division instruction being used to indicate the position of each sub-area in the target surveillance video in the image, each sub-area being at least two of the multiple initial sub-areas.

[0179] It should be noted that the shooting device can divide the image of the monitoring scene into regions by block segmentation, obtain multiple initial sub-regions, and send the image information of the monitoring scene carrying the multiple initial sub-regions to the terminal device.

[0180] The process of region division may be to segment the image according to the size of the equal-proportional blocks to obtain a plurality of equal-proportional blocks, and these blocks may be used as initial sub-regions.

[0181] The terminal device can determine the initial sub-areas included in each sub-area in the monitoring scene. For example, sub-area A can include 10 initial sub-areas, sub-area B can include 5 initial sub-areas, etc., without specific limitation here.

[0182] It should be noted that each initial sub-region may be, for example, a rectangular region whose length and width are preset fixed values, and each sub-region may be composed of at least two initial sub-regions.

[0183] After the terminal device determines the initial sub-area included in each sub-area, it can generate a second area division instruction and send the second area division instruction to the shooting device. The shooting device can determine the position of each sub-area in the target monitoring video in the image according to the second area division instruction.

[0184] For example, sub-region A is composed of ten initial sub-regions, namely, initial sub-region 1 to initial sub-region 10. Then, the position of sub-region A is the set of the positions of these ten initial sub-regions.

[0185] Figure 5This is a schematic diagram of sub-area division provided in the embodiment of the present application, please refer to Figure 5 , Figure 5 The sub-region is divided into four regions, namely, sub-region 1 to sub-region 4, and each sub-region is composed of multiple initial sub-regions.

[0186] in, Figure 5 Each cell in is one of the initial sub-regions, and each sub-region is composed of multiple initial sub-regions. For example, sub-region 1 is composed of the first 6 initial sub-regions in the first row and the first 3 initial sub-regions in the second row.

[0187] In one embodiment, multiple initial sub-regions are determined from an image of a surveillance scene corresponding to a shooting device, including: dividing the image of the surveillance scene corresponding to the shooting device into regions to obtain multiple initial sub-regions; or performing intelligent region recognition on the image of the surveillance scene corresponding to the shooting device to obtain multiple initial sub-regions.

[0188] Among them, region division can be region division based on preset rules, for example: determining the size of an initial sub-region based on a fixed length and width, thereby achieving division; intelligent recognition can be region division based on a preset algorithm, for example: determining the size of each initial sub-region based on the length and width of the image using a preset algorithm, thereby achieving division.

[0189] In the monitoring prompt method provided in the embodiment of the present application, the image of the monitoring scene corresponding to the shooting device can be divided into regions to obtain multiple initial sub-regions; the image information of the monitoring scene carrying multiple initial sub-regions is sent to the terminal device; and the second region division instruction determined based on the image information of the monitoring scene carrying multiple initial sub-regions sent by the terminal device is received. Among them, the shooting device can intelligently identify the monitoring scene, determine multiple initial sub-regions and then send them to the terminal device. By obtaining the second region division instruction, the shooting device can determine the position of each sub-region in the target monitoring video in the image, so that the target sub-region where the event occurred can be accurately determined.

[0190] In one embodiment, the method further includes: sending a video corresponding to the event to a terminal device.

[0191] It should be noted that the shooting device can also send a video to the terminal device, for example, it can be a video corresponding to the event.

[0192] It has been explained above that the time of occurrence of an event can be obtained, and the video within the time of occurrence of the event can be used as the video corresponding to the event, and the video can be sent to the terminal device.

[0193] It should be noted that the terminal device can enrich the content of the user prompt based on the video, for example: displaying the corresponding video while generating the text prompt.

[0194] In one possible implementation, the video only includes images of the sub-region where the event occurs.

[0195] In the monitoring prompt method provided in the embodiment of the present application, a video corresponding to the event can be sent to the terminal device. By sending the video corresponding to the event to the terminal device, the terminal device can generate more comprehensive and accurate prompt information in the process of prompting the user, thereby increasing the accuracy and completeness of the prompt information.

[0196] In a possible implementation, the video corresponding to the event sent to the terminal device only includes the image of the sub-region where the event occurred. In this embodiment, since only the image of the sub-region corresponding to the event is sent to the terminal device, the amount of data transmission can be reduced, and it is also helpful for the user to clearly and directly watch the video content of interest, effectively reducing the difficulty of the user repeatedly observing and looking for the target event in the large image.

[0197] The above implementation process is a plurality of steps executed in a process in which the photographing device is the execution subject. The following will explain the corresponding steps executed in a case in which the terminal device is the execution subject.

[0198] Figure 6 Another schematic diagram of the monitoring prompt method provided in the embodiment of the present application, please refer to Figure 6 , the method comprising:

[0199] S610: Sending a region division instruction to the photographing device.

[0200] The execution subject of the method may be the above-mentioned terminal device, and the terminal device may send a region division instruction to the shooting device.

[0201] The region division instruction is used to indicate the position of each sub-region in the image captured by the capturing device, and any two sub-regions in each sub-region correspond to different positions in the image.

[0202] It should be noted that the area division instruction can be a variety of forms of division, for example, it can be the aforementioned first area division instruction, or it can be a second area division instruction. There is no specific limitation here, and one of the methods can be selected to implement area division according to actual usage requirements.

[0203] For example, the region division instruction may indicate the position of each sub-region in the target surveillance video. For example, if there are five sub-regions in a target surveillance video, the position of each sub-region may be indicated by the region division instruction.

[0204] The position in the image may be represented by a specific coordinate range, or by the number and quantity of initial sub-regions constituting each sub-region, etc., which is not specifically limited here and can be set accordingly according to actual needs.

[0205] S620: Acquire information about events in the target surveillance video sent by the shooting device.

[0206] It should be noted that after the shooting device obtains the target surveillance video and determines the information of the events in the target surveillance video, the information of the events can be sent to the terminal device, where the information of the events can be the information of the events after the arrangement.

[0207] The target surveillance video includes at least two sub-areas determined according to the area division instruction, and the event information includes: the target sub-area where the event occurs in the at least two sub-areas, and at least one of the following information: event occurrence time, target object and event type.

[0208] S630: Generate prompt information according to the event information.

[0209] It should be noted that after receiving the event information, the terminal device may generate prompt information according to the event information, and the prompt information is used to provide prompts to the user.

[0210] In one embodiment, the prompt information may include a target sub-area where the event occurs.

[0211] It should be noted that the prompt information can be a warning information, and the warning prompt can be made in a pre-configured format, for example: it can prompt a specific behavior event generated by a user within a certain period of time, or it can also prompt a specific event occurring in a certain sub-area within a certain period of time.

[0212] For example, if prompt information is set for user A, the corresponding prompt information can be generated after obtaining that user A performs a certain behavior event; if prompt information is set for sub-area C, the corresponding prompt information can be generated after obtaining that a certain behavior event occurs in sub-area C.

[0213] In one embodiment, in addition to using the above-mentioned prompt method, a prompt can also be generated after any event occurs. For example, if user A is cooking in the kitchen at 8:00 in the morning, a prompt message for the event can be generated. In addition to prompting that an event that requires a prompt has occurred, corresponding prompt information can also be generated based on the information of the event, such as: prompting a specific time, prompting a specific object, prompting a sub-area where the event occurred, or prompting a specific event type, etc., without specific limitations here.

[0214] In the monitoring prompt method provided in the embodiment of the present application, event information based on the target sub-area in the target monitoring video can be obtained, and the information can also include at least one of the event occurrence time, target object and event type. The sub-area information can further make the event prompt content more accurate, thereby achieving more precise event monitoring.

[0215] The following is an explanation of the implementation process of generating prompt information provided in the embodiments of the present application.

[0216] In one embodiment, the event information provided by the shooting device can be input into the sentence arrangement model to obtain the corresponding prompt text, and then the corresponding prompt information can be generated based on the prompt text.

[0217] Figure 7 This is a schematic diagram of the structure of the sentence arrangement model provided in the embodiment of this application. Please refer to Figure 7 , Figure 7 The input of the sentence organization model can be event information, and the output of the sentence organization model can be a prompt text.

[0218] Among them, the sentence arrangement model is trained based on the information of sample events and sample prompt texts.

[0219] The sample prompt text can be prompted in a variety of different formats, such as phrases and complete sentences, without specific restrictions. Based on different types of sample prompt texts for training, different types of sentence arrangement models can be obtained, and one type of sentence arrangement model can be selected for use according to actual needs, for example, a sentence arrangement model that generates phrases or a sentence arrangement model that generates complete sentences.

[0220] For example, the prompt text output by the sentence arrangement model can be a phrase such as "User A, 9:00, living room, fitness" or "User B, 12:00, restaurant, dinner"; or, it can be a complete sentence such as "User A 9:00 in the living room fitness" or "User B 12:00 in the restaurant dinner".

[0221] In one embodiment, prompt information is generated based on event information, including: determining prompt text based on event information and a sentence organization model, wherein the sentence organization model is trained based on sample event information and sample prompt text; generating prompt information, the prompt information including the prompt text and a video clip of the target video in a time interval corresponding to the event.

[0222] It should be noted that, in the actual implementation process, the prompt text may be directly used as prompt information, or the prompt text may be combined with a video clip of the target video in the time interval corresponding to the event as prompt information.

[0223] The video segment of the target video in the time interval corresponding to the behavior event may be a corresponding segment captured during the event. For example, if the event is user A exercising in the living room, the corresponding segment in the target video may be extracted.

[0224] For the prompt information, a separate prompt text, such as a phrase, a complete sentence, etc., can be used to implement the prompt, or the prompt text can be combined with the above-extracted video clip, which can be a complete video clip or a partial screenshot of the video clip.

[0225] For example, the prompt information can be any one of a variety of ways such as "phrase", "complete sentence", "phrase + video clip", "complete sentence + video clip", "phrase + screenshot", "complete sentence + screenshot" and so on.

[0226] In one embodiment, for video clips and screenshots, key content may be annotated therein, for example, a target character may be annotated out of the video clip or screenshot for display.

[0227] Figure 8 For a schematic diagram of the generated prompt information provided in the embodiment of this application, please refer to Figure 8 , Figure 8 Part (a) is the prompting method for phrases, part (b) is the prompting method for complete sentences, part (c) is the prompting method for any type of sentence and video clip combination, and part (d) is the prompting method for any type of sentence and annotated screenshot combination.

[0228] In the actual implementation process, the following can be adopted Figure 8 The prompt can be performed in any of the multiple prompting methods shown, and no specific limitation is made here.

[0229] In a monitoring prompt method provided in an embodiment of the present application, a prompt text can be determined based on event information and a sentence arrangement model, and the sentence arrangement model is trained based on sample event information and sample prompt text; prompt information is generated, and the prompt information includes the prompt text and a video clip of the target video in the time interval corresponding to the behavior event. Among them, prompting by combining prompt text and video clips can increase the diversity of prompt information display, thereby meeting the prompt needs of different users. The prompt information can be clearly and clearly output according to the sentence arrangement model, so that users can obtain the content of the prompt more accurately. Even if they do not watch the video corresponding to the event, they can accurately obtain relatively accurate event content, which can improve the efficiency of users in obtaining event information.

[0230] After obtaining the prompt information in the above manner, other relevant information may be generated based on the prompt information, such as integrated information, recommended information, etc., which is not specifically limited here. The following specifically explains one of the feasible implementation methods for generating recommended information provided in the embodiments of the present application.

[0231] Fig. 9 This is a schematic diagram of one of the flow charts for generating recommendation information provided in the embodiments of this application. Please refer to Fig. 9 After generating prompt information according to the information of the target behavior event, the method further includes:

[0232] S910: Generate first integrated information of the target object according to prompt information for the same target object within a preset time period.

[0233] It should be noted that each prompt information may include information such as time, sub-area type and target object, and can be integrated based on one or more of these information. For example, various prompt information about the same target object can be obtained, and then the first integrated information of the target object can be generated.

[0234] Optionally, the first integrated information is used to characterize each event executed by the target object. It can be each specific behavior event, or the execution frequency of a certain event, or it can be the corresponding integration of the occurrence time and occurrence sub-area of ​​these events. There is no specific limitation here, and one of the integration methods can be selected according to actual needs to obtain the first integrated information after integration.

[0235] It should be noted that after the first integrated information is obtained, it can be displayed in a list format to facilitate user reference.

[0236] The integrated information can record the events of the target users within a period of time. For example, user A triggers the kitchen scene event for the longest time, which can be used to determine that user A uses the kitchen more frequently. User B triggers the living room scene event for the longest time, which can be used to determine that user B uses the living room more frequently.

[0237] S920: When the frequency of executing the target event in the first integrated information meets the preset frequency requirement, generate recommendation information for the target object and the target event.

[0238] It should be noted that after obtaining the first integrated information, recommendation information may be generated based on the frequency of a target user performing a certain target event.

[0239] For example, if the frequency of the target event is greater than a preset frequency requirement, recommendation information for the target object and the target event may be generated.

[0240] For example: if the target event is washing dishes, and the target object's duration of washing dishes is greater than the preset duration, or the target object's frequency of washing dishes within a month is greater than the preset frequency, corresponding recommendation information can be generated, such as recommending a dishwasher; if the target event is getting up at night, and the target object's number of times getting up at night in a month is greater than the preset number, corresponding recommendation information can be generated, such as recommending a night light.

[0241] In the monitoring prompt method provided in the embodiment of the present application, the first integrated information of the target object can be generated based on the prompt information for the same target object within a preset time period; when the frequency of executing the target event in the first integrated information meets the preset frequency requirement, the recommendation information for the target object and the target event is generated. Among them, the first integrated information for the target object can be generated based on the prompt information for the same target object, and the analysis of the behavior event for the target object is further provided, thereby obtaining information such as the behavior habits of the target object.

[0242] The above explanation is for recommendation information for the target object. In the actual implementation process, corresponding recommendations can also be made for the target sub-region. Another implementation method for generating recommendation information provided in an embodiment of the present application is explained below.

[0243] Fig.10 Please refer to the flowchart of another process for generating recommendation information provided in the embodiment of the present application. Fig.10 After generating prompt information according to the information of the target behavior event, the method further includes:

[0244] S1010: Generate second integrated information of the target sub-region according to the prompt information for the same sub-region within a preset time period.

[0245] Optionally, the second integrated information is used to characterize each event executed in the target sub-area. It can be each specific behavior event, or the execution frequency of a certain event, or it can be the corresponding integration of the occurrence time and occurrence sub-area of ​​these events. There is no specific limitation here. You can select one of the integration methods according to actual needs to obtain the second integrated information after integration.

[0246] It should be noted that after the second integrated information is obtained, it can be displayed in a list format to facilitate user reference.

[0247] The integrated information can record the behavioral events of the target sub-area within a period of time. For example, the kitchen is used every morning, noon, and evening, so it can be judged that the kitchen is used more frequently. User A goes to the balcony to exercise every Friday, so it can be judged that the balcony is often used on Fridays.

[0248] S1020: When the frequency of executing the target event in the second integrated information meets a preset frequency requirement, generate recommendation information for the target sub-region and the target event.

[0249] It should be noted that after obtaining the second integrated information, recommendation information may be generated based on the frequency of a certain target event in the target sub-region.

[0250] For example, if the frequency of the target event is greater than the preset frequency requirement, recommendation information for the target sub-region and the target event may be generated.

[0251] For example: if the target event is cooking, and the frequency of cooking in the kitchen every day is greater than or equal to the preset frequency, corresponding recommendation information can be generated, such as recommending kitchen utensils; if the target event is bathing, and the number of bathing events in the bathroom within a month is greater than the preset number, corresponding recommendation information can be generated, such as recommending bath products.

[0252] In the monitoring prompt method provided in the embodiment of the present application, the second integrated information of the target sub-area is generated according to the prompt information for the same sub-area within the preset time period; when the frequency of executing the target event in the second integrated information meets the preset frequency requirement, the recommendation information for the target sub-area and the target event is generated. Among them, the second integrated information for the target object can be generated according to the prompt information for the same sub-area, further providing the user with statistical information with reference value. For example, in the store monitoring scene, different shelf areas correspond to different sub-areas, and the popularity of different shelf areas can be obtained through the second integrated information.

[0253] It should be noted that the implementation process of the above steps S910-S920 is only one feasible method, and in actual application, it can be ended after obtaining the first integrated information, that is, after executing step S910, and no specific limitation is made here.

[0254] Accordingly, the implementation process of the above steps S1010-S1020 is only one feasible method, and in practical applications it can be ended after obtaining the second integrated information, that is, after executing step S1010, and no specific limitation is made here.

[0255] Accordingly, step S920 and step S1020 may also be performed as independent steps:

[0256] In one embodiment, the method also includes: generating recommendation information for the target object and the target event when the frequency of executing the target event in the first integrated information meets the preset frequency requirement; and / or generating recommendation information for the target sub-area and the target event when the frequency of executing the target event in the second integrated information meets the preset frequency requirement.

[0257] In this embodiment, recommendation information for a certain target object or a certain target sub-region can be implemented based on the first integrated information for the same target object or the second integrated information for the same sub-region, thereby improving the accuracy and rationality of the recommendation.

[0258] It should be noted that the above are only two feasible examples for generating recommendation information. In actual implementation, the above two methods can be combined or based on the above two methods, recommendations can be implemented according to other relevant information.

[0259] For integrated information, events that occurred in the living room within a month can be integrated to determine the frequency of use of the living room, the frequent use time and the frequent users, etc. For example, it can be judged that user B uses the living room more frequently. For example: the user who triggered the living room scene event for the longest time is user B, so it can be judged that the person who uses the living room more frequently is user B; the user who triggered the public health scene for the longest time is the pet at home, so it can be judged that the pet at home uses the public health scene more frequently.

[0260] For recommended information, personalized recommendations can be made for users, and corresponding reports can be generated based on user needs. For example, more than 90% of the time the living room is used is at 7:30 p.m., and the most triggered event is watching the news. Users can be recommended to use a certain news software. Or, based on the fact that the frequency of pets using public health this month is 50% higher than last month, users can be recommended to pay attention to their pets' health.

[0261] In one embodiment, before sending a region division instruction to the shooting device, the method also includes: acquiring an image captured by the shooting device; segmenting the image to obtain a plurality of blocks; and determining a sub-region based on a user's selection from the plurality of blocks, wherein the sub-region is a collection of target blocks selected by the user from the plurality of blocks.

[0262] It should be noted that the image information of the monitoring scene may be an image captured within the monitoring scene before the shooting device acquires the target monitoring video, and the scene displayed in the image may be the same as the scene in the target monitoring video.

[0263] Optionally, after the terminal device obtains the image information of the monitored scene, it can perform image recognition processing, for example: determining the position of each sub-area in the image based on the content in the scene. The recognition processing process can be implemented by a preset recognition algorithm or by artificial division, and no specific restrictions are made here.

[0264] For example: different regions may be automatically divided based on image recognition technology to obtain the position of each sub-region in the image; or, the image may be divided into regions in response to an input division operation to obtain division results of multiple sub-regions, thereby determining the position of each sub-region in the image.

[0265] It should be noted that the above method can realize the division of sub-areas, and can generate the aforementioned first area division instruction, and then send the first area division instruction to the shooting device.

[0266] In the monitoring prompt method provided in the embodiment of the present application, the image information of the monitoring scene sent by the shooting device can be received; the image information of the monitoring scene can be segmented and processed to determine the position of each sub-area in the monitoring scene in the image. Among them, the terminal device can segment the image information in the monitoring scene to obtain multiple blocks, and then determine the position of each sub-area in the target monitoring video in the image based on the user's selection, thereby improving the efficiency and accuracy of determining the sub-area position.

[0267] In a possible implementation, the terminal device divides the monitoring image into a plurality of equal-proportion blocks, and determines a set of the plurality of equal-proportion blocks as a sub-region according to user selection. Optionally, the user determines a set of blocks included in the plurality of sub-regions respectively.

[0268] The user may determine the block included in each sub-region by inputting a selection, and the block is the aforementioned initial sub-region.

[0269] In this embodiment, a convenient implementation method for determining a sub-region is provided, which can reduce the difficulty of setting a sub-region for a user.

[0270] In one embodiment, before sending a region division instruction to the shooting device, the method also includes: receiving image information of a monitoring scene carrying multiple initial sub-regions sent by the shooting device, the initial sub-regions being obtained by region division of the image shot by the shooting device; and determining, in response to a selection operation for the initial sub-regions, a position of each sub-region in the target monitoring video in the image, each sub-region in the target monitoring video being at least two of the multiple initial sub-regions.

[0271] It should be noted that the shooting device can divide the image of the monitoring scene into regions by block segmentation, obtain multiple initial sub-regions, and send the image information of the monitoring scene carrying the multiple initial sub-regions to the terminal device.

[0272] The process of region division may be to segment the image according to the size of the equal-proportional blocks to obtain a plurality of equal-proportional blocks, and these blocks may be used as initial sub-regions.

[0273] The terminal device can determine the initial sub-areas included in each sub-area in the monitoring scene. For example, sub-area A can include 10 initial sub-areas, sub-area B can include 5 initial sub-areas, etc., without specific limitation here.

[0274] It should be noted that each initial sub-region may be, for example, a rectangular region whose length and width are preset fixed values, and each sub-region may be composed of at least two initial sub-regions.

[0275] After the terminal device determines the initial sub-area included in each sub-area, it can generate a second area division instruction and send the second area division instruction to the shooting device. The shooting device can determine the position of each sub-area in the target monitoring video in the image according to the second area division instruction.

[0276] For example, sub-region A is composed of ten initial sub-regions, namely, initial sub-region 1 to initial sub-region 10. Then, the position of sub-region A is the set of the positions of these ten initial sub-regions.

[0277] It should be noted that the relationship between the initial sub-region and the sub-region can be seen in the above Figure 5 , which will not be repeated here.

[0278] The terminal device can determine the initial sub-areas included in each sub-area according to the user's selection operation, generate the aforementioned second area division instruction, and then send the second area division instruction to the shooting device.

[0279] In the monitoring prompt method provided in the embodiment of the present application, image information of a monitoring scene carrying multiple initial sub-areas sent by a shooting device can be received; in response to a selection operation for the initial sub-areas, the position of each sub-area in the target monitoring video in the image is determined. The user can select each sub-area in the target monitoring video from the multiple initial sub-areas, thereby improving the efficiency of determining the sub-area position.

[0280] The following is an explanation of the interaction process between the shooting device and the terminal device provided in the embodiment of the present application.

[0281] Fig.11For a schematic diagram of the interaction between the photographing device and the terminal device provided in the embodiment of the present application, please refer to Fig.11 , the method comprising:

[0282] S1110: The photographing device receives the area division instruction sent by the terminal device.

[0283] S1120: The shooting device obtains the target surveillance video.

[0284] S1130: The shooting device obtains information about events in the target surveillance video.

[0285] S1140: The shooting device sends information about events in the target surveillance video to the terminal device.

[0286] S1150: The terminal device generates prompt information according to the event information.

[0287] It should be noted that steps S1110-S1150 are the overall implementation process of the aforementioned steps S310-S330 and steps S610-S630, which is implemented through interaction between the shooting device and the terminal device.

[0288] The above steps have been explained in the above content and will not be repeated here. The implementation process of steps S1110-S1150 is as follows: Figure 2 In practical applications, in addition to the process in the middle and upper side, Figure 2 In addition to implementing the process in the middle and upper side, it can also be implemented through Figure 2 This is achieved through the process in the middle and lower sides.

[0289] Fig.12 This is another interactive diagram of the shooting device and the terminal device provided in the embodiment of the present application. Please refer to Fig.12 , the method comprising:

[0290] S1210: The shooting device obtains the target surveillance video.

[0291] S1220: The shooting device sends the target surveillance video to the terminal device.

[0292] S1230: The terminal device obtains information about events in the target surveillance video.

[0293] S1240: The terminal device generates prompt information according to the event information.

[0294] It should be noted that when using Figure 2 During the implementation of the middle and lower side processes, the shooting device only needs to acquire the target monitoring video, and the image analysis and prompts to the user can be implemented by the terminal device.

[0295] On the other hand, an embodiment of the present application provides a monitoring prompt method, which is applied to a shooting device, and the shooting device is communicatively connected to a terminal device. The method includes: obtaining a target monitoring video, the target monitoring video includes at least two sub-areas, and the at least two sub-areas correspond to different image positions in the monitoring video; sending the target monitoring video to the terminal device so that the terminal device determines information about an event in the target monitoring video, and provides an event prompt to the user based on the event information, wherein the event information includes: a target sub-area in which an event occurs in at least two sub-areas, and at least one of the following information: event occurrence time, target object, and event type.

[0296] On the other hand, an embodiment of the present application provides a monitoring prompt method, which is applied to a terminal device, and the terminal device is communicatively connected to a shooting device. The method includes: obtaining a target monitoring video sent by the shooting device, the target monitoring video includes at least two sub-areas, and the at least two sub-areas correspond to different image positions in the monitoring video; obtaining information about an event in the target monitoring video, the event information includes: a target sub-area where the event occurs in at least two sub-areas, and at least one of the following information: event occurrence time, target object, and event type; generating prompt information based on the event information, the prompt information includes the target sub-area where the event occurs.

[0297] It should be noted that if the Fig.12 The interactive process shown in , the terminal device does not need to send a region division instruction to the shooting device, and the terminal device can implement the region division by itself. The shooting device obtains the target monitoring video and sends it to the terminal device, and the rest of the work can be implemented by the terminal device.

[0298] It should be understood that, although the steps in the above-mentioned flowcharts are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a part of the steps in the above-mentioned flowcharts may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these sub-steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the sub-steps or stages of other steps.

[0299] Based on the foregoing embodiments, the embodiments of the present application provide a monitoring prompt device, which includes the modules included and the units included in the modules, which can be implemented by a processor; of course, it can also be implemented by a specific logic circuit; in the implementation process, the processor can be a central processing unit (CPU), a microprocessor (MPU), a digital signal processor (DSP) or a field programmable gate array (FPGA), etc.

[0300] Fig.13 This is a schematic diagram of the structure of the monitoring prompt device provided in the embodiment of the present application. Please refer to Fig.13 , the device comprises:

[0301] The video acquisition module 1310 is used to acquire a target surveillance video, where the target surveillance video includes at least two sub-areas, and the at least two sub-areas correspond to different image positions in the surveillance video respectively;

[0302] The information determination module 1320 is used to obtain information about an event in a target surveillance video, where the information about the event includes: a target sub-region where the event occurs in at least two sub-regions, and at least one of the following information: event occurrence time, target object, and event type;

[0303] The information transmission module 1330 is used to send the event information to the terminal device so that the terminal device provides the user with an event prompt according to the event information.

[0304] In one embodiment, the information determination module 1320 is specifically used to perform target detection on the target surveillance video and determine the position of the target detection frame in the image, where the target detection frame is used to indicate the object that appears in the image; if the position of the target detection frame in the image overlaps with the position of the target sub-area in the image, then information on events in the target sub-area in the target surveillance video is obtained.

[0305] In one embodiment, the information determination module 1320 is specifically used to identify the identity of the object and determine the target object in the target surveillance video; identify the action of the object and determine the event type according to the identified action; and obtain the event occurrence time when the object performs the action.

[0306] In one embodiment, the information transmission module 1330 is also used to input the target sub-region, and at least one of the target object, event occurrence time, and event type into a video understanding model for analysis, and output the information of the organized event. The video understanding model is trained based on the information of the sample event and the information of the organized sample event, wherein the information of the sample event includes: the sample sub-region, and at least one of the sample object, sample event occurrence time, and sample event type; and the organized event information is sent to the terminal device.

[0307] In one embodiment, the video acquisition module 1310 is also used to send image information of the monitoring scene corresponding to the shooting device to the terminal device; receive a first area division instruction determined based on the image information of the monitoring scene sent by the terminal device, and the first area division instruction is used to indicate the position of each sub-area in the target monitoring video in the image.

[0308] In one embodiment, the video acquisition module 1310 is also used to determine multiple initial sub-areas from an image of a monitoring scene corresponding to a shooting device; send image information of the monitoring scene carrying multiple initial sub-areas to a terminal device; receive a second area division instruction sent by the terminal device, the second area division instruction is used to indicate the position of each sub-area in the target monitoring video in the image, each sub-area being at least two of the multiple initial sub-areas.

[0309] In one embodiment, the information transmission module 1330 is further used to send a video corresponding to the event to the terminal device.

[0310] Fig.14 This is another structural diagram of the monitoring prompt device provided in the embodiment of the present application, please refer to Fig.14 , the device comprises:

[0311] The instruction issuing module 1410 is used to send a region division instruction to the shooting device, where the region division instruction is used to indicate the position of each sub-region in the target monitoring video in the image shot by the shooting device, and any two sub-regions in each sub-region correspond to different positions in the image;

[0312] The information acquisition module 1420 is used to acquire information about an event in a target surveillance video sent by a shooting device, the target surveillance video including at least two sub-areas determined according to the area division instruction, and the event information including: a target sub-area in which the event occurs in the at least two sub-areas, and at least one of the following information: event occurrence time, target object, and event type;

[0313] The prompt generating module 1430 is used to generate prompt information according to the event information, and the prompt information includes the target sub-region where the event occurs.

[0314] In one embodiment, the prompt generation module 1430 is specifically used to determine the prompt text based on the event information and the sentence organization model, where the sentence organization model is trained based on the sample event information and the sample prompt text; and generate prompt information, which includes the prompt text and a video clip of the target video in the time interval corresponding to the event.

[0315] In one embodiment, the prompt generating module 1430 is further used to generate first integrated information of the target object according to the prompt information for the same target object within a preset time period, where the first integrated information is used to characterize various events executed by the target object.

[0316] In one embodiment, the prompt generating module 1430 is further used to generate second integrated information of the target sub-region according to the prompt information for the same sub-region within a preset time period, and the second integrated information is used to characterize each event executed in the target sub-region.

[0317] In one embodiment, the prompt generation module 1430 is further configured to generate recommendation information for the target object and the target event when the frequency of executing the target event in the first integrated information meets the preset frequency requirement. And / or, generate recommendation information for the target sub-region and the target event when the frequency of executing the target event in the second integrated information meets the preset frequency requirement.

[0318] In one embodiment, the instruction issuing module 1410 is also used to obtain an image captured by a shooting device; segment the image to obtain multiple blocks; and determine a sub-region based on a user's selection from the multiple blocks, where the sub-region is a collection of target blocks selected by the user from the multiple blocks.

[0319] In one embodiment, the instruction issuing module 1410 is also used to receive an image carrying multiple initial sub-areas sent by a shooting device, where the initial sub-areas are obtained by dividing the image shot by the shooting device; in response to a selection operation on the initial sub-areas, the position of each sub-area in the target monitoring video in the image is determined, and each sub-area in the target monitoring video is at least two of the multiple initial sub-areas.

[0320] The description of the above device embodiment is similar to the description of the above method embodiment, and has similar beneficial effects as the method embodiment. For technical details not disclosed in the device embodiment of the present application, please refer to the description of the method embodiment of the present application for understanding.

[0321] It should be noted that in the embodiments of this application Fig.13 as well as Fig.14 The division of modules in the monitoring prompt device shown is schematic and is only a logical function division. There may be other division methods in actual implementation. In addition, each functional unit in each embodiment of the present application may be integrated into a processing unit, or may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of a software functional unit. It may also be implemented in the form of a combination of software and hardware.

[0322] It should be noted that in the embodiment of the present application, if the above method is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the embodiment of the present application can be essentially or partly embodied in the form of a software product that contributes to the relevant technology. The computer software product is stored in a storage medium, including several instructions to enable an electronic device to execute all or part of the methods of each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a U disk, a mobile hard disk, a read-only memory (ROM), a disk or an optical disk. In this way, the embodiment of the present application is not limited to any specific combination of hardware and software.

[0323] Fig.15 For a schematic diagram of the structure of the computer device provided in the embodiment of the present application, please refer to Fig.15 , an embodiment of the present application provides a computer device, which can be any electronic device in the above-mentioned shooting device or terminal device, and its internal structure diagram can be as shown in Fig.15 As shown. The computer device includes a processor 1520, a memory and a network interface 1540 connected via a system bus 1510. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium 1531 and an internal memory 1532. The non-volatile storage medium 1531 stores an operating system, a computer program and a database. The internal memory 1532 provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium 1531. The database of the computer device is used to store data. The network interface 1540 of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor 1520, the above method is implemented.

[0324] An embodiment of the present application provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps in the method provided in the above embodiment are implemented.

[0325] An embodiment of the present application provides a computer program product including instructions, which, when executed on a computer, enables the computer to execute the steps of the method provided in the above method embodiment.

[0326] Those skilled in the art will understand that Fig.15 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0327] In one embodiment, the monitoring prompt device provided by the present application can be implemented in the form of a computer program. The computer program can be Fig.15 The computer device shown in the figure is run. The memory of the computer device can store various program modules constituting the above-mentioned device. The computer program composed of various program modules enables the processor to execute the steps in the method of each embodiment of the present application described in this specification.

[0328] It should be noted here that the description of the above storage medium and device embodiments is similar to the description of the above method embodiments, and has similar beneficial effects as the method embodiments. For technical details not disclosed in the storage medium, storage medium and device embodiments of this application, please refer to the description of the method embodiments of this application for understanding.

[0329] It should be understood that "one embodiment" or "an embodiment" or "some embodiments" mentioned throughout the specification means that specific features, structures or characteristics related to the embodiment are included in at least one embodiment of the present application. Therefore, "in one embodiment" or "in one embodiment" or "in some embodiments" appearing throughout the specification may not necessarily refer to the same embodiment. In addition, these specific features, structures or characteristics can be combined in one or more embodiments in any suitable manner. It should be understood that in various embodiments of the present application, the size of the sequence number of the above-mentioned processes does not mean the order of execution, and the execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present application. The above-mentioned sequence numbers of the embodiments of the present application are only for description and do not represent the advantages and disadvantages of the embodiments. The above description of each embodiment tends to emphasize the differences between the various embodiments, and the same or similar aspects can be referenced to each other. For the sake of brevity, this article will not repeat them.

[0330] The term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there may be three relationships. For example, object A and / or object B can represent three situations: object A exists alone, object A and object B exist at the same time, and object B exists alone.

[0331] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.

[0332] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The embodiments described above are only schematic. For example, the division of the modules is only a logical function division. There may be other division methods in actual implementation, such as: multiple modules or components can be combined, or can be integrated into another system, or some features can be ignored, or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of devices or modules can be electrical, mechanical or other forms.

[0333] The modules described above as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical modules; they may be located in one place or distributed on multiple network units; some or all of the modules may be selected according to actual needs to achieve the purpose of the present embodiment.

[0334] In addition, all functional modules in the embodiments of the present application may be integrated into one processing unit, or each module may be a separate unit, or two or more modules may be integrated into one unit; the above-mentioned integrated modules may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0335] A person skilled in the art can understand that all or part of the steps of implementing the above method embodiment can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above method embodiment; and the aforementioned storage medium includes: mobile storage devices, read-only memories (ROM), magnetic disks or optical disks, etc., various media that can store program codes.

[0336] Alternatively, if the above-mentioned integrated unit of the present application is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present application can essentially or in other words, the part that contributes to the relevant technology can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for enabling an electronic device to execute all or part of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROMs, magnetic disks, or optical disks.

[0337] The methods disclosed in several method embodiments provided in this application can be arbitrarily combined without conflict to obtain new method embodiments.

[0338] The features disclosed in several product embodiments provided in this application can be arbitrarily combined without conflict to obtain new product embodiments.

[0339] The features disclosed in several method or device embodiments provided in this application can be arbitrarily combined without conflict to obtain new method embodiments or device embodiments.

[0340] The above is only an implementation method of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A monitoring prompt method, characterized in that: Applied to a photographing device, the photographing device is communicatively connected with a terminal device, and the method includes: Acquire a target surveillance video, wherein the target surveillance video includes at least two sub-areas, and the at least two sub-areas respectively correspond to different image positions in the surveillance video; Acquire information about an event in the target surveillance video, the information about the event including: a target sub-region where the event occurs in the at least two sub-regions, and at least one of the following information: event occurrence time, target object, and event type; The event information is sent to the terminal device, so that the terminal device provides an event prompt to the user according to the event information.

2. The method according to claim 1, characterized in that The obtaining of information about an event in the target surveillance video includes: Performing target detection on the target surveillance video to determine the position of a target detection frame in the image, wherein the target detection frame is used to indicate an object that appears in the image; If the position of the target detection frame in the image overlaps with the position of the target sub-region in the image, information about events in the target sub-region in the target monitoring video is obtained.

3. The method according to claim 2, characterized in that The obtaining of information about events in the target sub-area in the target surveillance video includes at least one of the following: Performing identity recognition on the object to determine the target object in the target surveillance video; Performing action recognition on the object, and determining the event type according to the recognized action; Get the time when the event occurs when the object performs an action.

4. The method according to claim 1, characterized in that Before sending the event information to the terminal device, the method further includes: Inputting the target sub-region, and at least one of the target object, the event occurrence time, and the event type into a video understanding model for analysis, and outputting the organized information of the event, wherein the video understanding model is trained based on the information of sample events and the organized information of sample events, wherein the information of the sample events includes: a sample sub-region, and at least one of the sample object, the sample event occurrence time, and the sample event type; The sending the event information to the terminal device includes: The information of the programmed event is sent to the terminal device.

5. The method according to any one of claims 1 to 4, characterized in that: Before acquiring the target surveillance video, the method further includes: Sending image information of the monitoring scene corresponding to the shooting device to the terminal device; A first area division instruction determined based on the image information of the monitoring scene and sent by the terminal device is received, where the first area division instruction is used to indicate the position of each of the sub-areas in the target monitoring video in the image.

6. The method according to any one of claims 1 to 4, characterized in that: Before acquiring the target surveillance video, the method further includes: Determining a plurality of initial sub-areas from an image of a monitoring scene corresponding to the photographing device; Sending image information of the monitoring scene carrying the multiple initial sub-areas to a terminal device; A second area division instruction sent by the terminal device is received, where the second area division instruction is used to indicate the position of each of the sub-areas in the target monitoring video in the image, and each of the sub-areas is at least two of the multiple initial sub-areas.

7. The method according to any one of claims 1 to 4, characterized in that: The method further comprises: The video corresponding to the event is sent to the terminal device.

8. A monitoring prompt method, characterized in that: Applied to a terminal device, the terminal device is communicatively connected with a photographing device, and the method includes: Sending a region division instruction to the photographing device, wherein the region division instruction is used to indicate positions of each sub-region in the image photographed by the photographing device, wherein any two sub-regions in each sub-region correspond to different positions in the image; Acquire information about an event in a target surveillance video sent by the shooting device, the target surveillance video including at least two sub-areas determined according to the area division instruction, the event information including: a target sub-area in which the event occurs among the at least two sub-areas, and at least one of the following information: event occurrence time, target object, and event type; Prompt information is generated according to the information of the event, and the prompt information includes a target sub-region where the event occurs.

9. The method according to claim 8, characterized in that The generating prompt information according to the information of the event includes: Determine a prompt text according to the event information and a sentence arrangement model, wherein the sentence arrangement model is trained according to the sample event information and the sample prompt text; The prompt information is generated, where the prompt information includes the prompt text and a video clip of the target video in a time interval corresponding to the event.

10. The method according to claim 8, characterized in that After generating the prompt information according to the information of the event, the method further includes: According to the prompt information for the same target object within a preset time period, first integrated information of the target object is generated, where the first integrated information is used to characterize various events executed by the target object.

11. The method according to claim 8, characterized in that After generating the prompt information according to the information of the event, the method further includes: According to the prompt information for the same sub-region within a preset time period, second integrated information of the target sub-region is generated, where the second integrated information is used to represent each event executed in the target sub-region.

12. The method according to claim 10 or 11, characterized in that: The method further comprises: When the frequency of executing the target event in the first integrated information meets the preset frequency requirement, generating recommendation information for the target object and the target event; and / or, When the frequency of executing the target event in the second integrated information meets the preset frequency requirement, recommendation information for the target sub-area and the target event is generated.

13. The method according to any one of claims 8 to 11, characterized in that Before sending the region division instruction to the shooting device, the method further includes: Acquire an image captured by the shooting device; Segmenting the image to obtain a plurality of blocks; The sub-region is determined according to a selection of the user from the plurality of blocks, and the sub-region is a set of target blocks selected by the user from the plurality of blocks.

14. The method according to claim 8 or 9, characterized in that: Before sending the region division instruction to the shooting device, the method further includes: receiving an image sent by the photographing device and carrying a plurality of initial sub-regions, wherein the initial sub-regions are obtained by dividing the image photographed by the photographing device; In response to a selection operation on the initial sub-region, a position of each of the sub-regions in the target surveillance video in the image is determined, and each of the sub-regions in the target surveillance video is at least two of the multiple initial sub-regions.

15. A monitoring prompt method, characterized in that: Applied to a photographing device, the photographing device is communicatively connected with a terminal device, and the method includes: Acquire a target surveillance video, wherein the target surveillance video includes at least two sub-areas, and the at least two sub-areas respectively correspond to different image positions in the surveillance video; The target surveillance video is sent to the terminal device so that the terminal device determines the information of the event in the target surveillance video and provides an event prompt to the user based on the information of the event, wherein the information of the event includes: the target sub-area where the event occurs in the at least two sub-areas, and at least one of the following information: the time of event occurrence, the target object and the event type.

16. A monitoring prompt method, characterized in that: Applied to a terminal device, the terminal device is communicatively connected with a photographing device, and the method includes: Acquire a target surveillance video sent by the shooting device, wherein the target surveillance video includes at least two sub-areas, and the at least two sub-areas respectively correspond to different image positions in the surveillance video; Acquire information about an event in the target surveillance video, the information about the event including: a target sub-region where the event occurs in the at least two sub-regions, and at least one of the following information: event occurrence time, target object, and event type; Prompt information is generated according to the information of the event, and the prompt information includes a target sub-region where the event occurs.

17. A computer device comprising a memory and a processor, wherein the memory stores a computer program executable on the processor, wherein: When the processor executes the program, the steps of the method according to any one of claims 1-7, 8-14, 15 or 16 are implemented.

18. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1-7, 8-14, 15 or 16 is implemented.

Images