Video monitoring streaming media decision-making method and device, equipment and storage medium

Abstract

The invention provides a video monitoring streaming media decision-making method, device and equipment and a storage medium, and relates to the technical field of video monitoring, and the method comprises the steps: obtaining a streaming starting decision-making factor corresponding to each video monitoring device in at least one video monitoring device; based on each stream starting decision factor, determining a stream starting probability of each video monitoring device; and based on each stream starting probability, determining a stream media decision result corresponding to the at least one video monitoring device, the stream media decision result being used for representing whether each video monitoring device starts stream in advance, and when the stream media decision result represents that each video monitoring device starts stream in advance, sending the stream media decision result to the at least one video monitoring device. The streaming media decision result carries a corresponding pre-start streaming prediction sequence. According to the invention, the stream starting waiting time can be shortened, and the user experience is improved.

Description

Technical Field

[0001] The present invention relates to the field of video surveillance technology, and in particular to a video surveillance streaming media decision method, device, equipment and storage medium. Background Art

[0002] With the rapid development of Internet of Things technology, video surveillance equipment has greatly shortened the distance between people. It is not restricted by geography. When the video surveillance equipment is installed, people and things of interest can be viewed anytime and anywhere. The specific process is as follows: after the user opens the APP (application) and logs in to the account, the APP obtains the device list, and after the APP logs in to the device through a hole, the user selects a device, and the APP requests the stream from the device, and then the device sends the stream to the APP. After receiving the video stream, the APP decodes and plays it. However, in the above process, the success rate of the APP logging in to the device through a hole is low, and the speed of starting the stream is slow after the hole is successfully punched, resulting in a long waiting time for starting the stream.

[0003] In the prior art, such as Figure 1 As shown, after the user opens the APP, logs in to the account, obtains the device list and selects a certain device, the APP requests the streaming media server to stream, thus avoiding the problem of the APP failing to log in to the device through a hole. The specific process is: the signaling server starts streaming to the selected device, the device sends the stream to the streaming media server, the streaming media server forwards the video stream to the APP, and the APP decodes and plays the video stream after receiving it. However, the above solution only solves the problem of the APP failing to log in to the device through a hole, and does not solve the problem of the long waiting time for starting the stream. Therefore, how to shorten the waiting time for starting the stream for video surveillance equipment is an urgent problem to be solved. Summary of the invention

[0004] The present invention provides a video surveillance streaming media decision method, device, equipment and storage medium, which are used to solve the defect of long waiting time for starting streaming of video surveillance equipment in the prior art, shorten the waiting time for starting streaming and improve user experience.

[0005] The present invention provides a video surveillance streaming media decision method, comprising:

[0006] Obtaining a flow start decision factor corresponding to each of the video surveillance devices in at least one video surveillance device;

[0007] Based on each of the start-flow decision factors, determining the start-flow probability of each of the video surveillance devices;

[0008] Based on each of the streaming start probabilities, a streaming media decision result corresponding to the at least one video surveillance device is determined, and the streaming media decision result is used to characterize whether each of the video surveillance devices has pre-started streaming. In the case where the streaming media decision result characterizes whether each of the video surveillance devices has pre-started streaming, the streaming media decision result carries a corresponding pre-start prediction order.

[0009] According to the video surveillance streaming media decision method provided by the present invention, the starting streaming probability of each video surveillance device is determined based on each starting streaming decision factor, including:

[0010] When the first preset condition is met, determining that the probability of starting the flow of the video surveillance device is 1;

[0011] In the case where the first preset condition is not met, determining, for each video surveillance device, the occurrence probability corresponding to the start-flow decision factor; and determining the start-flow probability of the video surveillance device based on the start-flow decision factor and the corresponding occurrence probability;

[0012] The first preset condition includes:

[0013] A preset alarm in the flow start decision factor occurs; or,

[0014] The number of devices corresponding to the at least one video surveillance device is one.

[0015] According to the video surveillance streaming media decision method provided by the present invention, the determining of the occurrence probability corresponding to the streaming decision factor includes:

[0016] Get historical start flow data;

[0017] Based on the historical flow-starting data, determining a preset probability of occurrence corresponding to the flow-starting decision factor;

[0018] When the video surveillance device does not meet the second preset condition, determining the occurrence probability corresponding to the start-flow decision factor to be 0;

[0019] When the start-up of the video surveillance device satisfies the second preset condition, the preset occurrence probability is determined as the occurrence probability corresponding to the start-up decision factor.

[0020] According to the video surveillance streaming media decision method provided by the present invention, the start-streaming decision factor includes an alarm factor and a user habit factor; the preset occurrence probability includes a first preset occurrence probability corresponding to the user habit factor and a second preset occurrence probability corresponding to the alarm factor; the occurrence probability includes a habit occurrence probability corresponding to the user habit factor and an alarm occurrence probability corresponding to the alarm factor;

[0021] When the start of the video surveillance device satisfies the second preset condition, determining the second preset occurrence probability as the occurrence probability corresponding to the start of the flow decision factor includes:

[0022] When the start of streaming of the video surveillance device satisfies the second preset condition that the start of streaming of the video surveillance device is in accordance with the location and / or time of user habits, the first preset occurrence probability is determined as the habitual occurrence probability corresponding to the user habit factor;

[0023] and / or,

[0024] In the case where the start of the video surveillance device satisfies the second preset condition that the video surveillance device generates alarm information, the second preset occurrence probability is determined as the alarm occurrence probability corresponding to the alarm factor.

[0025] According to the video surveillance streaming media decision method provided by the present invention, the streaming decision factor also includes the number of devices corresponding to the at least one video surveillance device;

[0026] The determining the probability of starting the flow of the video surveillance device based on the decision factor for starting the flow and the corresponding probability of occurrence includes:

[0027] Upon receiving the alarm information corresponding to the video surveillance device, determining the target alarm type based on the alarm information;

[0028] Determining an alarm weight corresponding to the alarm factor based on the target alarm type and the historical flow start data;

[0029] Determining a habit weight corresponding to the user habit factor based on the historical initiation data;

[0030] Determine a first flow start probability based on a preset constant, the alarm occurrence probability and the alarm weight corresponding to the alarm factor, and the habit occurrence probability and the habit weight corresponding to the user habit factor;

[0031] Based on the number of devices, determine the device weight corresponding to the video surveillance device;

[0032] Based on the first streaming start probability and the device weight, a streaming start probability of the video surveillance device is determined.

[0033] According to the video surveillance streaming media decision method provided by the present invention, the determining of the streaming media decision result corresponding to the at least one video surveillance device based on each of the streaming start probabilities includes:

[0034] Compare each of the flow start probabilities with a preset flow start threshold value to determine a flow start decision result;

[0035] Sorting the start-stream probabilities in descending order, and determining a start-stream priority sorting result corresponding to the at least one video surveillance device;

[0036] Based on the stream start priority sorting result and the stream start decision result, a streaming media decision result corresponding to the at least one video surveillance device is determined.

[0037] According to the video surveillance streaming media decision method provided by the present invention, the method further includes:

[0038] Get abnormal start flow data;

[0039] In the case where the abnormal start-up data is different from the start-up priority ranking result, based on the abnormal start-up data, determining the abnormal factor corresponding to the video surveillance device from the start-up decision factors; updating the abnormality occurrence probability corresponding to the abnormal factor, and updating the start-up probability of the video surveillance device based on the abnormality occurrence probability;

[0040] When the abnormal flow start data is different from the flow start decision result, the flow start preset threshold is updated based on the abnormal flow start data.

[0041] The present invention also provides a video surveillance streaming media decision device, comprising:

[0042] An acquisition module, used to acquire a flow start decision factor corresponding to each of the at least one video surveillance device;

[0043] A first determination module is used to determine the probability of starting the flow of each of the video surveillance devices based on each of the start-flow decision factors;

[0044] The second determination module is used to determine the streaming media decision result corresponding to the at least one video surveillance device based on each of the streaming start probabilities, and the streaming media decision result is used to characterize whether each of the video surveillance devices pre-starts the streaming, and when the streaming media decision result characterizes that each of the video surveillance devices pre-starts the streaming, the streaming media decision result carries the corresponding pre-start prediction order.

[0045] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, any of the above-mentioned video surveillance streaming media decision methods is implemented.

[0046] The present invention also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the video surveillance streaming media decision method as described in any one of the above is implemented.

[0047] The video surveillance streaming media decision method, device, equipment and storage medium provided by the present invention determine the streaming probability of each video surveillance device according to the streaming decision factor corresponding to each video surveillance device, and determine the streaming media decision results corresponding to all video surveillance devices according to each streaming probability, that is, determine whether each video surveillance device is pre-streamed. In the case that each video surveillance device is pre-streamed, the pre-streaming prediction order of each video surveillance device is further determined, and the video surveillance device is pre-streamed according to the pre-streaming prediction order, so that after the user triggers it, the live broadcast of the corresponding video surveillance device can be directly started, which greatly shortens the waiting time for streaming and improves the user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0049] Figure 1 It is a schematic diagram of the start-up flow of a video surveillance device provided by the prior art;

[0050] Figure 2 This is one of the flow charts of the video surveillance streaming media decision method provided by an embodiment of the present invention;

[0051] Figure 3 This is a second flow chart of the video surveillance streaming media decision method provided by an embodiment of the present invention;

[0052] Figure 4 It is a schematic diagram of the flow of starting a video surveillance device provided by an embodiment of the present invention;

[0053] Figure 5 It is a structural diagram of a video surveillance streaming media decision device provided by an embodiment of the present invention;

[0054] Figure 6 It is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0055] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0056] In view of the problem of long waiting time for starting streaming of video surveillance equipment in the prior art, an embodiment of the present invention provides a video surveillance streaming media decision method. Figure 2 FIG. 1 is one of the flow charts of the video surveillance streaming media decision method provided by an embodiment of the present invention. Figure 2 As shown, the method includes:

[0057] Step 210: Obtain a flow start decision factor corresponding to each of the at least one video surveillance device.

[0058] Specifically, the corresponding streaming decision factor can be obtained through the configuration file of each video surveillance device. The streaming decision factor is a factor that affects whether each video surveillance device starts streaming and the streaming start order.

[0059] It should be noted that starting the stream is used to indicate that after the target video surveillance device is determined and the target video surveillance device starts live broadcast, the target video surveillance device is requested to obtain the video stream collected by the target video surveillance device. The target video surveillance device is any one of the at least one video surveillance device.

[0060] Optionally, the video surveillance device is an IP Camera (IP Camera, IPC), which can transmit the collected video stream to the terminal through the network. The IP camera can include a hemispherical camera, a wired cylindrical camera, a high-speed ball camera, and a wireless camera, etc., which is not limited in the embodiment of the present invention.

[0061] Optionally, the start-up decision factor may include: an alarm factor, a user habit factor, and the number of devices corresponding to at least one video surveillance device. For example:

[0062] 1) The alarm factor can be: after device B generates a sound alarm, a fall alarm, a human alarm, a motion alarm, a video loss alarm, and a network fluctuation alarm, the user turns on the live broadcast of device B. The sound alarm can include a crying alarm, a pet sound alarm, a motor vehicle alarm, a non-motor vehicle alarm, etc. In addition, after device B generates a preset alarm such as a doorbell alarm, the streaming media server needs to pull the stream to device B and forward the video stream to the APP in the terminal, so that the user can turn on the live broadcast of device B.

[0063] 2) The user habit factor may be: the user is accustomed to opening the live broadcast of device C at location S and time T.

[0064] 3) The number of devices can be: if the user has only one video surveillance device, the user can only open the live broadcast of this video surveillance device; if the user has at least two video surveillance devices, the user can open the live broadcast of any video surveillance device.

[0065] Step 220: Determine the probability of starting streaming for each of the video surveillance devices based on the streaming decision factors.

[0066] Specifically, after determining each streaming decision factor, the streaming probability of each video surveillance device can be further determined, so as to determine whether each video surveillance device should pre-start streaming and the corresponding pre-start streaming prediction order when pre-start streaming is determined based on each streaming start probability, so that after the user logs into the account, the streaming media can be directly notified to start streaming, thereby shortening the waiting time for streaming and improving the user experience.

[0067] Furthermore, Figure 3 FIG. 2 is a flow chart of the video surveillance streaming media decision method provided by an embodiment of the present invention. Figure 3 As shown, the determining the probability of starting the flow of each of the video surveillance devices based on each of the start-flow decision factors includes:

[0068] When the first preset condition is met, determining that the probability of starting the flow of the video surveillance device is 1;

[0069] In the case where the first preset condition is not met, determining, for each video surveillance device, the occurrence probability corresponding to the start-flow decision factor; and determining the start-flow probability of the video surveillance device based on the start-flow decision factor and the corresponding occurrence probability;

[0070] The first preset condition includes:

[0071] A preset alarm in the flow start decision factor occurs; or,

[0072] The number of devices corresponding to the at least one video surveillance device is one.

[0073] Specifically, Figure 3 As shown in the figure, if a preset alarm occurs on the video surveillance device, for example, when a doorbell alarm occurs on device B, at this time, the probability of device B immediately opening the live broadcast to start streaming is 1, that is, the streaming media server needs to pull the stream to device B and forward the video stream to the APP. The probability of starting streaming is 1. If the video surveillance device does not have a preset alarm, the number of video surveillance devices is further determined. If the number of devices is one, then no matter what type of alarm occurs or what habits the user has, the corresponding probability of starting streaming of the video surveillance device is always 1. If the number of devices is at least two, the probability of occurrence of the start-streaming decision factor is further determined, and based on the occurrence probability and the start-streaming decision factor, the start-streaming probability of each video surveillance device is further determined.

[0074] Further, the determining the occurrence probability corresponding to the flow start decision factor includes:

[0075] Get historical start flow data;

[0076] Based on the historical flow-starting data, determining a preset probability of occurrence corresponding to the flow-starting decision factor;

[0077] When the video surveillance device does not meet the second preset condition, determining the occurrence probability corresponding to the start-flow decision factor to be 0;

[0078] When the start-up of the video surveillance device satisfies the second preset condition, the preset occurrence probability is determined as the occurrence probability corresponding to the start-up decision factor.

[0079] Specifically, when determining the occurrence probability corresponding to the start-up decision factor, it is necessary to first obtain historical start-up data. The historical start-up data may include the operation data of multiple users starting live streaming to the video surveillance device within a preset time period. By analyzing the historical start-up data, analyzing the alarms of different alarm types that occur on the video surveillance device and the situations in which the user starts live streaming to the video surveillance device according to the user's habits, the preset occurrence probability corresponding to the start-up decision factor is determined, that is, the preset occurrence probability corresponding to the start-up decision factor is determined based on the data ratio of the user starting live streaming to the video surveillance device under different circumstances. And then, based on whether the video surveillance device meets the second preset condition, it can be further determined whether the occurrence probability corresponding to the start-up decision factor is 0 or the preset occurrence probability.

[0080] It should be noted that, since the above-mentioned start-streaming decision factors include an alarm factor and a user habit factor, the above-mentioned preset occurrence probability includes a first preset occurrence probability corresponding to the user habit factor and a second preset occurrence probability corresponding to the alarm factor. The occurrence probability of the above-mentioned start-streaming decision factors includes a habit occurrence probability corresponding to the user habit factor and an alarm occurrence probability corresponding to the alarm factor. Among them, the alarm occurrence probability is the probability that the user immediately starts streaming to the video surveillance device to start live broadcasting after the video surveillance device generates an alarm. The habit occurrence probability is the probability that the user starts streaming to the video surveillance device to start live broadcasting at the habitual place and time according to his own habits.

[0081] Optionally, when determining the preset occurrence probability corresponding to the flow initiation decision factor, it may be determined based on all or part of the historical flow initiation data.

[0082] Optionally, when selecting historical flow start data corresponding to multiple users, users may be randomly selected.

[0083] For example, taking the historical streaming data including the operation data of 100 randomly selected users within 5 days as an example, a part of the historical streaming data of 10 users within 5 days is randomly selected. In this part of the historical streaming data, there are data in which the user starts streaming to the video surveillance device at a location and time that does not conform to the user's habit, and does not start streaming after the video surveillance device alarms, that is, there are data in which the user starts streaming without an alarm and irregularly. Therefore, in order to ensure the accuracy of the streaming probability, a preset constant can be pre-set according to the proportion of the data in which the user starts streaming without an alarm and irregularly in this part of the historical streaming data. For example, if the proportion of the data in which the user starts streaming without an alarm and irregularly in this part of the historical streaming data is 20%, then the preset constant is set to 0.2. Afterwards, in this part of the historical streaming data, the proportion of the data in which the user starts streaming to the video surveillance device and starts live streaming after the video surveillance device alarms is 50%, and the proportion of the data in which the user starts streaming to the video surveillance device and starts live streaming according to habit at a specific location and / or time is 50%. Therefore, the first preset occurrence probability corresponding to the user habit factor is The second preset occurrence probability corresponding to the alarm factor is

[0084] It should be noted that, since this part of the historical streaming data contains overlapping data of users starting streaming and live streaming to the video surveillance device at a specific location and / or time as usual, and starting streaming and live streaming to the video surveillance device after an alarm is generated in the video surveillance device, the sum of the data ratio of users starting streaming and live streaming to the video surveillance device after an alarm is generated in the video surveillance device and the data ratio of users starting streaming and live streaming to the video surveillance device at a specific location and / or time as usual may be less than 100% or greater than or equal to 100%.

[0085] Furthermore, the second preset condition includes: the location and / or time at which the video surveillance device starts streaming is consistent with the user's habits, or the video surveillance device generates alarm information. The location and / or time at which the video surveillance device starts streaming is consistent with the user's habits corresponds to the user's habit factor, and the alarm information generated by the video surveillance device corresponds to the alarm factor. When the video surveillance device starts streaming and does not meet all the conditions in the second preset condition, it can be determined that the habit occurrence probability corresponding to the user habit factor is 0, and the alarm occurrence probability corresponding to the alarm factor is 0.

[0086] In addition, when the start of the video surveillance device satisfies the second preset condition, determining the second preset occurrence probability as the occurrence probability corresponding to the start of the flow decision factor includes:

[0087] When the start of streaming of the video surveillance device satisfies the second preset condition that the start of streaming of the video surveillance device is in accordance with the location and / or time of user habits, the first preset occurrence probability is determined as the habitual occurrence probability corresponding to the user habit factor;

[0088] and / or,

[0089] In the case where the start of the video surveillance device satisfies the second preset condition that the video surveillance device generates alarm information, the second preset occurrence probability is determined as the alarm occurrence probability corresponding to the alarm factor.

[0090] Specifically, when the start of the video surveillance device meets the location and / or time of the user's habit but does not meet the requirement for the video surveillance device to generate alarm information, the first preset occurrence probability may be determined as the habitual occurrence probability corresponding to the user's habit factor, and the alarm occurrence probability corresponding to the alarm factor may be determined as 0. When the start of the video surveillance device meets the requirement for the video surveillance device to generate alarm information but does not meet the location and / or time of the user's habit, the second preset occurrence probability may be determined as the alarm occurrence probability corresponding to the alarm factor, and the habitual occurrence probability corresponding to the user's habit factor may be determined as 0. When the start of the video surveillance device meets the requirement for the video surveillance device to generate alarm information and meets the location and / or time of the user's habit, the first preset occurrence probability may be determined as the habitual occurrence probability corresponding to the user's habit factor, and the second preset occurrence probability may be determined as the alarm occurrence probability corresponding to the alarm factor.

[0091] For example, if the second preset occurrence probability corresponding to the alarm factor is 0.4, if the alarm information is received, it means that the video surveillance device has an alarm, and the alarm information is generated, then the second preset occurrence probability corresponding to the alarm factor is determined as the alarm occurrence probability corresponding to the alarm factor, that is, the alarm occurrence probability is 0.4. If the alarm information is not received, it means that the video surveillance device has not an alarm, and the alarm occurrence probability corresponding to the alarm factor is 0.

[0092] For example, the first preset occurrence probability corresponding to the user habit factor is 0.4. Among the user's multiple video surveillance devices, for each video surveillance device, if the start-up of the video surveillance device conforms to the location and / or time of the user's habits, that is, based on the historical start-up data of the video surveillance device, it is determined that the location where the video surveillance device often starts streaming conforms to the location of the user's habits, and / or the time when the video surveillance device often starts streaming conforms to the time of the user's habits, then the second preset occurrence probability of the habit is determined as the habit occurrence probability corresponding to the user habit factor corresponding to the video surveillance device, that is, the habit occurrence probability is 0.4. Conversely, if the start-up of the video surveillance device neither conforms to the location of the user's habits nor conforms to the time of the user's habits, then the habit occurrence probability corresponding to the user habit factor corresponding to the video surveillance device is determined to be 0.

[0093] Further, the determining the probability of starting the flow of the video surveillance device based on the decision factor for starting the flow and the corresponding probability of occurrence includes:

[0094] Upon receiving the alarm information corresponding to the video surveillance device, determining the target alarm type based on the alarm information;

[0095] Determining an alarm weight corresponding to the alarm factor based on the target alarm type and the historical flow start data;

[0096] Determining a habit weight corresponding to the user habit factor based on the historical initiation data;

[0097] Determine a first flow start probability based on a preset constant, the alarm occurrence probability and the alarm weight corresponding to the alarm factor, and the habit occurrence probability and the habit weight corresponding to the user habit factor;

[0098] The start-flow decision factor also includes the number of devices corresponding to the at least one video surveillance device;

[0099] Based on the number of devices, determine the device weight corresponding to the video surveillance device;

[0100] Based on the first streaming start probability and the device weight, a streaming start probability of the video surveillance device is determined.

[0101] Specifically, after obtaining the historical start-up data, the initial alarm weights corresponding to different alarm types in the alarm factor can be determined based on the data ratios corresponding to different alarm types in the historical start-up data, and when the video surveillance device receives the alarm information, the target alarm type is determined based on the alarm information, and then the alarm weight corresponding to the target alarm type is determined by traversing the different alarm types corresponding to the initial alarm weight.

[0102] Taking the historical streaming data including the operation data of 100 randomly selected users in one day, and the operation data are all data of receiving alarm information as an example, it can be known from the analysis of all the operation data that 1) after the alarm information indicates the occurrence of alarm situations such as crying alarm, pet sound alarm and fall alarm, the proportion of data that users start streaming to the video surveillance device at the first moment is 80%, and the proportion of data that users start streaming to the video surveillance device at a non-first moment is 20%. Therefore, the alarm type corresponding to the alarm situations such as crying alarm, pet sound alarm and fall alarm is determined as a Class I alarm. 2) After the occurrence of alarm situations such as human alarm, motor vehicle alarm and non-motor vehicle alarm, there are users who will start streaming to the video surveillance device at the first moment or at a non-first moment, and there are also users who do not start streaming to the video surveillance device. Therefore, the alarm type corresponding to the alarm situations such as human alarm, motor vehicle alarm and non-motor vehicle alarm is determined as a Class II alarm. 3) After the occurrence of motion alarm, video loss alarm, network fluctuation alarm and other alarm situations, there is no operation data of the user starting the live stream to the video surveillance device at the first moment, there is a user starting the live stream to the video surveillance device at a non-first moment, and there is also a user not starting the live stream to the video surveillance device. Therefore, the alarm types corresponding to the motion alarm, video loss alarm and network fluctuation alarm are determined as three types of alarms. The priority of the first type of alarm is higher than that of the second type of alarm, and the priority of the second type of alarm is higher than that of the third type of alarm. According to the above alarm types, the weight range and initial alarm weight corresponding to different alarm types can be further determined as shown in Table 1, wherein the initial alarm weight is determined according to the average of the upper limit and lower limit of the weight range.

[0103] Table 1 Weight ranges and initial alarm weights corresponding to different alarm types

[0104]

[0105] It should be noted that the above-mentioned first moment indicates that the time interval between the moment when the user starts streaming live to the video surveillance device and the moment when the alarm occurs is shorter, and the above-mentioned non-first moment indicates that the time interval between the moment when the user starts streaming live to the video surveillance device and the moment when the alarm occurs is longer.

[0106] Secondly, according to the data proportions corresponding to the user's accustomed locations and times in the historical start-up data, the initial habit weights corresponding to the locations and times that conform to the user's habits in the user habit factor are determined, and then according to the location or time that conforms to the user's habits of the video surveillance device, the habit weight corresponding to the user habit factor is further determined.

[0107] Table 2 Statistics of different users opening live broadcasts at the same location

[0108] user Day 1 Day 2 Day 3 Day 4 Day 5 The proportion of flow starting at the same location User 1 Location A Location A / Location A Location A 80％ User 2 Location B Location B Location B Location B Location B 100％ User 3 Location C / / Location C / 40％ User 4 Location D / Location D / Location D 60％ User 5 Location E Location E Location E Location E / 80％ User 6 Location F / Location F / / 40％ User 7 / Location G / Location G Location G 60％ User 8 Location H Location H / / / 40％ User 9 Location I Location I / / Location I 60％ User 10 Location J / Location J / / 40％

[0109] Taking the historical streaming data including the operation data of 10 randomly selected users within 5 days, and the operation data all conform to the location or time of the user's habits as an example, it can be known from the analysis of the time and location of the streaming in all the operation data, as shown in Table 2, among the above users 1 to 10, the average proportion of data starting live streaming to the video surveillance device at the same location is 60%. Therefore, the initial habit weight corresponding to the location that conforms to the user's habits is determined to be 0.6. Similarly, it can be determined that the average proportion of data starting live streaming to the video surveillance device at the same time is 40%, and the initial habit weight corresponding to the time that conforms to the user's habits is determined to be 0.4.

[0110] Secondly, according to the number of video surveillance devices corresponding to the user, the device weight corresponding to the video surveillance device is determined. When the number of devices is 1, the probability of selecting the video surveillance device is 1. When the number of devices is N, under the same conditions, the probability of the user selecting each video surveillance device is 1 / N. Therefore, according to the number of devices, it can be further determined that the device weight corresponding to the video surveillance device is 1 or 1 / N.

[0111] After determining the preset constant, habit occurrence probability, alarm occurrence probability, habit weight and alarm weight, the first start-up probability can be determined according to formula (1), and then the start-up probability of the video surveillance device can be determined according to formula (2) based on the product of the first start-up probability and the device weight. Formula (1) is:

[0112] P1=P c +P a *w a +P b *w b

[0113] Among them, P1 represents the first flow start probability, P c Indicates the preset constant, P b represents the probability of habit occurrence, P a represents the probability of alarm occurrence, and P c +P a +P b =1,w b represents the habit weight, w a Indicates the alarm weight.

[0114] Formula (2) is:

[0115] P t =P1*w s

[0116] Among them, P trepresents the start-up probability of the video surveillance device, w s Indicates the device weight.

[0117] Optionally, the historical flow start data may be all the historical flow start data, or may be part of the historical flow start data randomly selected from all the historical flow start data. For example, if all the historical flow start data includes the operation data of 100 randomly selected users within 5 days, then the historical flow start data may be the operation data of 100 randomly selected users within any day.

[0118] Step 230: Based on each of the streaming start probabilities, determine the streaming media decision result corresponding to the at least one video surveillance device, wherein the streaming media decision result is used to characterize whether each of the video surveillance devices has pre-started streaming, and when the streaming media decision result characterizes whether each of the video surveillance devices has pre-started streaming, the streaming media decision result carries the corresponding pre-start streaming prediction order.

[0119] Specifically, after determining the streaming start probability corresponding to each video surveillance device, it can be determined whether each video surveillance device should pre-start the streaming according to the size of the streaming start probability. If the video surveillance device determines to pre-start the streaming, the pre-start prediction order of each video surveillance device that pre-starts the streaming is further determined based on the sorting result of the streaming start probability. That is, when each video surveillance device pre-starts the streaming, the streaming media decision result also carries the pre-start prediction order of each video surveillance device that pre-starts the streaming.

[0120] Furthermore, the determining, based on each of the streaming start probabilities, a streaming media decision result corresponding to the at least one video surveillance device includes:

[0121] Compare each of the flow start probabilities with a preset flow start threshold value to determine a flow start decision result;

[0122] Sorting the start-stream probabilities in descending order, and determining a start-stream priority sorting result corresponding to the at least one video surveillance device;

[0123] Based on the stream start priority sorting result and the stream start decision result, a streaming media decision result corresponding to the at least one video surveillance device is determined.

[0124] Specifically, after determining the start-up probability corresponding to each video surveillance device, each start-up probability can be sorted in order from large to small to obtain the start-up priority sorting result corresponding to each video surveillance device. The start-up priority sorting result is used to determine the start-up order of each video surveillance device, that is, the greater the start-up probability, the higher the start-up order. At the same time, in order to ensure the prediction accuracy of the streaming media decision result, each start-up probability can be compared with the start-up preset threshold to obtain the start-up decision result. The start-up decision result is used to determine whether the video surveillance device needs to start the stream in advance. If the start-up probability is greater than the start-up preset threshold, the video surveillance device is started in advance. If the start-up probability is less than or equal to the start-up preset threshold, the video surveillance device is not started in advance. According to the intersection of the start-up decision result and the start-up priority sorting result, the streaming media decision result corresponding to at least one video surveillance device can be further determined.

[0125] For example, the different situations corresponding to user A, user B, user C and user D are taken as an example, and the preset threshold for starting streaming is 0.1. To simplify the description, the following devices are all video surveillance devices, and the determination processes of the streaming media decision results are as follows:

[0126] 1) User A has only one device. Regardless of whether the preset constant, habit occurrence probability, alarm occurrence probability, habit weight and alarm weight change, the start-up probability PRA1 of device A1 is always 1, that is, PRA1=1, and PRA1>0.1, then the user will always actively click on the device A1, that is, the device A1 will be pre-started when user A logs in to the APP.

[0127] 2) User B has two devices, among which device B1 generates a preset alarm, then the streaming probability PRB1 of device B1 is 1, that is, PRB1=1; device B2 does not generate an alarm, but the location and time are in line with the user's habits, that is, user B is accustomed to clicking device B2 at a specific location and time to watch live broadcasts in the historical time. From the above, it can be seen that the preset constant is 0.2, the alarm occurrence probability is 0, the habit occurrence probability is 0.4, the initial habit weight corresponding to the location that is in line with the user's habits is 0.6, and the initial habit weight corresponding to the time that is in line with the user's habits is 0.4. User B has two devices, and the device weight is 1 / 2. According to formula (1) and formula (2), it can be determined that the streaming probability of device B2 is PRB2=(0.2+0+(0.6+0.4)*0.4)*(1 / 2)=0.3. Since PRB1>PRB2>0.1, the determined streaming media decision result is that both device B1 and device B2 are pre-started, and the order of starting the streaming is to start the streaming on device B1 first and then on device B2.

[0128] 3) User C has 3 devices, among which device C1 has a crying alarm and the location is consistent with the user's habit, that is, in the historical time, user C is only accustomed to clicking device C1 at a specific location to watch the live broadcast. Then, the start-streaming probability of device C1 can be determined as PRC1=(0.2+0.8*0.4+0.6*0.4)*(1 / 3)=0.253; device C2 has no alarm, but it is consistent with the user's habitual time, then the start-streaming probability of device C2 is PRC2=(0.2+0.4*0.4)*(1 / 3)=0.12; device C3 has no alarm and does not conform to the user's habitual location and time, then the start-streaming probability of device C3 is PRC3=(0.2)*(1 / 3)=0.067. Since PRC1>PRC2>0.1>PRC3, the determined streaming media decision result is to pre-start streaming to devices C1 and C2, and the streaming start order is to start streaming to device C1 first, then to device C2, and device C3 is not pre-started.

[0129] 4) User D has 4 devices, none of which have any alarms, and none of which conform to the user's usual locations and times. The probability of starting streaming for device D1, device D2, device D3, and device D4 is PRD1=PRD2=PRD3=PRD4=(0.2)*(1 / 4)=0.04. Since PRD1=PRD2=PRD3=PRD4<0.1, the determined streaming media decision result is that device D1, device D2, device D3, and device D4 will not be pre-started.

[0130] Further, such as Figure 3 As shown, the method also includes:

[0131] Get abnormal start flow data;

[0132] In the case where the abnormal start-up data is different from the start-up priority ranking result, based on the abnormal start-up data, determining the abnormal factor corresponding to the video surveillance device from the start-up decision factors; updating the abnormality occurrence probability corresponding to the abnormal factor, and updating the start-up probability of the video surveillance device based on the abnormality occurrence probability;

[0133] and / or,

[0134] When the abnormal flow start data is different from the flow start decision result, the flow start preset threshold is updated based on the abnormal flow start data.

[0135] Specifically, after the signaling server obtains the streaming media decision result and starts streaming based on the streaming media decision result and the user trigger result, the user trigger result can be obtained as feedback data, and abnormal streaming start data different from the streaming media decision result can be obtained from the feedback data to optimize according to the abnormal streaming start data.

[0136] If the abnormal start-up data is different from the start-up priority sorting result, that is, the start-up order of the device is different, the abnormal start-up data can be further analyzed to determine the abnormal factors that cause the different start-up orders of the devices, and according to the abnormal start-up data, the abnormal occurrence probability corresponding to the abnormal factor is updated, and then the start-up probability corresponding to the video surveillance device is updated. For example, user E has two devices, and device E1 has a crying alarm, but it does not meet the user's habitual location and time, then the start-up probability of device E1 PRE1 = (0.2 + 0.8 * 0.4) * (1 / 2) = 0.26; device E2 has no alarm, but meets the user's habitual location and time, then the start-up probability of device E2 PRE2 = (0.2 + (0.6 + 0.4) * 0.4) * (1 / 2) = 0.3; since PRE2> PRE1> 0.1, it can be determined that the streaming media decision result is that both device E1 and device E2 are pre-started, and the start-up order is to start the stream on device E2 first, and then start the stream on device E1. However, in actual use, the user first turns on the live broadcast of device E1, and then turns on the live broadcast of device E2, making the user trigger result different from the streaming media decision result. Therefore, the abnormal streaming data corresponding to the user trigger result is analyzed to determine that the user is more concerned about the alarm factor rather than the habit factor. Therefore, the abnormal factors are determined to be the alarm factor and the habit factor. After that, the alarm occurrence probability is gradually increased, and at the same time, the habit occurrence probability is gradually reduced. After each increase in the alarm occurrence probability and the reduction in the habit occurrence probability, the streaming probability PRE1 of device E1 and the streaming probability PRE2 of device E2 are determined to determine whether the streaming probability PRE1 of device E1 is greater than the streaming probability PRE2 of device E2. If the streaming probability PRE1 of device E1 is greater than the streaming probability PRE2 of device E2, the alarm occurrence probability is updated to the increased alarm occurrence probability, and the habit occurrence probability is updated to the reduced habit occurrence probability. If the start-up probability PRE1 of device E1 is less than or equal to the start-up probability PRE2 of device E2, the alarm occurrence probability continues to increase, and at the same time, the habit occurrence probability continues to decrease until the start-up probability PRE1 of device E1 is greater than the start-up probability PRE2 of device E2 and the iteration is stopped.For example, taking a step of 0.025 as an example, update the alarm occurrence probability to 0.425 and the habit occurrence probability to 0.375. After the update, the start flow probability PRE1 of device E1 = (0.2 + 0.8 * 0.425) * (1 / 2) = 0.27, and the start flow probability PRE2 of device E2 = (0.2 + (0.6 + 0.4) * 0.375) * (1 / 2) = 0.2875. Since PRE1 < PRE2, continue to update the alarm occurrence probability to 0.45 and the habit occurrence probability to 0.35. After the update, the start flow probability PRE1 of device E1 = (0.2 + 0.8 * 0.45) * (1 / 2) = 0.28, and the start flow probability PRE2 of device E2 = (0.2 + (0.6 + 0.4) * 0.35) * (1 / 2) = 0.275. At this time, PRE1 > PRE2 > 0.1, making the optimized streaming media decision result the same as the user trigger result in the abnormal start flow data. Therefore, update the alarm occurrence probability to 0.45 and the habit occurrence probability to 0.35.

[0137] It should be noted that, since the above abnormal factors are alarm factors and habit factors, the increase in the alarm probability is equal to the decrease in the habit probability, but the increase in two adjacent increases may be equal or unequal, and the embodiment of the present invention does not limit this. In addition, if the two devices of the above user E do not conform to the user's habitual location and / or time, and no alarm occurs, at this time, the preset constant can be updated, and according to the increase or decrease in the preset constant, the alarm probability and the habit probability can be updated based on the ratio of the alarm probability and the habit probability, respectively, to ensure that the sum of the preset constant, the alarm probability and the habit probability is 1. For example, if the preset constant is 0.2, the probability of alarm occurrence and the probability of habit occurrence are both 0.4, and the preset threshold for starting streaming is 0.06, if the user has 4 devices, and no alarm occurs, and none of them conform to the user's habitual location and / or time, then the probability of starting streaming for each device is 0.2*(1 / 4)=0.05. Since 0.05<0.06, the determined streaming decision result is that each device does not start streaming in advance. However, in actual use, the user turned on the live broadcast of three of the devices, making the user trigger result different from the streaming decision result. Therefore, the preset constant is increased by 0.05 each time with a step of 0.05, and then the new start probability of each device is calculated to be 0.25*(1 / 4)=0.0625. At this time, since 0.0625>0.06, the preset constant is updated to 0.25 while the preset start probability remains unchanged. At the same time, since the preset constant is increased by 0.05, and the ratio of the habit occurrence probability to the alarm occurrence probability is 1:1, in order to ensure that the sum of the preset constant, the alarm occurrence probability and the habit occurrence probability is 1, the habit occurrence probability and the alarm occurrence probability need to be reduced by 0.025, that is, the habit occurrence probability and the alarm occurrence probability are both reduced to 0.375. After updating the above preset constants, habit occurrence probability and alarm occurrence probability, the streaming media decision result is updated to pre-start streaming for each device, and the start probability of each device is equal, so that the optimized streaming media decision result is the same as the user trigger result in actual use.

[0138] In addition, the preset start threshold can also be updated to make the optimized streaming media decision result the same as the user trigger result in actual use. Specifically, if the abnormal start data is different from the start decision result, that is, because the start probability of the device is less than or equal to the preset start threshold, the device will not be pre-started, but in actual use, if there is a user who starts the device with a start probability less than or equal to the preset start threshold, the preset start threshold can be updated to optimize the streaming media decision result. For example, user F has 5 devices, only device F1 has a motion alarm, and other devices have no alarm, and it does not meet the user's usual location and time. Then the corresponding streaming probability of device F1 is PRF1 = (0.2 + 0.2 * 0.4) * (1 / 5) = 0.06, and the streaming probability of devices F2-F5 is PRF2 = PRF3 = PRF4 = PRF5 = (0.2 * 1) * (1 / 5) = 0.04. Since 0.1>PRF1>PRF2 = PRF3 = PRF4 = PRF5, the determined streaming decision result is not to start streaming in advance for any device. However, in actual use, the user turned on the live broadcast of device F1. Therefore, the preset threshold for starting streaming can be updated to a value greater than 0.04 and less than 0.06. For example, the preset threshold for starting streaming is updated to 0.05. After updating the preset threshold for starting streaming, since PRF1>0.05>PRF2=PRF3=PRF4=PRF5, it can be determined that the streaming media decision result is to pre-start streaming only for device F1, and not for other devices.

[0139] Furthermore, Figure 4 Schematic diagram of the video surveillance device start-up process provided by the embodiment of the present invention. Figure 4 As shown, the execution subject corresponding to the video surveillance streaming media decision method provided by the embodiment of the present invention is a streaming media decision server. After the streaming media decision result is determined in advance and the signaling server detects that the user has logged in to the account, the streaming media decision server can send the streaming media decision result to the signaling server. After obtaining the streaming media decision result, the signaling server pre-starts the stream to each video surveillance device according to the streaming media decision result. After that, the start-up address corresponding to the streaming media server is sent to the user's client, and the start-up result is sent to the streaming media server. After starting the stream, the video surveillance device sends the video stream to the streaming media server, so that after the user triggers the video surveillance device that has started the stream, the live broadcast can be directly started, which greatly shortens the waiting time for starting the stream. In addition, after the user triggers, the signaling server will feedback the user triggering effect to optimize the occurrence probability corresponding to the start-up decision factor.

[0140] The video surveillance streaming media decision method provided by the embodiment of the present invention determines the streaming probability of each video surveillance device according to the streaming decision factor corresponding to each video surveillance device, and determines the streaming media decision results corresponding to all video surveillance devices according to each streaming probability, that is, determines whether each video surveillance device starts streaming in advance, and in the case where each video surveillance device starts streaming in advance, further determines the pre-streaming prediction order of each video surveillance device, and pre-streams the video surveillance devices according to the pre-streaming prediction order, so that after the user triggers it, the live broadcast of the corresponding video surveillance device can be directly started, which greatly shortens the waiting time for starting streaming and improves the user experience.

[0141] The video surveillance streaming media decision device provided by the present invention is described below. The video surveillance streaming media decision device described below and the video surveillance streaming media decision method described above can be referenced to each other.

[0142] The embodiment of the present invention also provides a video surveillance streaming media decision device, Figure 5 is a schematic diagram of the structure of a video surveillance streaming media decision device provided by an embodiment of the present invention, such as Figure 5 As shown, the video surveillance streaming media decision device 500 includes: an acquisition module 510, a first determination module 520 and a second determination module 530, wherein:

[0143] An acquisition module 510 is used to acquire a flow start decision factor corresponding to each of the at least one video surveillance device;

[0144] A first determination module 520, configured to determine a probability of starting a flow for each of the video surveillance devices based on each of the start-flow decision factors;

[0145] The second determination module 530 is used to determine the streaming media decision result corresponding to the at least one video surveillance device based on each of the streaming start probabilities, and the streaming media decision result is used to characterize whether each of the video surveillance devices pre-starts the streaming, and when the streaming media decision result characterizes that each of the video surveillance devices pre-starts the streaming, the streaming media decision result carries the corresponding pre-start prediction order.

[0146] The video surveillance streaming media decision device provided by the embodiment of the present invention determines the streaming probability of each video surveillance device according to the streaming decision factor corresponding to each video surveillance device, and determines the streaming media decision results corresponding to all video surveillance devices according to each streaming probability, that is, determines whether each video surveillance device starts streaming in advance, and in the case where each video surveillance device starts streaming in advance, further determines the pre-streaming prediction order of each video surveillance device, and pre-streams the video surveillance devices according to the pre-streaming prediction order, so that after the user triggers it, the live broadcast of the corresponding video surveillance device can be directly started, which greatly shortens the waiting time for starting streaming and improves the user experience.

[0147] Optionally, the first determining module 520 is specifically configured to:

[0148] When the first preset condition is met, determining that the probability of starting the flow of the video surveillance device is 1;

[0149] In the case where the first preset condition is not met, determining, for each video surveillance device, the occurrence probability corresponding to the start-flow decision factor; and determining the start-flow probability of the video surveillance device based on the start-flow decision factor and the corresponding occurrence probability;

[0150] The first preset condition includes:

[0151] A preset alarm in the flow start decision factor occurs; or,

[0152] The number of devices corresponding to the at least one video surveillance device is one.

[0153] Optionally, the first determining module 520 is specifically configured to:

[0154] Get historical start flow data;

[0155] Based on the historical flow-starting data, determining a preset probability of occurrence corresponding to the flow-starting decision factor;

[0156] When the video surveillance device does not meet the second preset condition, determining the occurrence probability corresponding to the start-flow decision factor to be 0;

[0157] When the start-up of the video surveillance device satisfies the second preset condition, the preset occurrence probability is determined as the occurrence probability corresponding to the start-up decision factor.

[0158] Optionally, the start-up decision factor includes an alarm factor and a user habit factor; the preset occurrence probability includes a first preset occurrence probability corresponding to the user habit factor and a second preset occurrence probability corresponding to the alarm factor; the occurrence probability includes the habit occurrence probability corresponding to the user habit factor and the alarm occurrence probability corresponding to the alarm factor.

[0159] Optionally, the first determining module 520 is specifically configured to:

[0160] When the start of streaming of the video surveillance device satisfies the second preset condition that the start of streaming of the video surveillance device is in accordance with the location and / or time of user habits, the first preset occurrence probability is determined as the habitual occurrence probability corresponding to the user habit factor;

[0161] and / or,

[0162] In the case where the start of the video surveillance device satisfies the second preset condition that the video surveillance device generates alarm information, the second preset occurrence probability is determined as the alarm occurrence probability corresponding to the alarm factor.

[0163] Optionally, the flow initiation decision factor also includes the number of devices corresponding to the at least one video surveillance device.

[0164] Optionally, the first determining module 520 is specifically configured to:

[0165] Upon receiving the alarm information corresponding to the video surveillance device, determining the target alarm type based on the alarm information;

[0166] Determining an alarm weight corresponding to the alarm factor based on the target alarm type and the historical flow start data;

[0167] Determining a habit weight corresponding to the user habit factor based on the historical initiation data;

[0168] Determine a first flow start probability based on a preset constant, the alarm occurrence probability and the alarm weight corresponding to the alarm factor, and the habit occurrence probability and the habit weight corresponding to the user habit factor;

[0169] Based on the number of devices, determine the device weight corresponding to the video surveillance device;

[0170] Based on the first streaming start probability and the device weight, a streaming start probability of the video surveillance device is determined.

[0171] Optionally, the second determining module 530 is specifically configured to:

[0172] Compare each of the flow start probabilities with a preset flow start threshold value to determine a flow start decision result;

[0173] Sorting the start-stream probabilities in descending order, and determining a start-stream priority sorting result corresponding to the at least one video surveillance device;

[0174] Based on the stream start priority sorting result and the stream start decision result, a streaming media decision result corresponding to the at least one video surveillance device is determined.

[0175] Optionally, the video surveillance streaming media decision device 500 further includes an update module, which is specifically used to:

[0176] Get abnormal start flow data;

[0177] In the case where the abnormal start-up data is different from the start-up priority ranking result, based on the abnormal start-up data, determining the abnormal factor corresponding to the video surveillance device from the start-up decision factors; updating the abnormality occurrence probability corresponding to the abnormal factor, and updating the start-up probability of the video surveillance device based on the abnormality occurrence probability;

[0178] and / or,

[0179] When the abnormal flow start data is different from the flow start decision result, the flow start preset threshold is updated based on the abnormal flow start data.

[0180] Figure 6 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention, such as Figure 6 As shown, the electronic device may include: a processor 610, a communication interface 620, a memory 630 and a communication bus 640, wherein the processor 610, the communication interface 620 and the memory 630 communicate with each other through the communication bus 640. The processor 610 may call the logic instructions in the memory 630 to execute the video surveillance streaming media decision method, which includes:

[0181] Obtaining a flow start decision factor corresponding to each of the video surveillance devices in at least one video surveillance device;

[0182] Based on each of the start-flow decision factors, determining the start-flow probability of each of the video surveillance devices;

[0183] Based on each of the streaming start probabilities, a streaming media decision result corresponding to the at least one video surveillance device is determined, and the streaming media decision result is used to characterize whether each of the video surveillance devices has pre-started streaming. In the case where the streaming media decision result characterizes whether each of the video surveillance devices has pre-started streaming, the streaming media decision result carries a corresponding pre-start prediction order.

[0184] In addition, the logic instructions in the above-mentioned memory 630 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, 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 a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0185] On the other hand, the present invention further provides a computer program product, the computer program product includes a computer program, the computer program can be stored in a computer-readable storage medium, when the computer program is executed by a processor, the computer can execute the video surveillance streaming media decision method provided by the above methods, the method includes:

[0186] Obtaining a flow start decision factor corresponding to each of the video surveillance devices in at least one video surveillance device;

[0187] Based on each of the start-flow decision factors, determining the start-flow probability of each of the video surveillance devices;

[0188] Based on each of the streaming start probabilities, a streaming media decision result corresponding to the at least one video surveillance device is determined, and the streaming media decision result is used to characterize whether each of the video surveillance devices has pre-started streaming. In the case where the streaming media decision result characterizes whether each of the video surveillance devices has pre-started streaming, the streaming media decision result carries a corresponding pre-start prediction order.

[0189] In another aspect, the present invention further provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the video surveillance streaming media decision method provided by the above methods is implemented, and the method includes:

[0190] Obtaining a flow start decision factor corresponding to each of the video surveillance devices in at least one video surveillance device;

[0191] Based on each of the start-flow decision factors, determining the start-flow probability of each of the video surveillance devices;

[0192] Based on each of the streaming start probabilities, a streaming media decision result corresponding to the at least one video surveillance device is determined, and the streaming media decision result is used to characterize whether each of the video surveillance devices has pre-started streaming. In the case where the streaming media decision result characterizes whether each of the video surveillance devices has pre-started streaming, the streaming media decision result carries a corresponding pre-start prediction order.

[0193] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0194] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0195] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A video surveillance streaming media decision method, characterized in that: include: Obtaining a flow start decision factor corresponding to each of the video surveillance devices in at least one video surveillance device; Based on each of the start-flow decision factors, determining the start-flow probability of each of the video surveillance devices; Based on each of the streaming start probabilities, a streaming media decision result corresponding to the at least one video surveillance device is determined, and the streaming media decision result is used to characterize whether each of the video surveillance devices has pre-started streaming. In the case where the streaming media decision result characterizes whether each of the video surveillance devices has pre-started streaming, the streaming media decision result carries a corresponding pre-start prediction order.

2. The video surveillance streaming media decision method according to claim 1 is characterized in that: The determining the probability of starting the flow of each video surveillance device based on each of the start-flow decision factors includes: When the first preset condition is met, determining that the probability of starting the flow of the video surveillance device is 1; In the case where the first preset condition is not met, determining, for each video surveillance device, the occurrence probability corresponding to the start-flow decision factor; and determining the start-flow probability of the video surveillance device based on the start-flow decision factor and the corresponding occurrence probability; The first preset condition includes: A preset alarm in the flow start decision factor occurs; or, The number of devices corresponding to the at least one video surveillance device is one.

3. The video surveillance streaming media decision method according to claim 2 is characterized in that: The determining the occurrence probability corresponding to the flow start decision factor includes: Get historical start flow data; Based on the historical flow-starting data, determining a preset probability of occurrence corresponding to the flow-starting decision factor; When the video surveillance device does not meet the second preset condition, determining the occurrence probability corresponding to the start-flow decision factor to be 0; When the start-up of the video surveillance device satisfies the second preset condition, the preset occurrence probability is determined as the occurrence probability corresponding to the start-up decision factor.

4. The video surveillance streaming media decision method according to claim 3 is characterized in that: The start-up decision factor includes an alarm factor and a user habit factor; the preset occurrence probability includes a first preset occurrence probability corresponding to the user habit factor and a second preset occurrence probability corresponding to the alarm factor; the occurrence probability includes a habit occurrence probability corresponding to the user habit factor and an alarm occurrence probability corresponding to the alarm factor; When the start of the video surveillance device satisfies the second preset condition, determining the second preset occurrence probability as the occurrence probability corresponding to the start of the flow decision factor includes: When the start of streaming of the video surveillance device satisfies the second preset condition that the start of streaming of the video surveillance device is in accordance with the location and / or time of user habits, the first preset occurrence probability is determined as the habitual occurrence probability corresponding to the user habit factor; and / or, In the case where the start of the video surveillance device satisfies the second preset condition that the video surveillance device generates alarm information, the second preset occurrence probability is determined as the alarm occurrence probability corresponding to the alarm factor.

5. The video surveillance streaming media decision method according to claim 4 is characterized in that: The start-flow decision factor also includes the number of devices corresponding to the at least one video surveillance device; The determining the probability of starting the flow of the video surveillance device based on the decision factor for starting the flow and the corresponding probability of occurrence includes: Upon receiving the alarm information corresponding to the video surveillance device, determining the target alarm type based on the alarm information; Determining an alarm weight corresponding to the alarm factor based on the target alarm type and the historical flow start data; Determining a habit weight corresponding to the user habit factor based on the historical initiation data; Determine a first flow start probability based on a preset constant, the alarm occurrence probability and the alarm weight corresponding to the alarm factor, and the habit occurrence probability and the habit weight corresponding to the user habit factor; Based on the number of devices, determine the device weight corresponding to the video surveillance device; Based on the first streaming start probability and the device weight, a streaming start probability of the video surveillance device is determined.

6. The video surveillance streaming media decision method according to any one of claims 1 to 5, characterized in that: The determining, based on each of the streaming start probabilities, a streaming media decision result corresponding to the at least one video surveillance device includes: Compare each of the flow start probabilities with a preset flow start threshold value to determine a flow start decision result; Sorting the start-stream probabilities in descending order, and determining a start-stream priority sorting result corresponding to the at least one video surveillance device; Based on the stream start priority sorting result and the stream start decision result, a streaming media decision result corresponding to the at least one video surveillance device is determined.

7. The video surveillance streaming media decision method according to claim 6 is characterized in that: The method further comprises: Get abnormal start flow data; In the case where the abnormal start-up data is different from the start-up priority ranking result, based on the abnormal start-up data, determining the abnormal factor corresponding to the video surveillance device from the start-up decision factors; updating the abnormality occurrence probability corresponding to the abnormal factor, and updating the start-up probability of the video surveillance device based on the abnormality occurrence probability; and / or, When the abnormal flow start data is different from the flow start decision result, the flow start preset threshold is updated based on the abnormal flow start data.

8. A video surveillance streaming media decision device, characterized in that: include: An acquisition module, used to acquire a flow start decision factor corresponding to each of the at least one video surveillance device; A first determination module is used to determine the probability of starting the flow of each of the video surveillance devices based on each of the start-flow decision factors; The second determination module is used to determine the streaming media decision result corresponding to the at least one video surveillance device based on each of the streaming start probabilities, and the streaming media decision result is used to characterize whether each of the video surveillance devices pre-starts the streaming, and when the streaming media decision result characterizes that each of the video surveillance devices pre-starts the streaming, the streaming media decision result carries the corresponding pre-start prediction order.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the video surveillance streaming media decision method as described in any one of claims 1-7 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the video surveillance streaming media decision method as described in any one of claims 1 to 7 is implemented.

Images