Multi-channel video transmission method, device and equipment based on polling mechanism

Abstract

The invention provides a multi-channel video transmission method, device and equipment based on a polling mechanism, and relates to the technical field of video transmission. The method comprises the following steps: firstly, determining a target polling transmission mechanism corresponding to a plurality of video acquisition devices; secondly, based on a target polling transmission mechanism, determining a video acquisition device which needs to carry out video transmission currently from the plurality of video acquisition devices, so as to form at least one target video acquisition device; and then, performing a pulling operation on a video stream acquired and formed by each target video acquisition device in the at least one target video acquisition device in the current time period, so that the target video acquisition device transmits the video stream acquired and formed in the current time period to the electronic device. Based on the content, the problem of excessive resource consumption of multi-channel video transmission in the prior art can be improved.

Description

Technical Field

[0001] The present application relates to the field of video transmission technology, and in particular to a method, device and equipment for multi-channel video transmission based on a polling mechanism. Background Art

[0002] With the widespread development of video surveillance and multimedia applications, video transmission technology has been widely used in various fields. Especially in the scenario of multi-channel video transmission, how to effectively manage and transmit multiple video streams has become one of the key issues to improve system efficiency and performance. At present, in the process of multi-channel video transmission, traditional video transmission methods usually adopt continuous transmission to transmit multiple video streams to the receiving end at the same time. This method will lead to higher bandwidth occupancy and computing resource consumption. The more video channels there are, the more serious the problem of excessive resource consumption will be. Summary of the invention

[0003] In view of this, the purpose of the present application is to provide a multi-channel video transmission method, device and equipment based on a polling mechanism to improve the problem of excessive resource consumption in multi-channel video transmission in the prior art.

[0004] To achieve the above objectives, this application adopts the following technical solutions:

[0005] A multi-channel video transmission method based on a polling mechanism is applied to an electronic device, wherein the electronic device is communicatively connected to a plurality of video acquisition devices, each of which is used to perform a video acquisition operation. The multi-channel video transmission method based on a polling mechanism comprises:

[0006] Determine a target polling transmission mechanism corresponding to the multiple video acquisition devices, wherein the target polling transmission mechanism is used to sequentially control video transmission of some of the multiple video acquisition devices, and the video acquisition devices polled twice adjacently are at least partially different;

[0007] Based on the target polling transmission mechanism, determining a video acquisition device that currently needs to perform video transmission among the multiple video acquisition devices to form at least one target video acquisition device;

[0008] In the current time period, a pulling operation is performed on the video stream collected by each of the at least one target video collection device, so that the target video collection device transmits the video stream collected in the current time period to the electronic device.

[0009] In a preferred embodiment of the present application, in the above-mentioned multi-channel video transmission method based on the polling mechanism, the step of determining the target polling transmission mechanism corresponding to the multiple video acquisition devices includes:

[0010] Based on the correlation between every two video acquisition devices among the multiple video acquisition devices, the multiple video acquisition devices are classified and combined to form multiple device classification sets, wherein each of the device classification sets includes at least one video acquisition device, and the video acquisition devices included in every two device classification sets are different;

[0011] Based on the multiple device classification sets, a target polling transmission mechanism corresponding to the multiple video acquisition devices is determined, wherein when the video acquisition device that currently needs to perform video transmission is determined based on the target polling transmission mechanism, at least one target video acquisition device currently determined includes each video acquisition device in a corresponding one of the device classification sets.

[0012] In a preferred embodiment of the present application, in the above-mentioned multi-channel video transmission method based on the polling mechanism, the step of classifying and combining the multiple video acquisition devices based on the correlation between every two video acquisition devices among the multiple video acquisition devices to form multiple device classification sets includes:

[0013] Building a corresponding initial device relationship network based on the multiple video acquisition devices, wherein the initial device relationship network includes the multiple video acquisition devices;

[0014] Respectively obtain device deployment location information, video acquisition area description information, and at least one historical video frame corresponding to each of the multiple video acquisition devices;

[0015] The corresponding device deployment location information, video acquisition area description information and at least one historical video frame pulled out are used as device attribute information of the video acquisition device;

[0016] At least based on the correlation between the corresponding device deployment location information, the initial device relationship network is updated to form a candidate device relationship network corresponding to the initial device relationship network, wherein the correlation between the device deployment location information corresponding to two video acquisition devices having a connection relationship in the candidate device relationship network belongs to a predetermined first correlation relationship;

[0017] Based on the device attribute information of each of the video acquisition devices, the connection relationship in the candidate device relationship network is updated to form a target device relationship network;

[0018] Based on the target device relationship network, the multiple video acquisition devices are classified and combined to form multiple device classification sets.

[0019] In a preferred embodiment of the present application, in the above-mentioned multi-channel video transmission method based on the polling mechanism, the step of updating the connection relationship in the candidate device relationship network based on the device attribute information of each of the video acquisition devices to form a target device relationship network includes:

[0020] Extracting features from the device attribute information of each of the video acquisition devices respectively, and obtaining a first device attribute feature, a second device attribute feature, and a third device attribute feature corresponding to each of the video acquisition devices, wherein the first device attribute feature is obtained by embedding the device deployment location information in the device attribute information, the second device attribute feature is obtained by embedding the video acquisition area description information in the device attribute information, and the third device attribute feature is obtained by convolving the historical video frames in the device attribute information;

[0021] In the multi-level deep feature mining process, the first device attribute feature and the second device attribute feature are alternately aggregated into the third device attribute feature to form a deep device attribute feature corresponding to the video acquisition device;

[0022] Based on the connection relationship in the candidate device relationship network, the deep device attribute feature corresponding to each of the video acquisition devices is updated to form the target device attribute feature corresponding to each of the video acquisition devices;

[0023] Based on the similarity between the target device attribute features corresponding to every two of the video acquisition devices, the connection relationship in the candidate device relationship network is updated to form a target device relationship network.

[0024] In a preferred embodiment of the present application, in the above-mentioned multi-channel video transmission method based on the polling mechanism, the step of updating the deep device attribute characteristics corresponding to each of the video acquisition devices based on the connection relationship in the candidate device relationship network to form the target device attribute characteristics corresponding to each video acquisition device includes:

[0025] Randomly determine a video acquisition device in the candidate device relationship network as a central video acquisition device corresponding to the candidate device relationship network;

[0026] Determine a connection relationship characterization parameter between each video acquisition device in the candidate device relationship network and the central video acquisition device, wherein the connection relationship characterization parameter is positively correlated to the length of the shortest traversal path between the video acquisition device and the central video acquisition device;

[0027] Determining a feature update level corresponding to each of the video acquisition devices based on the connection relationship representation parameter, wherein the connection relationship representation parameter is positively correlated with the feature update level;

[0028] For two video acquisition devices that are adjacent in feature update level and have a connection relationship, the updated depth device attribute features of the video acquisition device with a lower feature update level among the two video acquisition devices are aggregated into the current depth device attribute features of the video acquisition device with a higher feature update level, so as to obtain the updated depth device attribute features of the video acquisition device with a higher feature update level, wherein the current depth device attribute features of the video acquisition device with the lowest feature update level are directly used as the updated depth device attribute features;

[0029] After obtaining the updated depth device attribute feature of a video acquisition device with the highest feature update level, use the updated depth device attribute feature of each video acquisition device as the new current depth device attribute feature, and rotate to execute the step of randomly determining a video acquisition device in the candidate device relationship network as the central video acquisition device corresponding to the candidate device relationship network, wherein the central video acquisition device determined after executing this step twice is different, and after the depth device attribute feature of each video acquisition device in the candidate device relationship network is updated at least once, stop rotating to execute this step, and use the current depth device attribute feature of each video acquisition device as the corresponding target device attribute feature.

[0030] In a preferred embodiment of the present application, in the above-mentioned multi-channel video transmission method based on the polling mechanism, the step of determining the target polling transmission mechanism corresponding to the multiple video acquisition devices based on the multiple device classification sets includes:

[0031] Based on the correlation between every two video acquisition devices in the multiple video acquisition devices, determining the correlation between every two device classification sets in the multiple device classification sets;

[0032] Based on the correlation between every two device classification sets in the multiple device classification sets, the multiple device classification sets are serialized to form a corresponding classification set sequence, wherein the classification set sequence belongs to a ring sequence, and in the classification set sequence, the correlation between every two adjacent device classification sets belongs to a predetermined second correlation;

[0033] Based on the classification set sequence, a target polling transmission mechanism corresponding to the multiple video acquisition devices is determined, wherein when the target polling transmission mechanism performs polling, the order of polling is determined based on the order of the device classification sets in the classification set sequence.

[0034] In a preferred embodiment of the present application, in the above-mentioned multi-channel video transmission method based on the polling mechanism, the step of determining the target polling transmission mechanism corresponding to the multiple video acquisition devices based on the classification set sequence includes:

[0035] Based on the order relationship between the device classification sets in the classification set sequence, determine the candidate polling transmission mechanisms corresponding to the multiple video acquisition devices, wherein the candidate polling transmission mechanisms have a polling order relationship between the video acquisition devices;

[0036] For each device classification set in the multiple device classification sets, each video acquisition device in the device classification set is analyzed to determine a video importance parameter corresponding to the device classification set, and based on the video importance parameter, a polling duration corresponding to the device classification set is determined, wherein the video importance parameter is positively correlated with the polling duration;

[0037] The polling duration corresponding to each of the device classification sets is merged into the candidate polling transmission mechanism to form a target polling transmission mechanism corresponding to the multiple video acquisition devices, wherein the polling duration is used to characterize the time length of the pulled video stream.

[0038] In a preferred embodiment of the present application, in the above-mentioned multi-channel video transmission method based on the polling mechanism, the steps of analyzing each video acquisition device in the device classification set for each of the multiple device classification sets, determining the video importance parameter corresponding to the device classification set, and determining the polling duration corresponding to the device classification set based on the video importance parameter include:

[0039] For each device classification set in the multiple device classification sets, respectively obtain a video stream analysis result of at least one historical video stream corresponding to each video acquisition device in the device classification set, wherein the electronic device is further used to analyze the pulled video stream to obtain a corresponding video stream analysis result, and the video stream analysis result is at least used to characterize whether there is an abnormality;

[0040] Performing statistical analysis on the video stream analysis results of at least one historical video stream corresponding to each video acquisition device in the device classification set to obtain corresponding abnormal distribution parameters, wherein the abnormal distribution parameters are at least used to reflect the number or proportion of abnormal video stream analysis results;

[0041] Based on the abnormal distribution parameters, the video importance parameters corresponding to the device classification set are determined, and based on the video importance parameters, the polling duration corresponding to the device classification set is determined, wherein there is a positive correlation between the quantity or proportion reflected by the abnormal distribution parameters and the video importance parameters corresponding to the device classification set.

[0042] The present application also provides a multi-channel video transmission device based on a polling mechanism, which is applied to an electronic device, wherein the electronic device is communicatively connected to a plurality of video acquisition devices, each of which is used to perform a video acquisition operation, and the multi-channel video transmission device based on a polling mechanism comprises:

[0043] A transmission mechanism determination module, used to determine a target polling transmission mechanism corresponding to the plurality of video acquisition devices, wherein the target polling transmission mechanism is used to sequentially control video transmission of some of the plurality of video acquisition devices, and the video acquisition devices polled twice adjacently are at least partially different;

[0044] A collection device determination module, used to determine a video collection device that currently needs to perform video transmission among the multiple video collection devices based on the target polling transmission mechanism, so as to form at least one target video collection device;

[0045] The video stream pulling module is used to pull the video stream collected by each target video acquisition device in the at least one target video acquisition device within the current time period, so that the target video acquisition device transmits the video stream collected within the current time period to the electronic device.

[0046] Based on the above, the present application also provides an electronic device, including:

[0047] Memory for storing computer programs;

[0048] The processor connected to the memory is used to execute the computer program stored in the memory to implement the above-mentioned multi-channel video transmission method based on the polling mechanism.

[0049] The multi-channel video transmission method, device and equipment based on the polling mechanism provided by the present application firstly determine the target polling transmission mechanism corresponding to multiple video acquisition devices; secondly, based on the target polling transmission mechanism, determine the video acquisition device that currently needs to perform video transmission among the multiple video acquisition devices to form at least one target video acquisition device; then, in the current time period, respectively pull the video stream formed by each target video acquisition device in the at least one target video acquisition device, so that the target video acquisition device transmits the video stream formed by the acquisition in the current time period to the electronic device. Based on the above content, since the pulling of the video stream is determined by polling the multiple video streams corresponding to the multiple video acquisition devices, it can have a higher video pulling reliability compared with the conventional technical solution of fixedly selecting part of the video streams for pulling. In addition, compared with the conventional technical solution of selecting all video streams for pulling, it can reduce the bandwidth consumption and computing resource consumption, so that the resource consumption problem of multi-channel video transmission is improved, that is, it can improve the problem of excessive resource consumption of multi-channel video transmission existing in the prior art. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, preferred embodiments are specifically cited below and described in detail with reference to the attached drawings.

[0051] Figure 1 A structural block diagram of an electronic device provided in an embodiment of the present application.

[0052] Figure 2 A flowchart of a multi-channel video transmission method based on a polling mechanism provided in an embodiment of the present application.

[0053] Figure 3 A schematic diagram of the feature update hierarchy provided in an embodiment of the present application.

[0054] Figure 4 A block diagram of a multi-channel video transmission device based on a polling mechanism provided in an embodiment of the present application. DETAILED DESCRIPTION

[0055] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. The components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various different configurations.

[0056] Therefore, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the present application for which protection is sought, but merely represents selected embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present application.

[0057] like Figure 1 As shown, an embodiment of the present application provides an electronic device, wherein the electronic device may include a memory, a processor, and a multi-channel video transmission device based on a polling mechanism.

[0058] In detail, the memory and the processor are electrically connected directly or indirectly to realize data transmission or interaction. For example, the memory and the processor can be electrically connected through one or more communication buses or signal lines. The multi-channel video transmission device based on the polling mechanism includes at least one software function module stored in the memory in the form of software or firmware. The processor is used to execute an executable computer program stored in the memory, for example, the software function module and computer program included in the multi-channel video transmission device based on the polling mechanism, so as to realize the multi-channel video transmission method based on the polling mechanism provided in the embodiment of the present application.

[0059] Optionally, the memory may be, but not limited to, a random access memory (RAM), a read only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), etc. Furthermore, the processor may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), a system on chip (SoC), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0060] Understandably, Figure 1 The structure shown is for illustration only, and the electronic device may also include Figure 1 More or fewer components as shown, or with Figure 1 The different configurations shown, for example, may also include a communication unit for exchanging information with other devices (such as a video acquisition device, etc.).

[0061] Combination Figure 2 The embodiment of the present application also provides a multi-channel video transmission method based on a polling mechanism that can be applied to the above-mentioned electronic device. The method steps defined in the process related to the multi-channel video transmission method based on a polling mechanism can be implemented by the electronic device; in addition, the electronic device can be communicatively connected to multiple video acquisition devices, each of which is used to perform video acquisition operations, such as video monitoring of a corresponding area.

[0062] The following will Figure 2 The specific process shown is explained in detail.

[0063] Step S110: determining the target polling transmission mechanism corresponding to the plurality of video acquisition devices.

[0064] In an embodiment of the present application, the electronic device can determine the target polling transmission mechanism corresponding to the multiple video acquisition devices. The target polling transmission mechanism is used to control the video transmission of some of the multiple video acquisition devices in turn, and the video acquisition devices polled twice are at least partially different, such as completely different or partially the same. For example, the video acquisition devices polled for the first time include video acquisition device 1, video acquisition device 2, and video acquisition device 3, and the video acquisition devices polled for the second time include video acquisition device 4 and video acquisition device 5; for another example, the video acquisition devices polled for the first time include video acquisition device 1, video acquisition device 2, and video acquisition device 3, and the video acquisition devices polled for the second time include video acquisition device 3, video acquisition device 4, and video acquisition device 5.

[0065] Step S120: Based on the target polling transmission mechanism, a video acquisition device that currently needs to perform video transmission is determined from the multiple video acquisition devices to form at least one target video acquisition device.

[0066] In an embodiment of the present application, after determining the target polling transmission mechanism, the electronic device can determine the video acquisition device that currently needs to perform video transmission among the multiple video acquisition devices based on the target polling transmission mechanism to form at least one target video acquisition device. As mentioned above, in the first polling, video acquisition device 1, video acquisition device 2, and video acquisition device 3 can be used as target video acquisition devices; in the second polling, video acquisition device 4 and video acquisition device 5 can be used as target video acquisition devices.

[0067] Step S130, performing a pull operation on the video stream collected by each of the at least one target video collection device in the current time period, so that the target video collection device transmits the video stream collected in the current time period to the electronic device.

[0068] In an embodiment of the present application, after determining the target video acquisition device, the electronic device can perform a pull operation on the video stream formed by each target video acquisition device in the at least one target video acquisition device in the current time period, so that the target video acquisition device transmits the video stream formed by the acquisition in the current time period to the electronic device. Among them, the specific time length of the current time period is not limited and can be selected according to actual needs, and the time length corresponding to each polled target video acquisition device can be the same or different. For example, the importance of the video streams corresponding to video acquisition devices 1, 2 and 3 is higher than the importance of the video streams corresponding to video acquisition devices 4 and 5. Therefore, the pulling time length of the video streams corresponding to video acquisition devices 1, 2 and 3 is greater than the pulling time length of the video streams corresponding to video acquisition devices 4 and 5, that is, the important video streams can obtain more video frames to avoid the loss of important information.

[0069] Based on the above content, since the pulling of video streams is determined by polling multiple video streams corresponding to multiple video acquisition devices, it can have higher video pulling reliability compared to the conventional technical solution of fixedly selecting part of the video streams for pulling. In addition, compared to the conventional technical solution of selecting all video streams for pulling, it can reduce bandwidth consumption and computing resource consumption, so that the resource consumption problem of multi-channel video transmission is improved, that is, the problem of excessive resource consumption of multi-channel video transmission in the prior art can be improved.

[0070] It should be noted that for step S110 , the specific method of determining the target polling transmission mechanism corresponding to the multiple video acquisition devices is not limited and can be selected according to actual needs.

[0071] For example, in an alternative implementation, in order to ensure the reliability of video stream pulling while reducing resource consumption, the above-mentioned step S110 may further include step S111 and step S112, and the specific content of each step is described as follows.

[0072] Step S111: based on the correlation between every two video acquisition devices among the multiple video acquisition devices, classify and combine the multiple video acquisition devices to form multiple device classification sets.

[0073] In an embodiment of the present application, the multiple video acquisition devices can be classified and combined based on the correlation between every two video acquisition devices in the multiple video acquisition devices to form multiple device classification sets. Each of the device classification sets includes at least one video acquisition device, and the video acquisition devices included between every two device classification sets are different. In other words, related video acquisition devices can be assigned to the same device classification set, and unrelated video acquisition devices can be assigned to different device classification sets.

[0074] Step S112: determining target polling transmission mechanisms corresponding to the multiple video acquisition devices based on the multiple device classification sets.

[0075] In an embodiment of the present application, a target polling transmission mechanism corresponding to the multiple video acquisition devices can be determined based on the multiple device classification sets. Wherein, when the video acquisition device that currently needs to perform video transmission is determined based on the target polling transmission mechanism, the at least one target video acquisition device currently determined includes each video acquisition device in a corresponding one of the device classification sets. For example, the video acquisition devices polled for the first time all belong to the first device classification set, the video acquisition devices polled for the second time all belong to the second device classification set, and the video acquisition devices polled for the third time all belong to the third device classification set. In this way, since the video acquisition devices in the same device classification set are related, when the video stream is pulled, it can be done synchronously, so that in some application scenarios, the pulled video stream can be subjected to correlation analysis, etc.

[0076] It is understandable that, in the above step S111, the specific method of classifying and combining the multiple video acquisition devices is not limited and can be selected according to actual needs. For example, in an alternative implementation, in order to ensure the reliability of the classification combination so that the multiple device classification sets formed can have a higher reliability, the above step S111 can further include step S111a, step S111b, step S111c, step S111d, step S111e and step S111f, and the specific content of each step is described as follows.

[0077] Step S111a: constructing a corresponding initial device relationship network based on the multiple video acquisition devices.

[0078] In the embodiment of the present application, a corresponding initial device relationship network can be constructed based on the multiple video acquisition devices, wherein the initial device relationship network includes the multiple video acquisition devices.

[0079] Step S111b, respectively obtaining device deployment location information, video acquisition area description information, and at least one historical video frame corresponding to each of the multiple video acquisition devices.

[0080] In an embodiment of the present application, the device deployment location information (such as coordinate information such as longitude, latitude and altitude) corresponding to each of the multiple video acquisition devices, the video acquisition area description information (i.e., used to introduce and illustrate the corresponding area, such as streets, highways, scenic spots, etc.) and at least one historical video frame pulled (such as the most recent one or more video frames pulled in history, or, it can also be screened and determined based on other mechanisms) can be obtained respectively.

[0081] Step S111c: Use the corresponding device deployment location information, video acquisition area description information and at least one historical video frame pulled as device attribute information of the video acquisition device.

[0082] In the embodiment of the present application, the corresponding device deployment location information, video acquisition area description information and at least one historical video frame pulled are used as the device attribute information of the video acquisition device. For example, the device deployment location information, video acquisition area description information and at least one historical video frame pulled corresponding to the video acquisition device 1 can be used as the device attribute information of the video acquisition device 1. In this way, the attributes of the video acquisition device can be characterized by information of multiple dimensions, thereby improving the richness of the attribute information and improving the reliability of subsequent correlation classification.

[0083] Step S111d: updating the initial device relationship network based at least on the correlation between the corresponding device deployment location information to form a candidate device relationship network corresponding to the initial device relationship network.

[0084] In an embodiment of the present application, the initial device relationship network can be updated based on at least the correlation between the corresponding device deployment location information to form a candidate device relationship network corresponding to the initial device relationship network. Among them, the correlation between the device deployment location information corresponding to two video acquisition devices with a connection relationship in the candidate device relationship network belongs to a predetermined first correlation. In other words, the two video acquisition devices with related device deployment location information can be connected, so that a candidate device relationship network can be formed. In addition, the device deployment location information correlation can mean that the location distance corresponding to the device deployment location information is less than a preset distance, and the preset distance can be selected according to actual needs, such as 50 meters, etc., and the preset distance can be different for different application scenarios, that is, in different application scenarios, the criteria for determining the correlation can be different.

[0085] Step S111e: based on the device attribute information of each of the video acquisition devices, the connection relationship in the candidate device relationship network is updated to form a target device relationship network.

[0086] In the embodiment of the present application, after the candidate device relationship network is formed, the connection relationship in the candidate device relationship network can be updated based on the device attribute information of each of the video acquisition devices to form a target device relationship network. In other words, since the connection relationship in the candidate device relationship network is determined only based on the device deployment location information, its reliability is relatively low. Therefore, the connection relationship can be updated based on richer device attribute information, thereby forming a target device relationship network with more reliable connection relationships.

[0087] Step S111f, based on the target device relationship network, classify and combine the multiple video acquisition devices to form multiple device classification sets.

[0088] In an embodiment of the present application, after the target device relationship network is formed, the multiple video acquisition devices can be classified and combined based on the target device relationship network to form multiple device classification sets. The specific classification combination is not restricted and can be selected according to actual needs. As described in the following text, when the target device relationship network is formed, the target device attribute characteristics corresponding to each of the video acquisition devices can be obtained. Based on this, the multiple video acquisition devices can be clustered based on the corresponding target device attribute characteristics (the specific clustering algorithm is not restricted). In this way, multiple corresponding clustering clusters are obtained. Then, each clustering cluster can be used as a device classification set to obtain multiple corresponding device classification sets.

[0089] It is understandable that, in the above step S111e, the specific method of updating the connection relationship in the candidate device relationship network is not limited and can be selected according to actual needs. For example, in an alternative implementation, in order to ensure the reliability of updating the connection relationship in the candidate device relationship network and form a target device relationship network, the above step S111e can further include step e1, step e2, step e3 and step e4, the specific contents of which are as follows.

[0090] Step e1, extracting features from device attribute information of each of the video acquisition devices respectively, and obtaining first device attribute features, second device attribute features and third device attribute features corresponding to each of the video acquisition devices.

[0091] In an embodiment of the present application, feature extraction can be performed on the device attribute information of each of the video acquisition devices to obtain the first device attribute feature, the second device attribute feature and the third device attribute feature corresponding to each of the video acquisition devices. The first device attribute feature is obtained by embedding the device deployment location information in the device attribute information (for example, it can be implemented by a corresponding word embedding model. For example, after the embedding feature corresponding to each attribute information word is obtained by the word embedding model, the mean of each embedded feature can be calculated to obtain the first device attribute feature), the second device attribute feature is obtained by embedding the video acquisition area description information in the device attribute information, and the third device attribute feature is obtained by convolving the historical video frames in the device attribute information (for example, it can be implemented by a corresponding convolutional neural network model. In addition, when there are multiple historical video frames, the corresponding multiple convolution features can be averaged to obtain the third device attribute feature).

[0092] Step e2, in the multi-level deep feature mining process, alternately aggregate the first device attribute feature and the second device attribute feature into the third device attribute feature to form the deep device attribute feature corresponding to the video acquisition device.

[0093] In an embodiment of the present application, after obtaining three device attribute features, the first device attribute features and the second device attribute features can be alternately aggregated into the third device attribute features in a multi-level deep feature mining process to form a deep device attribute feature corresponding to the video acquisition device. Exemplarily, in the first level of deep feature mining, the third device attribute feature can be cross-attention processed based on the first device attribute feature to obtain the corresponding cross-attention feature, and then the cross-attention feature and the third device attribute feature can be averaged to obtain the output feature of the first level; in the second level of deep feature mining, the output feature of the first level can be cross-attention processed based on the second device attribute feature to obtain the corresponding cross-attention feature, and then the cross-attention feature and the output feature of the first level can be averaged to obtain the output feature of the second level; in the third level of deep feature mining, the output feature of the second level can be cross-attention processed based on the first device attribute feature to obtain the corresponding cross-attention feature, and then the cross-attention feature and the output feature of the second level can be averaged to obtain the output feature of the third level, and in this way, the aggregation is alternately performed in sequence to obtain the output feature of the last level, and then the output feature of the last level is used as the deep device attribute feature corresponding to the video acquisition device. Based on this, the deep device attribute feature corresponding to each video acquisition device can be obtained.

[0094] Step e3: based on the connection relationship in the candidate device relationship network, the deep device attribute characteristics corresponding to each of the video acquisition devices are updated to form the target device attribute characteristics corresponding to each of the video acquisition devices.

[0095] In the embodiment of the present application, after obtaining the corresponding deep device attribute features, the deep device attribute features corresponding to each of the video acquisition devices can be updated based on the connection relationship in the candidate device relationship network to form the target device attribute features corresponding to each video acquisition device. In other words, the corresponding deep device attribute features can be updated based on the deep device attribute features corresponding to the video acquisition devices with connection relationships, so that the target device attribute features with better semantic representation capabilities can be obtained.

[0096] Step e4: based on the similarity between the target device attribute features corresponding to every two of the video acquisition devices, the connection relationship in the candidate device relationship network is updated to form a target device relationship network.

[0097] In an embodiment of the present application, after obtaining the corresponding target device attribute characteristics, the connection relationship in the candidate device relationship network can be updated based on the similarity between the target device attribute characteristics corresponding to each two of the video acquisition devices to form a target device relationship network. Exemplarily, if the similarity (such as cosine similarity, etc.) between the target device attribute characteristics corresponding to two video acquisition devices is greater than a preset similarity (such as 0.5, 0.6, 0.7, etc.), the two video acquisition devices can be connected to form a connection relationship. In addition, the original connection relationship in the candidate device relationship network can be retained or discarded, and can be configured according to actual needs. In addition, as described in the previous text, the obtained target device attribute characteristics can also be used for classification in step S111f.

[0098] It can be understood that in the above step e3, the specific method of updating the depth device attribute features corresponding to each of the video acquisition devices is not limited and can be selected according to actual needs. For example, in an alternative implementation, in order to ensure the update reliability of the depth device attribute features and make the semantic representation ability of the formed target device attribute features better, the above step e3 can further include the following contents:

[0099] First, a video acquisition device is randomly determined in the candidate device relationship network as a central video acquisition device corresponding to the candidate device relationship network;

[0100] Secondly, determining a connection relationship characterization parameter between each video acquisition device in the candidate device relationship network and the central video acquisition device, wherein the connection relationship characterization parameter is positively correlated to the length of the shortest traversal path between the video acquisition device and the central video acquisition device, that is, the smaller the length of the shortest traversal path, the smaller the corresponding connection relationship characterization parameter;

[0101] Then, the feature update level corresponding to each of the video acquisition devices can be determined based on the connection relationship representation parameter, wherein the connection relationship representation parameter is positively correlated with the feature update level, that is, the smaller the connection relationship representation parameter, the lower the corresponding feature update level. Figure 3 ;

[0102] Furthermore, for two video acquisition devices with adjacent feature update levels and a connection relationship, the updated depth device attribute features of the video acquisition device with a lower feature update level among the two video acquisition devices are aggregated into the current depth device attribute features of the video acquisition device with a higher feature update level, so as to obtain the updated depth device attribute features of the video acquisition device with a higher feature update level, wherein the current depth device attribute features of the video acquisition device with the lowest feature update level are directly used as the updated depth device attribute features. For example, the current depth device attribute features of the central video acquisition device (corresponding to the feature update level of 0) can be directly used as the updated depth device attribute features, and then the updated depth device attribute features of the central video acquisition device can be aggregated into the current depth device attribute features of the video acquisition device with the feature update level of 1. In the feature, the updated depth device attribute feature of the video acquisition device with a feature update level of 1 is obtained, wherein the feature aggregation may refer to performing cross-attention processing on the current depth device attribute feature of a video acquisition device with a higher feature update level based on the updated depth device attribute feature of a video acquisition device with a lower feature update level, and performing mean calculation on the feature obtained by the cross-attention processing and the current depth device attribute feature of a video acquisition device with a higher feature update level to obtain the updated depth device attribute feature; in addition, in some embodiments, a video acquisition device with a higher feature update level may have a connection relationship with multiple video acquisition devices with a lower feature update level, so that after performing cross-attention processing and mean calculation respectively, each calculation result may be averaged to obtain the updated depth device attribute feature;

[0103] Finally, after obtaining the updated depth device attribute feature of a video acquisition device with the highest feature update level, use the updated depth device attribute feature of each video acquisition device as the new current depth device attribute feature, and rotate to execute the step of randomly determining a video acquisition device in the candidate device relationship network as the central video acquisition device corresponding to the candidate device relationship network, wherein the central video acquisition device determined every two times this step is executed is different, and after the depth device attribute feature of each video acquisition device in the candidate device relationship network is updated at least once (i.e., to ensure that the depth device attribute feature of each video acquisition device is updated), stop rotating to execute this step, and use the current depth device attribute feature of each video acquisition device (i.e., the depth device attribute feature after the last update) as the corresponding target device attribute feature.

[0104] It can be understood that, in the above-mentioned step S112, the specific method of determining the target polling transmission mechanism corresponding to the multiple video acquisition devices is not limited and can be selected according to actual needs. For example, in an alternative implementation, in order to ensure that the reliability of the determined target polling transmission mechanism is relatively high, the above-mentioned step S112 can further include step S112a, step S112b and step S112c, and the specific content of each step is described as follows.

[0105] Step S112a: determining the correlation between every two device classification sets in the multiple device classification sets based on the correlation between every two video acquisition devices in the multiple video acquisition devices.

[0106] In an embodiment of the present application, the correlation between every two device classification sets in the multiple device classification sets can be determined based on the correlation between every two video acquisition devices in the multiple video acquisition devices. Exemplarily, the multiple device classification sets can be combined in pairs to obtain multiple collection groups, and then, for each collection group, the two device classification sets included in the collection group can be combined in pairs with the video acquisition devices between the sets to obtain the corresponding multiple video acquisition device groups, and then the similarity between the two video acquisition devices included in each video acquisition device group is calculated, and finally, the mean of each similarity is calculated to obtain the similarity between the two device classification sets included in the collection group. Among them, the similarity between video acquisition devices can refer to the similarity between the attribute characteristics of the target device or the similarity between the corresponding device deployment location information, or the similarity can also be calculated based on other information.

[0107] Step S112b: based on the correlation between every two device classification sets in the multiple device classification sets, serialize the multiple device classification sets to form a corresponding classification set sequence.

[0108] In an embodiment of the present application, after the correlation between device classification sets is obtained, the multiple device classification sets can be serialized based on the correlation between every two device classification sets in the multiple device classification sets to form a corresponding classification set sequence. Wherein, the classification set sequence belongs to a circular sequence, and in the classification set sequence, the correlation between every two adjacent device classification sets belongs to a predetermined second correlation. Exemplarily, multiple device classification sets can be randomly sorted to form multiple candidate sequences, and then, for each candidate sequence, the mean of the similarities between every two adjacent device classification sets in the candidate sequence is calculated to obtain the mean similarity corresponding to the candidate sequence, and finally, among the multiple candidate sequences, a candidate sequence with the largest mean similarity is determined as the corresponding classification set sequence.

[0109] Step S112c: determining the target polling transmission mechanism corresponding to the plurality of video acquisition devices based on the classification set sequence.

[0110] In an embodiment of the present application, after obtaining the classification set sequence, the target polling transmission mechanism corresponding to the multiple video acquisition devices can be determined based on the classification set sequence. Wherein, when the target polling transmission mechanism is used for polling, the order of polling is determined based on the order of the device classification sets in the classification set sequence, such as the order of polling is consistent with the order of the device classification sets in the classification set sequence.

[0111] It can be understood that, in the above-mentioned step S112c, the specific method of determining the target polling transmission mechanism corresponding to the multiple video acquisition devices is not limited and can be selected according to actual needs. For example, in an alternative implementation, in order to further improve the reliability of the target polling transmission mechanism, the above-mentioned step S112c can further include step c1, step c2 and step c3, and the specific content of each step is described as follows.

[0112] Step c1, determining candidate polling transmission mechanisms corresponding to the plurality of video acquisition devices based on the order relationship between the device classification sets in the classification set sequence.

[0113] In the embodiment of the present application, the candidate polling transmission mechanisms corresponding to the plurality of video acquisition devices may be determined based on the order relationship between the device classification sets in the classification set sequence, wherein the candidate polling transmission mechanisms have a polling order relationship between the video acquisition devices.

[0114] Step c2, for each device classification set in the multiple device classification sets, analyze each video acquisition device in the device classification set, determine the video importance parameter corresponding to the device classification set, and determine the polling duration corresponding to the device classification set based on the video importance parameter.

[0115] In an embodiment of the present application, for each device classification set in the multiple device classification sets, each video acquisition device in the device classification set can be analyzed to determine the video importance parameter corresponding to the device classification set, and the polling duration corresponding to the device classification set can be determined based on the video importance parameter. The video importance parameter is positively correlated with the polling duration. In other words, the higher the video importance parameter, the higher the polling duration, and the more video frames are acquired in one polling.

[0116] Step c3: merging the polling duration corresponding to each of the device classification sets into the candidate polling transmission mechanism to form a target polling transmission mechanism corresponding to the multiple video acquisition devices.

[0117] In an embodiment of the present application, after the polling duration, the polling duration corresponding to each of the device classification sets can be merged into the candidate polling transmission mechanism to form a target polling transmission mechanism corresponding to the multiple video acquisition devices. The polling duration is used to characterize the time length of the pulled video stream. In other words, multiple video acquisition devices can be polled according to the order relationship and polling duration in the target polling transmission mechanism.

[0118] It is understandable that, in the above step c2, the specific method of determining the polling duration corresponding to the device classification set is not limited and can be selected according to actual needs. For example, in an alternative implementation, in order to ensure that the determined polling duration has a high reliability, the above step c2 may further include the following contents:

[0119] First, for each device classification set in the multiple device classification sets, a video stream analysis result of at least one historical video stream corresponding to each video acquisition device in the device classification set is obtained respectively, wherein the electronic device is further used to analyze the pulled video stream to obtain a corresponding video stream analysis result, and the video stream analysis result is at least used to characterize whether there is an abnormality (or, it can also be used to characterize the degree of the abnormality, etc.), and the specific analysis method is not limited and can be determined according to actual monitoring needs. For example, the video frame can be analyzed based on the corresponding artificial intelligence technology to determine whether there is any violation, etc.;

[0120] Secondly, a statistical analysis may be performed on the video stream analysis results of at least one historical video stream corresponding to each video acquisition device in the device classification set (such as the video stream analysis results of a historical video stream recently pulled) to obtain a corresponding abnormal distribution parameter, wherein the abnormal distribution parameter is at least used to reflect the number or proportion of abnormal video stream analysis results;

[0121] Finally, the video importance parameters corresponding to the device classification set can be determined based on the abnormal distribution parameters, and the polling duration corresponding to the device classification set can be determined based on the video importance parameters, wherein the quantity or proportion reflected by the abnormal distribution parameters and the video importance parameters corresponding to the device classification set have a positive correlation, that is, the larger the quantity or proportion reflected by the abnormal distribution parameters, the higher the video importance parameter.

[0122] Combination Figure 4 The embodiment of the present application also provides a multi-channel video transmission device based on a polling mechanism that can be applied to the above electronic device. The multi-channel video transmission device based on a polling mechanism can include a transmission mechanism determination module, a collection device determination module and a video stream pulling module.

[0123] In detail, the transmission mechanism determination module can be used to determine the target polling transmission mechanism corresponding to the multiple video acquisition devices, wherein the target polling transmission mechanism is used to sequentially control video transmission of some of the multiple video acquisition devices, and the video acquisition devices polled twice are at least partially different. Figure 2 As shown in step S110, the relevant contents of the transmission mechanism determination module can refer to the above description of step S110.

[0124] In detail, the acquisition device determination module can be used to determine the video acquisition device that currently needs to perform video transmission among the multiple video acquisition devices based on the target polling transmission mechanism to form at least one target video acquisition device. Figure 2 As shown in step S120, the relevant contents of the acquisition device determination module can refer to the above description of step S120.

[0125] In detail, the video stream pulling module can be used to pull the video stream collected by each of the at least one target video acquisition device in the current time period, so that the target video acquisition device transmits the video stream collected in the current time period to the electronic device. Figure 2 As shown in step S130, the relevant contents of the video stream pulling module can refer to the above description of step S130.

[0126] In an embodiment of the present application, corresponding to the above-mentioned multi-channel video transmission method based on the polling mechanism applied to the electronic device, a computer-readable storage medium is also provided, in which a computer program is stored, and when the computer program is run, each step of the multi-channel video transmission method based on the polling mechanism is executed.

[0127] The steps executed when the aforementioned computer program is running will not be described one by one here, and reference may be made to the explanation of the aforementioned multi-channel video transmission method based on the polling mechanism in the previous text.

[0128] In summary, the multi-channel video transmission method, device and equipment based on the polling mechanism provided by the present application firstly determine the target polling transmission mechanism corresponding to multiple video acquisition devices; secondly, based on the target polling transmission mechanism, determine the video acquisition device that currently needs to perform video transmission among the multiple video acquisition devices to form at least one target video acquisition device; then, in the current time period, respectively pull the video stream formed by each target video acquisition device in the at least one target video acquisition device, so that the target video acquisition device transmits the video stream formed by the acquisition in the current time period to the electronic device. Based on the above content, since the pulling of the video stream is determined by polling the multiple video streams corresponding to the multiple video acquisition devices, it can have a higher video pulling reliability compared with the conventional technical solution of fixedly selecting part of the video stream for pulling. In addition, compared with the conventional technical solution of selecting all video streams for pulling, it can reduce the bandwidth consumption and computing resource consumption, so that the resource consumption problem of multi-channel video transmission is improved, that is, it can improve the problem of excessive resource consumption of multi-channel video transmission existing in the prior art.

[0129] In several embodiments provided in the embodiments of the present application, it should be understood that the disclosed device and method can also be implemented in other ways. The device and method embodiments described above are merely schematic, for example, the flowcharts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the device, method and computer program product according to multiple embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of a code, and the module, a program segment or a part of a code contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0130] In addition, the functional modules in the various embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.

[0131] If the function is implemented in the form of a software function module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present application is essentially 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, which is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, an electronic device, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a disk or an optical disk. It should be noted that, in this article, the term "include", "include" or any other variant thereof is intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements includes not only those elements, but also includes other elements that are not explicitly listed, or also includes elements inherent to such a process, method, article or device. Without more constraints, an element defined by the phrase "comprising a..." does not exclude the existence of other identical elements in the process, method, article or apparatus comprising the element.

[0132] The above description is only the preferred embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A multi-channel video transmission method based on a polling mechanism, characterized in that: Applied to an electronic device, the electronic device is communicatively connected to a plurality of video acquisition devices, each of the video acquisition devices is used to perform a video acquisition operation, and the multi-channel video transmission method based on the polling mechanism includes: Determine a target polling transmission mechanism corresponding to the multiple video acquisition devices, wherein the target polling transmission mechanism is used to sequentially control video transmission of some of the multiple video acquisition devices, and the video acquisition devices polled twice adjacently are at least partially different; Based on the target polling transmission mechanism, determining a video acquisition device that currently needs to perform video transmission among the multiple video acquisition devices to form at least one target video acquisition device; In the current time period, a pulling operation is performed on the video stream collected by each of the at least one target video collection device, so that the target video collection device transmits the video stream collected in the current time period to the electronic device.

2. The multi-channel video transmission method based on polling mechanism according to claim 1, characterized in that: The step of determining the target polling transmission mechanism corresponding to the multiple video acquisition devices includes: Based on the correlation between every two video acquisition devices among the multiple video acquisition devices, the multiple video acquisition devices are classified and combined to form multiple device classification sets, wherein each of the device classification sets includes at least one video acquisition device, and the video acquisition devices included in every two device classification sets are different; Based on the multiple device classification sets, a target polling transmission mechanism corresponding to the multiple video acquisition devices is determined, wherein when the video acquisition device that currently needs to perform video transmission is determined based on the target polling transmission mechanism, at least one target video acquisition device currently determined includes each video acquisition device in a corresponding one of the device classification sets.

3. The multi-channel video transmission method based on polling mechanism according to claim 2 is characterized in that: The step of classifying and combining the multiple video acquisition devices based on the correlation between every two video acquisition devices in the multiple video acquisition devices to form multiple device classification sets includes: Building a corresponding initial device relationship network based on the multiple video acquisition devices, wherein the initial device relationship network includes the multiple video acquisition devices; Respectively obtain device deployment location information, video acquisition area description information, and at least one historical video frame corresponding to each of the multiple video acquisition devices; The corresponding device deployment location information, video acquisition area description information and at least one historical video frame pulled out are used as device attribute information of the video acquisition device; At least based on the correlation between the corresponding device deployment location information, the initial device relationship network is updated to form a candidate device relationship network corresponding to the initial device relationship network, wherein the correlation between the device deployment location information corresponding to two video acquisition devices having a connection relationship in the candidate device relationship network belongs to a predetermined first correlation relationship; Based on the device attribute information of each of the video acquisition devices, the connection relationship in the candidate device relationship network is updated to form a target device relationship network; Based on the target device relationship network, the multiple video acquisition devices are classified and combined to form multiple device classification sets.

4. The multi-channel video transmission method based on polling mechanism according to claim 3 is characterized in that: The step of updating the connection relationship in the candidate device relationship network based on the device attribute information of each of the video acquisition devices to form a target device relationship network includes: Extracting features from the device attribute information of each of the video acquisition devices respectively, and obtaining a first device attribute feature, a second device attribute feature, and a third device attribute feature corresponding to each of the video acquisition devices, wherein the first device attribute feature is obtained by embedding the device deployment location information in the device attribute information, the second device attribute feature is obtained by embedding the video acquisition area description information in the device attribute information, and the third device attribute feature is obtained by convolving the historical video frames in the device attribute information; In the multi-level deep feature mining process, the first device attribute feature and the second device attribute feature are alternately aggregated into the third device attribute feature to form a deep device attribute feature corresponding to the video acquisition device; Based on the connection relationship in the candidate device relationship network, the deep device attribute feature corresponding to each of the video acquisition devices is updated to form the target device attribute feature corresponding to each of the video acquisition devices; Based on the similarity between the target device attribute features corresponding to every two of the video acquisition devices, the connection relationship in the candidate device relationship network is updated to form a target device relationship network.

5. The multi-channel video transmission method based on polling mechanism according to claim 4 is characterized in that: The step of updating the depth device attribute features corresponding to each of the video acquisition devices based on the connection relationship in the candidate device relationship network to form the target device attribute features corresponding to each of the video acquisition devices includes: Randomly determine a video acquisition device in the candidate device relationship network as a central video acquisition device corresponding to the candidate device relationship network; Determine a connection relationship characterization parameter between each video acquisition device in the candidate device relationship network and the central video acquisition device, wherein the connection relationship characterization parameter is positively correlated to the length of the shortest traversal path between the video acquisition device and the central video acquisition device; Determining a feature update level corresponding to each of the video acquisition devices based on the connection relationship representation parameter, wherein the connection relationship representation parameter is positively correlated with the feature update level; For two video acquisition devices that are adjacent in feature update level and have a connection relationship, the updated depth device attribute features of the video acquisition device with a lower feature update level among the two video acquisition devices are aggregated into the current depth device attribute features of the video acquisition device with a higher feature update level, so as to obtain the updated depth device attribute features of the video acquisition device with a higher feature update level, wherein the current depth device attribute features of the video acquisition device with the lowest feature update level are directly used as the updated depth device attribute features; After obtaining the updated depth device attribute feature of a video acquisition device with the highest feature update level, use the updated depth device attribute feature of each video acquisition device as the new current depth device attribute feature, and rotate to execute the step of randomly determining a video acquisition device in the candidate device relationship network as the central video acquisition device corresponding to the candidate device relationship network, wherein the central video acquisition device determined after executing this step twice is different, and after the depth device attribute feature of each video acquisition device in the candidate device relationship network is updated at least once, stop rotating to execute this step, and use the current depth device attribute feature of each video acquisition device as the corresponding target device attribute feature.

6. The multi-channel video transmission method based on polling mechanism according to claim 2 is characterized in that: The step of determining the target polling transmission mechanism corresponding to the multiple video acquisition devices based on the multiple device classification sets includes: Based on the correlation between every two video acquisition devices in the multiple video acquisition devices, determining the correlation between every two device classification sets in the multiple device classification sets; Based on the correlation between every two device classification sets in the multiple device classification sets, the multiple device classification sets are serialized to form a corresponding classification set sequence, wherein the classification set sequence belongs to a ring sequence, and in the classification set sequence, the correlation between every two adjacent device classification sets belongs to a predetermined second correlation; Based on the classification set sequence, a target polling transmission mechanism corresponding to the multiple video acquisition devices is determined, wherein when the target polling transmission mechanism performs polling, the order of polling is determined based on the order of the device classification sets in the classification set sequence.

7. The multi-channel video transmission method based on polling mechanism according to claim 6 is characterized in that: The step of determining the target polling transmission mechanism corresponding to the plurality of video acquisition devices based on the classification set sequence comprises: Based on the order relationship between the device classification sets in the classification set sequence, determine the candidate polling transmission mechanisms corresponding to the multiple video acquisition devices, wherein the candidate polling transmission mechanisms have a polling order relationship between the video acquisition devices; For each device classification set in the multiple device classification sets, each video acquisition device in the device classification set is analyzed to determine a video importance parameter corresponding to the device classification set, and based on the video importance parameter, a polling duration corresponding to the device classification set is determined, wherein the video importance parameter is positively correlated with the polling duration; The polling duration corresponding to each of the device classification sets is merged into the candidate polling transmission mechanism to form a target polling transmission mechanism corresponding to the multiple video acquisition devices, wherein the polling duration is used to characterize the time length of the pulled video stream.

8. The multi-channel video transmission method based on polling mechanism according to claim 7, characterized in that: The steps of analyzing each video acquisition device in each device classification set in the multiple device classification sets, determining a video importance parameter corresponding to the device classification set, and determining a polling duration corresponding to the device classification set based on the video importance parameter, include: For each device classification set in the multiple device classification sets, respectively obtain a video stream analysis result of at least one historical video stream corresponding to each video acquisition device in the device classification set, wherein the electronic device is further used to analyze the pulled video stream to obtain a corresponding video stream analysis result, and the video stream analysis result is at least used to characterize whether there is an abnormality; Performing statistical analysis on the video stream analysis results of at least one historical video stream corresponding to each video acquisition device in the device classification set to obtain corresponding abnormal distribution parameters, wherein the abnormal distribution parameters are at least used to reflect the number or proportion of abnormal video stream analysis results; Based on the abnormal distribution parameters, the video importance parameters corresponding to the device classification set are determined, and based on the video importance parameters, the polling duration corresponding to the device classification set is determined, wherein there is a positive correlation between the quantity or proportion reflected by the abnormal distribution parameters and the video importance parameters corresponding to the device classification set.

9. A multi-channel video transmission device based on a polling mechanism, characterized in that: Applied to an electronic device, the electronic device is communicatively connected to a plurality of video acquisition devices, each of the video acquisition devices is used to perform a video acquisition operation, and the multi-channel video transmission device based on the polling mechanism includes: A transmission mechanism determination module, used to determine a target polling transmission mechanism corresponding to the plurality of video acquisition devices, wherein the target polling transmission mechanism is used to sequentially control video transmission of some of the plurality of video acquisition devices, and the video acquisition devices polled twice adjacently are at least partially different; A collection device determination module, used to determine a video collection device that currently needs to perform video transmission among the multiple video collection devices based on the target polling transmission mechanism, so as to form at least one target video collection device; The video stream pulling module is used to pull the video stream collected by each target video acquisition device in the at least one target video acquisition device within the current time period, so that the target video acquisition device transmits the video stream collected within the current time period to the electronic device.

10. An electronic device, characterized in that: include: Memory for storing computer programs; A processor connected to the memory is used to execute the computer program stored in the memory to implement the multi-channel video transmission method based on the polling mechanism as described in any one of claims 1-8.

Images