Video service perception method, device, video transmission equipment and receiving equipment

By segmenting stable segments with poor quality at the video source end and calculating their QoE evaluation values, and combining complex and simple models, the problem of insufficient accuracy and complexity in existing video service perception methods is solved, and efficient video quality assessment is achieved.

CN114125496BActive Publication Date: 2024-11-08CHINA MOBILE COMM LTD RES INST +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202010905409.3
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2020-09-01
Publication Date
2024-11-08
Estimated Expiration
2040-09-01

AI Technical Summary

Technical Problem

Existing video service perception methods are insufficient in terms of accuracy and computational complexity, making it difficult to simultaneously meet the requirements of high accuracy and low complexity. In particular, artificial intelligence training methods are greatly affected by the amount of training samples, while analytical methods are greatly affected by factors other than input parameters.

Method used

By analyzing video quality parameters at the video source, segments with stable poor quality are segmented, and their Quality of Experience (QoE) evaluation values ​​under different playback parameters are calculated. High-precision business resolution coefficients are obtained using a complex model, and then a simple model is used for fitting, simplifying the fitting process of video business perception parameters.

Benefits of technology

This approach simplifies the fitting process for video service perception parameters while ensuring real-time accuracy, reduces the impact of video content on the accuracy of the simplified algorithm, and improves the efficiency of video quality assessment.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114125496B_ABST
    Figure CN114125496B_ABST
Patent Text Reader

Abstract

The present invention provides a video service perception method, device, video transmission equipment and receiving equipment. The method comprises: analyzing the quality parameters of the target video source to obtain a plurality of stable segments with poor quality and stable quality parameters in the target video source; calculating the experience quality QoE evaluation value of each stable segment with poor quality under different playback parameters; fitting the QoE evaluation value of each stable segment with poor quality under different playback parameters with the preset analysis parameters for video service perception to obtain the service analysis coefficient for video service perception of each stable segment with poor quality; transmitting the target video source, wherein the video segment corresponding to each stable segment with poor quality in the target video source includes the service analysis coefficient. The method can simplify the fitting process of video service perception parameters, while ensuring the real-time accuracy of the perception algorithm and reducing the influence of video content on the accuracy of the simplified algorithm.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of data services, and in particular to a video service perception method, device, video transmission equipment and receiving equipment. Background Art

[0002] At present, the methods of video service perception include analytical methods and artificial intelligence training methods. Among them, the analytical method selects appropriate coefficients according to input parameters and calculates the perceived quality of experience (QoE) according to a given formula. The model of this method is relatively fixed and the calculation efficiency is high, but the accuracy is greatly affected by factors other than the input parameters. The artificial intelligence training method trains the model based on the manual annotation of QoE combined with the input parameters in the hope of achieving a model that better fits the manual judgment. However, with this method, the model accuracy is greatly affected by the number of training samples, and the calculation complexity is affected by the complexity of the model, which requires more computing resources.

[0003] Of the two perception methods mentioned above, the accuracy and efficiency of the artificial intelligence training method depend on the size of the data set and the complexity of the algorithm, and it is difficult to achieve perception of large-scale concurrent services; the analysis method is affected by factors other than the input parameters, making it difficult to achieve high accuracy; therefore, it is necessary to improve the method of video service perception in order to simultaneously achieve the goals of improving accuracy and reducing computational complexity. Summary of the invention

[0004] The technical solution of the present invention aims to provide a video service perception method, device, video transmission equipment and receiving equipment, which are used to simplify the fitting process of video service perception parameters, while ensuring the real-time accuracy of the perception algorithm and reducing the influence of video content on the accuracy of the simplified algorithm.

[0005] An embodiment of the present invention provides a video service perception method, which includes:

[0006] Analyze the quality parameters of the target video source to obtain a plurality of stable quality-poor segments of the target video source with stable quality parameters;

[0007] Calculate the quality of experience (QoE) evaluation value of each of the stable quality difference segments under different playback parameters;

[0008] Fitting the QoE evaluation value of each stable segment with poor quality under different playback parameters with the preset analysis parameters for video service perception to obtain a service analysis coefficient for video service perception for each stable segment with poor quality;

[0009] The target video source is transmitted, wherein the video segment corresponding to each of the poor quality stable segments in the target video source includes the business resolution coefficient.

[0010] Optionally, in the video service perception method, the step of analyzing the quality parameters of the target video source to obtain a plurality of stable quality-poor segments with stable quality parameters in the target video source comprises:

[0011] Calculate the QoE evaluation value of each time period by a model training method according to the quality parameter of the target video source in each time period;

[0012] According to the QoE evaluation value of each time period, the transition segments obtained after the target video source is segmented for transition are analyzed to obtain a plurality of quality difference stable segments.

[0013] Optionally, the video service perception method, wherein, according to the quality parameter of the target video source in each time period, the QoE evaluation value of each time period is calculated by a model training method, comprises:

[0014] Classifying the scene corresponding to each quality parameter in the target video source;

[0015] The QoE evaluation value of the quality parameter of the target video source in each of the time periods in the corresponding different grading combinations is calculated by a model training method.

[0016] Optionally, the video service perception method, wherein, according to the QoE evaluation value of each time period, the transition segments obtained after the target video source is segmented for transition, to obtain multiple quality difference stable segments, includes:

[0017] When the duration of the transition segment is greater than the duration period, determining whether the QoE evaluation value obtained corresponding to each duration period in the transition segment meets a preset stability condition;

[0018] Determine the video segment corresponding to the time period that meets the preset stability condition as the poor-quality stable segment;

[0019] When the duration of the transition segment is less than or equal to the duration period, and the corresponding QoE evaluation value obtained within the duration period corresponding to the transition segment meets the preset stability condition, the transition segment is determined to be the poor quality stable segment.

[0020] Optionally, in the video service perception method, the step of calculating a quality of experience (QoE) evaluation value of each of the poor quality stable segments under different playback parameters includes:

[0021] Selecting a target stable quality segment from at least one stable quality segment in each transition segment; the difference between the QoE evaluation value of the target stable quality segment and the average value of the QoE evaluation value of the transition segment is the smallest;

[0022] The quality of experience (QoE) evaluation value of each target quality difference stable segment under different playback parameters is calculated.

[0023] Optionally, the video service perception method further comprises:

[0024] When transmitting the target video source, the starting time point, the ending time point and / or the segment duration of each transition segment are marked.

[0025] Optionally, the video service perception method further comprises:

[0026] When transmitting the target video source, the poor quality stable segment is identified.

[0027] The embodiment of the present invention further provides a video service perception method, which includes:

[0028] Acquire a target video source; wherein the video segment corresponding to each of the poor quality stable segments in the target video source includes the business resolution coefficient;

[0029] According to the service analysis coefficient, the service quality of each of the poor-quality stable segments of the target video source is analyzed to obtain a service perception evaluation result of the target video source.

[0030] Optionally, the video service perception method, wherein, according to the service analysis coefficient, performing analysis on the service quality of each of the poor quality stable segments of the target video source, comprises:

[0031] The business analysis coefficient corresponding to the poor quality stable segment is used to replace the calculation coefficient in the business quality simplification algorithm to obtain the business perception evaluation result of the target video source.

[0032] An embodiment of the present invention further provides a video transmission device, including a processor, wherein the processor is used to:

[0033] Analyze the quality parameters of the target video source to obtain a plurality of stable quality-poor segments of the target video source with stable quality parameters;

[0034] Calculate the quality of experience (QoE) evaluation value of each of the stable quality difference segments under different playback parameters;

[0035] Fitting the QoE evaluation value of each stable segment with poor quality under different playback parameters with the preset analysis parameters for video service perception to obtain a service analysis coefficient for video service perception for each stable segment with poor quality;

[0036] The target video source is transmitted, wherein the video segment corresponding to each of the poor quality stable segments in the target video source includes the business resolution coefficient.

[0037] Optionally, in the video transmission device, the processor analyzes the quality parameters of the target video source to obtain a plurality of stable quality-poor segments with stable quality parameters in the target video source, including:

[0038] Calculate the QoE evaluation value of each time period by a model training method according to the quality parameter of the target video source in each time period;

[0039] According to the QoE evaluation value of each time period, the transition segments obtained after the target video source is segmented for transition are analyzed to obtain a plurality of quality difference stable segments.

[0040] Optionally, in the video transmission device, the processor calculates the QoE evaluation value of each time period by a model training method according to the quality parameter of the target video source in each time period, including:

[0041] Classifying the scene corresponding to each quality parameter in the target video source;

[0042] The QoE evaluation value of the quality parameter of the target video source in each of the time periods in the corresponding different grading combinations is calculated by a model training method.

[0043] Optionally, in the video transmission device, the processor analyzes the transition segments obtained after the target video source is segmented according to the QoE evaluation value of each time period to obtain a plurality of quality difference stable segments, including:

[0044] When the duration of the transition segment is greater than the duration period, determining whether the QoE evaluation value obtained corresponding to each duration period in the transition segment meets a preset stability condition;

[0045] Determine the video segment corresponding to the time period that meets the preset stability condition as the poor-quality stable segment;

[0046] When the duration of the transition segment is less than or equal to the duration period, and the corresponding QoE evaluation value obtained within the duration period corresponding to the transition segment meets the preset stability condition, the transition segment is determined to be the poor quality stable segment.

[0047] Optionally, in the video transmission device, the processor calculates the quality of experience (QoE) evaluation value of each of the poor quality stable segments under different playback parameters, including:

[0048] Selecting a target stable quality segment from at least one stable quality segment in each transition segment; the difference between the QoE evaluation value of the target stable quality segment and the average value of the QoE evaluation value of the transition segment is the smallest;

[0049] The quality of experience (QoE) evaluation value of each target quality difference stable segment under different playback parameters is calculated.

[0050] Optionally, in the video transmission device, the processor is further configured to:

[0051] When transmitting the target video source, the starting time point, the ending time point and / or the segment duration of each transition segment are marked.

[0052] Optionally, in the video transmission device, the processor is further configured to:

[0053] When transmitting the target video source, the poor quality stable segment is identified.

[0054] An embodiment of the present invention further provides a video receiving device, including a processor, wherein the processor is used to:

[0055] Acquire a target video source; wherein the video segment corresponding to each of the poor quality stable segments in the target video source includes the business resolution coefficient;

[0056] According to the service analysis coefficient, the service quality of each of the poor-quality stable segments of the target video source is analyzed to obtain a service perception evaluation result of the target video source.

[0057] Optionally, in the video transmission device, the processor performs analytical analysis on the service quality of each poor quality stable segment of the target video source according to the service analysis coefficient, including:

[0058] The business analysis coefficient corresponding to the poor quality stable segment is used to replace the calculation coefficient in the business quality simplification algorithm to obtain the business perception evaluation result of the target video source.

[0059] The embodiment of the present invention further provides a video service perception device, which includes:

[0060] An analysis module, configured to analyze the quality parameters of the target video source and obtain a plurality of stable quality-poor segments with stable quality parameters in the target video source;

[0061] A calculation module, used to calculate the experience quality QoE evaluation value of each of the poor quality stable segments under different playback parameters;

[0062] A fitting module, used for fitting the QoE evaluation value of each of the poor quality stable segments under different playback parameters with the preset parsing parameters for video service perception, to obtain a service parsing coefficient for video service perception for each of the poor quality stable segments;

[0063] A transmission module is used to transmit the target video source, wherein the video segment corresponding to each of the poor-quality stable segments in the target video source includes the business resolution coefficient.

[0064] The embodiment of the present invention further provides a video service perception device, which includes:

[0065] An acquisition module is used to acquire a target video source; wherein the video segment corresponding to each of the poor quality stable segments in the target video source includes the business resolution coefficient;

[0066] The parsing module is used to perform parsing analysis on the service quality of each of the poor-quality stable segments of the target video source according to the service parsing coefficient, so as to obtain a service perception evaluation result of the target video source.

[0067] An embodiment of the present invention further provides a network device, which includes a processor, a memory, and a program stored in the memory and executable on the processor, wherein the program, when executed by the processor, implements the steps in the video service perception method as described in any one of the above items.

[0068] An embodiment of the present invention further provides a readable storage medium, wherein a program is stored on the readable storage medium, and when the program is executed by a processor, the steps in the video service perception method as described in any one of the above items are implemented.

[0069] At least one of the above technical solutions of the present invention has the following beneficial effects:

[0070] The video service perception method described in the embodiment of the present invention uses a complex model to calculate the QoE evaluation value of the stable quality segment corresponding to different playback parameters, thereby ensuring the real-time accuracy of the subsequent video quality evaluation; a simple model is used to fit the QoE evaluation value with the preset analytical coefficient of video service perception, thereby obtaining the service analytical coefficient for video service perception, thereby simplifying the fitting process of the video service perception parameters. Therefore, the method described in the embodiment of the present invention can simplify the fitting process of the video service perception parameters, while ensuring the real-time accuracy of the perception algorithm and reducing the influence of the video content on the accuracy of the simplified algorithm. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] Figure 1 This is a flow chart of implementation mode 1 of the video service perception method according to an embodiment of the present invention;

[0072] Figure 2This is a flow chart of implementation mode 2 of the video service perception method according to an embodiment of the present invention;

[0073] Figure 3 A schematic diagram of the structure of a video transmission device according to an embodiment of the present invention;

[0074] Figure 4 A schematic diagram of the structure of a video receiving device according to an embodiment of the present invention;

[0075] Figure 5 This is a structural diagram of implementation mode 1 of the video service perception device according to an embodiment of the present invention;

[0076] Figure 6 This is a structural diagram of implementation mode 2 of the video service perception device according to an embodiment of the present invention;

[0077] Figure 7 This is a schematic diagram of the structure of implementation mode 1 of the network device according to an embodiment of the present invention;

[0078] Figure 8 This is a structural diagram of implementation mode 2 of the network device described in an embodiment of the present invention. DETAILED DESCRIPTION

[0079] In order to make the technical problems, technical solutions and advantages to be solved by the present invention more clear, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0080] In order to simplify the fitting process of video service perception parameters while ensuring the real-time accuracy of the perception algorithm and reducing the influence of video content on the accuracy of the simplified algorithm, an embodiment of the present invention provides a method for video service perception, which transfers the complexity of video quality evaluation during video transmission or reception to a one-time calculation at the video source end, segments the video source at the video source end, and evaluates to obtain multiple stable quality-difference segments. Using a complex algorithm with higher accuracy, the experience quality QoE evaluation value of the stable quality-difference segment under different playback parameters can be calculated; thereafter, a simple model can be used to fit the QoE evaluation value under different playback parameters with the preset analytical coefficients of video service perception to obtain the model coefficients of the simple model as the simple model evaluation coefficients of the video source segment, so that it can be used for the subsequent network listening end or the video receiving end to perform video quality evaluation on the video segment.

[0081] The above-mentioned video service perception method of the embodiment of the present invention can use a complex model to calculate the QoE evaluation value of the stable quality segment corresponding to different playback parameters, ensuring the real-time accuracy of the subsequent video quality evaluation; a simple model can be used to fit the QoE evaluation value with the preset analytical coefficient of video service perception, obtain the service analytical coefficient for video service perception, and simplify the fitting process of video service perception parameters. Therefore, the method described in the embodiment of the present invention can simplify the fitting process of video service perception parameters, while ensuring the real-time accuracy of the perception algorithm and reducing the influence of video content on the accuracy of the simplified algorithm.

[0082] Specifically, Figure 1 As shown, the method for video service perception described in the embodiment of the present invention includes:

[0083] S110, analyzing the quality parameters of the target video source to obtain a plurality of stable quality-poor segments with stable quality parameters in the target video source;

[0084] S120, calculating a quality of experience (QoE) evaluation value of each of the stable quality difference segments under different playback parameters;

[0085] S130, fitting the QoE evaluation value of each stable segment with poor quality under different playback parameters with preset analysis parameters for video service perception, to obtain a service analysis coefficient for video service perception for each stable segment with poor quality;

[0086] S140, transmitting the target video source, wherein the video segment corresponding to each of the poor quality stable segments in the target video source includes the service resolution coefficient.

[0087] Optionally, in step S110, the target video source is a video transmitted by a video source end. The method for video service perception described in an embodiment of the present invention analyzes quality parameters of the video at the video source end, and can calculate the QoE evaluation value of each time period through a model training method. Further, based on the QoE evaluation value of each time period, multiple stable quality difference segments can be analyzed and obtained.

[0088] In the embodiment of the present invention, optionally, the quality parameters of the target video source include but are not limited to only including packet loss rate, delay and / or bit error rate and the like.

[0089] Specifically, step S110, analyzing the quality parameters of the target video source to obtain a plurality of stable quality-poor segments with stable quality parameters in the target video source, includes:

[0090] Calculate the QoE evaluation value of each time period by a model training method according to the quality parameter of the target video source in each time period;

[0091] According to the QoE evaluation value of each time period, the transition segments obtained after the target video source is segmented for transition are analyzed to obtain a plurality of quality difference stable segments.

[0092] According to the quality parameter of the target video source in each time period, the QoE evaluation value of each time period is calculated by a model training method, including:

[0093] Classifying the scene corresponding to each quality parameter in the target video source;

[0094] The QoE evaluation value of the quality parameter of the target video source in each of the time periods in the corresponding different grading combinations is calculated by a model training method.

[0095] For example, when the quality parameters include network packet loss rate and delay rate, the corresponding scenarios with network packet loss can be graded according to different packet loss rates, such as dividing the packet loss rate into several levels including 0.1%, 0.4%, 1%, 3%, 5% and 10%, and the level of network packet loss in the corresponding scenario is determined according to the packet loss rate in the network packet loss scenario; the corresponding scenarios with network delay can also be graded according to different delay rates, such as dividing the delay rate into multiple levels, and determining the level of network delay in the corresponding scenario according to the delay situation in the network delay scenario.

[0096] In the case where each quality parameter corresponding to the scene is graded separately, a fixed time period T1 can be further set, and the QoE evaluation value of the quality parameters of the target video source in different time periods T1 in different graded combinations can be obtained through a model training method.

[0097] It should be noted that, in the embodiment of the present invention, the above-mentioned method for calculating the QoE evaluation value of the quality parameters of the target video source in different time periods T1 in different graded combinations can be calculated using a conventional complex model training method. Optionally, the complex model training method can be a full-media artificial intelligence training method, that is, model training is performed based on manual annotation and combined with input parameters, so as to achieve a model with better manual judgment and obtain a higher evaluation accuracy value.

[0098] Those skilled in the art should be able to understand the specific methods of all-media artificial intelligence training, which will not be described in detail here.

[0099] The method for video service perception described in the embodiment of the present invention can obtain the QoE evaluation values ​​of the quality parameters of the target video source in different time periods T1 in different graded combinations, and can further analyze the transition segments obtained after the target video source is transition-segmented according to the QoE evaluation values ​​of each time period to obtain multiple stable quality-difference segments.

[0100] Specifically, the method further includes: performing transition segmentation on the target video source.

[0101] Among them, the method of performing transition segmentation on the target video source can be based on the video playback scene in the video source as the division basis, dividing videos belonging to different playback scenes into different transition segments, and determining the transition time point of each transition segment.

[0102] Optionally, the method further comprises: marking each transition segment, such as marking a start time point, an end time point and / or segment duration of each transition segment.

[0103] For example, according to the playback time sorting, the end time point of each transition segment is marked in sequence as: t_context(0), ..., t_context(i) and t_context(i+1), etc.; and the segment duration of each transition segment is marked in sequence as: duration_t_context(0), ..., duration_t_context(i) and duration_t_context(i+1), etc.

[0104] In the embodiment of the present invention, after the target video source is segmented for transition, the transition segments obtained after the target video source is segmented for transition are analyzed according to the QoE evaluation value of each time period, and a plurality of quality difference stable segments are obtained, including:

[0105] When the duration of the transition segment is greater than the duration period, determining whether the QoE evaluation value obtained corresponding to each duration period in the transition segment meets a preset stability condition;

[0106] Determine the video segment corresponding to the time period that meets the preset stability condition as the poor quality stable segment;

[0107] When the duration of the transition segment is less than or equal to the duration period, and the corresponding QoE evaluation value obtained within the duration period corresponding to the transition segment meets the preset stability condition, the transition segment is determined to be the poor quality stable segment.

[0108] Specifically, when the duration of one of the transition segments is greater than the duration period T1, the stability of the QoE evaluation value in each duration period T1 in the transition segment is determined, and the video segment corresponding to the duration period that meets the preset stability condition is determined as a poor quality stable segment.

[0109] Optionally, when the QoE evaluation value calculated within one of the time periods T1 is less than a preset threshold, it can be determined that the quality of the corresponding video segment is stable under the same damage, otherwise it is considered that the quality is unstable, that is, it does not belong to a poor quality stable segment.

[0110] When the duration of one of the transition segments is less than or equal to the duration period T1, if the corresponding QoE evaluation value obtained within the duration period corresponding to the transition segment meets the preset stability condition, the transition segment is determined to be the poor quality stable segment, otherwise the transition segment is considered not to be a poor quality stable segment.

[0111] Optionally, the method further comprises:

[0112] When transmitting the target video source, the poor quality stable segment is identified.

[0113] Optionally, the method further includes: recording each identified poor-quality stable segment to obtain a poor-quality stable segment group.

[0114] In the embodiment of the present invention, optionally, in step S120, calculating the quality of experience (QoE) evaluation value of each of the poor quality stable segments under different playback parameters includes:

[0115] Selecting a target stable quality segment from at least one stable quality segment in each transition segment; the difference between the QoE evaluation value of the target stable quality segment and the average value of the QoE evaluation value of the transition segment is the smallest;

[0116] The quality of experience (QoE) evaluation value of each target quality difference stable segment under different playback parameters is calculated.

[0117] Specifically, in the method described in the embodiment of the present invention, for the quality difference stable segment determined in each transition segment, a target quality difference stable segment with the smallest difference between the QoE evaluation value and the average value of the QoE evaluation value of the transition segment is selected to calculate the QoE evaluation value under different playback parameters.

[0118] Since the difference between the QoE evaluation value of the target quality difference stable segment and the average value of the QoE evaluation value of the corresponding transition segment is the smallest, the method described in the embodiment of the present invention can accurately reflect the QoE evaluation value of the entire transition segment under different playback parameters and other combination conditions by only selecting the target quality difference stable segment with the smallest difference in QoE evaluation value to perform QoE evaluation under different playback parameters and other combination conditions. In this way, the purpose of simplifying the algorithm is achieved while ensuring the calculation accuracy.

[0119] In the embodiment of the present invention, optionally, different playback parameters include but are not limited to resolution, bit rate and / or frame rate, etc.

[0120] Using the above implementation mode, the target quality difference stable segment is a representative segment used to calculate the QoE evaluation value under different combinations of playback parameter conditions, which can also be called a representative period. The overall QoE evaluation value of the quality difference stable segment in which the target quality difference stable segment is located is represented by calculating the QoE evaluation value of more parameter combinations for the representative period.

[0121] The method for video service perception described in the embodiment of the present invention, after obtaining the overall QoE evaluation value of each poor-quality stable segment, fits the QoE evaluation value of each poor-quality stable segment under different playback parameters with the preset analysis parameters for video service perception through step S130, so as to obtain the service analysis coefficient for video service perception of each poor-quality stable segment.

[0122] Furthermore, based on the obtained service resolution coefficient, when the video source end transmits the target video source, the video segment corresponding to each of the poor quality stable segments in the target video source includes the service resolution coefficient, so that it can be used for subsequent network monitoring end or video receiving end to perform video quality assessment of the video segment.

[0123] In this way, by using a simple model to fit the preset analytical parameters, the model coefficients of the simple model are obtained, which can be used as the simple model evaluation coefficients of the poor-quality stable segment.

[0124] Optionally, in step S130, a service analysis coefficient for service perception of each stable quality difference segment may be obtained by fitting algorithms such as least squares and random forest.

[0125] Optionally, the method further comprises: when transmitting the target video source, identifying the poor quality stable segment, so that the poor quality stable segment can be identified at the network monitoring end or the video receiving end.

[0126] The method for video service perception described in the embodiment of the present invention can simplify the fitting process of video service perception parameters, while ensuring the real-time accuracy of the perception algorithm and reducing the influence of video content on the accuracy of the simplified algorithm.

[0127] It should be noted that in the embodiments of the present invention, the simple algorithm mentioned in the embodiments of the present invention generally refers to a network planning model of the transmission layer and an evaluation algorithm with fewer input parameters; the complex algorithm generally refers to a bit stream model or a hybrid model. The embodiments of the present invention do not specifically limit the algorithm model used in each implementation process.

[0128] The embodiment of the present invention also provides a method for video service perception, such as Figure 2 As shown, including:

[0129] S210, obtaining a target video source; wherein the video segment corresponding to each of the poor quality stable segments in the target video source includes the business resolution coefficient;

[0130] S220: Analyze the service quality of each of the poor-quality stable segments of the target video source according to the service analysis coefficient to obtain a service perception evaluation result of the target video source.

[0131] The method for video service perception described in the embodiment of the present invention is adopted to obtain the network monitoring end or video receiving end of the target video source, apply the service analysis coefficient in the target video source to the video quality analysis algorithm, and obtain the service perception evaluation result of the target video source.

[0132] Optionally, a network monitoring end or a video receiving end of the target video source is obtained, and a simplified video quality analysis algorithm is used to perform a business perception evaluation analysis of the target video source; specifically, by replacing the coefficients in the simplified video quality analysis algorithm with the business analysis coefficients of each stable quality-poor segment, the business perception evaluation result of the target video source can be obtained.

[0133] Optionally, the video quality analysis simplification algorithm uses the delay, bit rate and packet loss rate of the monitored target video source, combined with the business analysis coefficient in the obtained video clip, to obtain the business perception evaluation result of the target video source

[0134] Therefore, optionally, in the method for video service perception, in step S220, the service quality of each of the poor quality stable segments of the target video source is analyzed according to the service analysis coefficient, including:

[0135] The business analysis coefficient corresponding to the poor quality stable segment is used to replace the calculation coefficient in the business quality simplification algorithm to obtain the business perception evaluation result of the target video source.

[0136] It should be noted that, in the embodiment of the present invention, the simplified video quality analysis algorithm does not specifically refer to a certain simplified analysis algorithm, but generally refers to a calculation model that requires fewer input parameters and some fitting coefficients.

[0137] The method of video service perception described in the embodiment of the present invention is adopted to transfer the complexity of video quality assessment during video transmission or reception to the video source end for one-time calculation. After the video is segmented and judged as a stable segment with poor quality, the data of a specific time period (target stable segment with poor quality) is selected to form a data set. The data set is combined according to multiple levels such as different resolutions, different bit rates, different delays and different packet losses, and the evaluation quality of each data in the data set is annotated using a complex algorithm with high accuracy. Then, a simple model is used to fit the annotated data to obtain the model coefficient of the simple model, that is, the business analysis coefficient of each stable segment with poor quality is obtained as the simple model evaluation coefficient of the segment.

[0138] In video transmission, the network monitoring end or the receiving end can perform analytical calculations based on the simple evaluation model of the stable quality-poor segment and the service analysis coefficient to obtain the video quality of the current segment.

[0139] The method can simplify the fitting process of video service perception parameters, while ensuring the real-time accuracy of the perception algorithm and reducing the impact of video content on the accuracy of the simplified algorithm.

[0140] The embodiment of the present invention also provides a video transmission device, such as Figure 3 As shown, the processor 310 is included, wherein the processor 310 is used for:

[0141] Analyze the quality parameters of the target video source to obtain a plurality of stable quality-poor segments of the target video source with stable quality parameters;

[0142] Calculate the quality of experience (QoE) evaluation value of each of the stable quality difference segments under different playback parameters;

[0143] Fitting the QoE evaluation value of each stable segment with poor quality under different playback parameters with the preset analysis parameters for video service perception to obtain a service analysis coefficient for video service perception for each stable segment with poor quality;

[0144] The target video source is transmitted, wherein the video segment corresponding to each of the poor quality stable segments in the target video source includes the business resolution coefficient.

[0145] Optionally, in the video transmission device, the processor 310 analyzes the quality parameters of the target video source to obtain a plurality of stable quality-poor segments with stable quality parameters in the target video source, including:

[0146] Calculate the QoE evaluation value of each time period by a model training method according to the quality parameter of the target video source in each time period;

[0147] According to the QoE evaluation value of each time period, the transition segments obtained after the target video source is segmented for transition are analyzed to obtain a plurality of quality difference stable segments.

[0148] Optionally, in the video transmission device, the processor 310 calculates the QoE evaluation value of each time period by a model training method according to the quality parameter of the target video source in each time period, including:

[0149] Classifying the scene corresponding to each quality parameter in the target video source;

[0150] The QoE evaluation value of the quality parameter of the target video source in each of the time periods in the corresponding different grading combinations is calculated by a model training method.

[0151] Optionally, in the video transmission device, the processor 310 analyzes the transition segments obtained after the target video source is segmented according to the QoE evaluation value of each time period to obtain a plurality of quality difference stable segments, including:

[0152] When the duration of the transition segment is greater than the duration period, determining whether the QoE evaluation value obtained corresponding to each duration period in the transition segment meets a preset stability condition;

[0153] Determine the video segment corresponding to the time period that meets the preset stability condition as the poor quality stable segment;

[0154] When the duration of the transition segment is less than or equal to the duration period, and the corresponding QoE evaluation value obtained within the duration period corresponding to the transition segment meets the preset stability condition, the transition segment is determined to be the poor quality stable segment.

[0155] Optionally, in the video transmission device, the processor 310 calculates the quality of experience (QoE) evaluation value of each of the poor quality stable segments under different playback parameters, including:

[0156] Selecting a target stable quality segment from at least one stable quality segment in each transition segment; the difference between the QoE evaluation value of the target stable quality segment and the average value of the QoE evaluation value of the transition segment is the smallest;

[0157] The quality of experience (QoE) evaluation value of each target quality difference stable segment under different playback parameters is calculated.

[0158] Optionally, in the video transmission device, the processor 310 is further configured to:

[0159] When transmitting the target video source, the starting time point, the ending time point and / or the segment duration of each transition segment are marked.

[0160] Optionally, in the video transmission device, the processor 310 is further configured to:

[0161] When transmitting the target video source, the poor quality stable segment is identified.

[0162] The embodiment of the present invention also provides a video receiving device, such as Figure 4 As shown, a processor 410 is included, wherein the processor 410 is used for:

[0163] Acquire a target video source; wherein the video segment corresponding to each of the poor quality stable segments in the target video source includes the business resolution coefficient;

[0164] According to the service analysis coefficient, the service quality of each of the poor-quality stable segments of the target video source is analyzed to obtain a service perception evaluation result of the target video source.

[0165] Optionally, in the video receiving device, the processor 410 performs analytical analysis on the service quality of each poor quality stable segment of the target video source according to the service analytical coefficient, including:

[0166] The business analysis coefficient corresponding to the poor quality stable segment is used to replace the calculation coefficient in the business quality simplification algorithm to obtain the business perception evaluation result of the target video source.

[0167] The embodiment of the present invention also provides a video service perception device, such as Figure 5 As shown, including:

[0168] An analysis module 510 is used to analyze the quality parameters of the target video source to obtain a plurality of stable quality-poor segments with stable quality parameters in the target video source;

[0169] A calculation module 520, configured to calculate a quality of experience (QoE) evaluation value of each of the poor quality stable segments under different playback parameters;

[0170] A fitting module 530, configured to fit the QoE evaluation value of each of the poor quality stable segments under different playback parameters with preset analysis parameters for video service perception, to obtain a service analysis coefficient for video service perception for each of the poor quality stable segments;

[0171] The transmission module 540 is configured to transmit the target video source, wherein the video segment corresponding to each of the poor quality stable segments in the target video source includes the service resolution coefficient.

[0172] Optionally, in the video service perception device, the analyzing module 510 analyzes the quality parameters of the target video source to obtain a plurality of stable quality-poor segments with stable quality parameters in the target video source, including:

[0173] Calculate the QoE evaluation value of each time period by a model training method according to the quality parameter of the target video source in each time period;

[0174] According to the QoE evaluation value of each time period, the transition segments obtained after the target video source is segmented for transition are analyzed to obtain a plurality of quality difference stable segments.

[0175] Optionally, in the video service perception device, the analysis module 510 calculates the QoE evaluation value of each time period by a model training method according to the quality parameter of the target video source in each time period, including:

[0176] Classifying the scene corresponding to each quality parameter in the target video source;

[0177] The QoE evaluation value of the quality parameter of the target video source in each of the time periods in the corresponding different grading combinations is calculated by a model training method.

[0178] Optionally, in the video service perception device, the analysis module 510 analyzes the transition segments obtained after the target video source is segmented according to the QoE evaluation value of each time period to obtain multiple quality difference stable segments, including:

[0179] When the duration of the transition segment is greater than the duration period, determining whether the QoE evaluation value obtained corresponding to each duration period in the transition segment meets a preset stability condition;

[0180] Determine the video segment corresponding to the time period that meets the preset stability condition as the poor-quality stable segment;

[0181] When the duration of the transition segment is less than or equal to the duration period, and the corresponding QoE evaluation value obtained within the duration period corresponding to the transition segment meets the preset stability condition, the transition segment is determined to be the poor quality stable segment.

[0182] Optionally, in the video service perception device, the calculation module 520 calculates the experience quality QoE evaluation value of each of the poor quality stable segments under different playback parameters, including:

[0183] Selecting a target stable quality segment from at least one stable quality segment in each transition segment; the difference between the QoE evaluation value of the target stable quality segment and the average value of the QoE evaluation value of the transition segment is the smallest;

[0184] The quality of experience (QoE) evaluation value of each target quality difference stable segment under different playback parameters is calculated.

[0185] Optionally, in the video service perception device, the transmission module 540 is further used for:

[0186] When transmitting the target video source, the starting time point, the ending time point and / or the segment duration of each transition segment are marked.

[0187] Optionally, in the video service perception device, the transmission module 540 is further used for:

[0188] When transmitting the target video source, the poor quality stable segment is identified.

[0189] The embodiment of the present invention also provides a video service perception device, such as Figure 6 As shown, including:

[0190] An acquisition module 610 is configured to acquire a target video source; wherein the video segment corresponding to each of the poor quality stable segments in the target video source includes the business resolution coefficient;

[0191] The analysis module 620 is used to analyze the service quality of each of the poor-quality stable segments of the target video source according to the service analysis coefficient to obtain a service perception evaluation result of the target video source.

[0192] Optionally, in the video service perception device, the parsing module 620 performs parsing analysis on the service quality of each of the poor quality stable segments of the target video source according to the service parsing coefficient, including:

[0193] The business analysis coefficient corresponding to the poor quality stable segment is used to replace the calculation coefficient in the business quality simplification algorithm to obtain the business perception evaluation result of the target video source.

[0194] Another aspect of the present invention is to provide a network device, such as Figure 7 As shown, it includes: a processor 701; and a memory 703 connected to the processor 701 through a bus interface 702, the memory 703 is used to store programs and data used by the processor 701 when performing operations, and the processor 701 calls and executes the programs and data stored in the memory 703.

[0195] The transceiver 704 is connected to the bus interface 702 and is used to receive and send data under the control of the processor 701. Specifically, the processor 701 is used to read the program in the memory 703 and execute the following process:

[0196] Analyze the quality parameters of the target video source to obtain a plurality of stable quality-poor segments of the target video source with stable quality parameters;

[0197] Calculate the quality of experience (QoE) evaluation value of each of the stable quality difference segments under different playback parameters;

[0198] Fitting the QoE evaluation value of each stable segment with poor quality under different playback parameters with the preset analysis parameters for video service perception to obtain a service analysis coefficient for video service perception for each stable segment with poor quality;

[0199] The target video source is transmitted, wherein the video segment corresponding to each of the poor quality stable segments in the target video source includes the business resolution coefficient.

[0200] Optionally, in the network device, the processor 701 analyzes the quality parameters of the target video source to obtain a plurality of stable quality-poor segments with stable quality parameters in the target video source, including:

[0201] Calculate the QoE evaluation value of each time period by a model training method according to the quality parameter of the target video source in each time period;

[0202] According to the QoE evaluation value of each time period, the transition segments obtained after the target video source is segmented for transition are analyzed to obtain a plurality of quality difference stable segments.

[0203] Optionally, in the network device, the processor 701 calculates the QoE evaluation value of each time period by a model training method according to the quality parameter of the target video source in each time period, including:

[0204] Classifying the scene corresponding to each quality parameter in the target video source;

[0205] The QoE evaluation value of the quality parameter of the target video source in each of the time periods in the corresponding different grading combinations is calculated by a model training method.

[0206] Optionally, in the network device, the processor 701 analyzes the transition segments obtained after the target video source is segmented according to the QoE evaluation value of each time period, and obtains a plurality of quality difference stable segments, including:

[0207] When the duration of the transition segment is greater than the duration period, determining whether the QoE evaluation value obtained corresponding to each duration period in the transition segment meets a preset stability condition;

[0208] Determine the video segment corresponding to the time period that meets the preset stability condition as the poor quality stable segment;

[0209] When the duration of the transition segment is less than or equal to the duration period, and the corresponding QoE evaluation value obtained within the duration period corresponding to the transition segment meets the preset stability condition, the transition segment is determined to be the poor quality stable segment.

[0210] Optionally, in the network device, the processor 701 calculates the experience quality QoE evaluation value of each of the poor quality stable segments under different playback parameters, including:

[0211] Selecting a target stable quality segment from at least one stable quality segment in each transition segment; the difference between the QoE evaluation value of the target stable quality segment and the average value of the QoE evaluation value of the transition segment is the smallest;

[0212] The quality of experience (QoE) evaluation value of each target quality difference stable segment under different playback parameters is calculated.

[0213] Optionally, in the network device, the processor 701 is further configured to:

[0214] When transmitting the target video source, the starting time point, the ending time point and / or the segment duration of each transition segment are marked.

[0215] Optionally, in the network device, the processor 701 is further configured to:

[0216] When transmitting the target video source, the poor quality stable segment is identified.

[0217] Another aspect of the present invention is to provide a network device, such as Figure 8 As shown, it includes: a processor 801; and a memory 803 connected to the processor 801 through a bus interface 802, the memory 803 is used to store programs and data used by the processor 801 when performing operations, and the processor 801 calls and executes the programs and data stored in the memory 803.

[0218] The transceiver 804 is connected to the bus interface 802 and is used to receive and send data under the control of the processor 801. Specifically, the processor 801 is used to read the program in the memory 803 and execute the following process: obtaining a target video source; wherein the video segment corresponding to each of the poor quality stable segments in the target video source includes the business resolution coefficient;

[0219] According to the service analysis coefficient, the service quality of each of the poor-quality stable segments of the target video source is analyzed to obtain a service perception evaluation result of the target video source.

[0220] Optionally, in the network device, the processor 801 performs analytical analysis on the service quality of each of the poor quality stable segments of the target video source according to the service analysis coefficient, including:

[0221] The business analysis coefficient corresponding to the poor quality stable segment is used to replace the calculation coefficient in the business quality simplification algorithm to obtain the business perception evaluation result of the target video source.

[0222] In addition, a specific embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored, wherein the program, when executed by a processor, implements the steps in any one of the above-described video service perception methods.

[0223] Specifically, the computer-readable storage medium is applied to the above-mentioned terminal. When applied to the terminal, the execution steps in the method for reporting smoke alarms are described in detail above and will not be repeated here.

[0224] In the several embodiments provided in the present application, it should be understood that the disclosed methods and devices can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0225] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may be physically included separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0226] The above-mentioned integrated unit implemented in the form of a software functional unit can be stored in a computer-readable storage medium. The above-mentioned software functional unit 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 some steps of the sending and receiving methods described in various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, referred to as ROM), random access memory (Random Access Memory, referred to as RAM), disk or optical disk and other media that can store program codes.

[0227] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary personnel in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A video service perception method, characterized in that: include: Analyze the quality parameters of the target video source to obtain a plurality of stable quality-poor segments of the target video source with stable quality parameters; Calculate the quality of experience (QoE) evaluation value of each stable quality segment under different playback parameters according to a complex model training method; wherein the complex model training method includes an all-media artificial intelligence training method; Fitting the QoE evaluation value of each stable segment with poor quality under different playback parameters with the preset analysis parameters for video service perception to obtain a service analysis coefficient for video service perception for each stable segment with poor quality; The target video source is transmitted, wherein the video segment corresponding to each of the poor quality stable segments in the target video source includes the business resolution coefficient.

2. The video service perception method according to claim 1, characterized in that: The analyzing the quality parameters of the target video source to obtain a plurality of stable quality-poor segments with stable quality parameters in the target video source includes: Calculate the QoE evaluation value of each time period by a model training method according to the quality parameter of the target video source in each time period; According to the QoE evaluation value of each time period, the transition segments obtained after the target video source is segmented for transition are analyzed to obtain a plurality of quality difference stable segments.

3. The video service perception method according to claim 2, characterized in that: According to the quality parameter of the target video source in each time period, a QoE evaluation value of each time period is calculated by a model training method, including: Classifying the scene corresponding to each quality parameter in the target video source; The QoE evaluation value of the quality parameter of the target video source in each of the time periods in the corresponding different grading combinations is calculated by a model training method.

4. The video service perception method according to claim 3, characterized in that: According to the QoE evaluation value of each time period, the transition segments obtained after the target video source is segmented and transitioned are analyzed to obtain multiple quality difference stable segments, including: When the duration of the transition segment is greater than the duration period, determining whether the QoE evaluation value obtained corresponding to each duration period in the transition segment meets a preset stability condition; Determine the video segment corresponding to the time period that meets the preset stability condition as the poor-quality stable segment; When the duration of the transition segment is less than or equal to the duration period, and the corresponding QoE evaluation value obtained within the duration period corresponding to the transition segment meets the preset stability condition, the transition segment is determined to be the poor quality stable segment.

5. The video service perception method according to any one of claims 2 to 4, characterized in that: The calculating of the quality of experience (QoE) evaluation value of each stable quality segment under different playback parameters includes: Selecting a target stable quality segment from at least one stable quality segment in each transition segment; the difference between the QoE evaluation value of the target stable quality segment and the average value of the QoE evaluation value of the transition segment is the smallest; The quality of experience (QoE) evaluation value of each target quality difference stable segment under different playback parameters is calculated.

6. The video service perception method according to claim 2, characterized in that: The method further comprises: When transmitting the target video source, the starting time point, the ending time point and / or the segment duration of each transition segment are marked.

7. The video service perception method according to any one of claims 1 to 4, characterized in that: The method further comprises: When transmitting the target video source, the poor quality stable segment is identified.

8. A video service perception method, characterized in that: include: Obtain a target video source; wherein the video segment corresponding to each stable quality segment in the target video source includes a business analysis coefficient; the target video source includes a plurality of stable quality segments with stable quality parameters; the business analysis coefficient is obtained by fitting the QoE evaluation value of each stable quality segment under different playback parameters with the preset analysis parameters for video business perception; the QoE evaluation value of each stable quality segment under different playback parameters is calculated according to a complex model training method; the complex model training method includes an all-media artificial intelligence training method; According to the service analysis coefficient, the service quality of each of the poor-quality stable segments of the target video source is analyzed to obtain a service perception evaluation result of the target video source.

9. The video service perception method according to claim 8, characterized in that: According to the service analysis coefficient, the service quality of each of the poor quality stable segments of the target video source is analyzed, including: The business analysis coefficient corresponding to the poor quality stable segment is used to replace the calculation coefficient in the business quality simplification algorithm to obtain the business perception evaluation result of the target video source.

10. A video transmission device, comprising a processor, characterized in that: The processor is used to: Analyze the quality parameters of the target video source to obtain a plurality of stable quality-poor segments of the target video source with stable quality parameters; Calculate the quality of experience (QoE) evaluation value of each stable quality segment under different playback parameters according to a complex model training method; wherein the complex model training method includes an all-media artificial intelligence training method; Fitting the QoE evaluation value of each stable segment with poor quality under different playback parameters with the preset analysis parameters for video service perception to obtain a service analysis coefficient for video service perception for each stable segment with poor quality; The target video source is transmitted, wherein the video segment corresponding to each of the poor quality stable segments in the target video source includes the business resolution coefficient.

11. The video transmission device according to claim 10, characterized in that: The processor analyzes the quality parameters of the target video source to obtain a plurality of stable quality-poor segments with stable quality parameters in the target video source, including: Calculate the QoE evaluation value of each time period by a model training method according to the quality parameter of the target video source in each time period; According to the QoE evaluation value of each time period, the transition segments obtained after the target video source is segmented for transition are analyzed to obtain a plurality of quality difference stable segments.

12. The video transmission device according to claim 11, characterized in that: The processor calculates the QoE evaluation value of each time period by a model training method according to the quality parameter of the target video source in each time period, including: Classifying the scene corresponding to each quality parameter in the target video source; The QoE evaluation value of the quality parameter of the target video source in each of the time periods in the corresponding different grading combinations is calculated by a model training method.

13. The video transmission device according to claim 12, characterized in that: The processor analyzes the transition segments obtained after the target video source is segmented according to the QoE evaluation value of each time period, and obtains a plurality of quality difference stable segments, including: When the duration of the transition segment is greater than the duration period, determining whether the QoE evaluation value obtained corresponding to each duration period in the transition segment meets a preset stability condition; Determine the video segment corresponding to the time period that meets the preset stability condition as the poor-quality stable segment; When the duration of the transition segment is less than or equal to the duration period, and the corresponding QoE evaluation value obtained within the duration period corresponding to the transition segment meets the preset stability condition, the transition segment is determined to be the poor quality stable segment.

14. The video transmission device according to any one of claims 11 to 13, characterized in that: The processor calculates a quality of experience (QoE) evaluation value of each stable quality segment under different playback parameters, including: Selecting a target stable quality segment from at least one stable quality segment in each transition segment; the difference between the QoE evaluation value of the target stable quality segment and the average value of the QoE evaluation value of the transition segment is the smallest; The quality of experience (QoE) evaluation value of each target quality difference stable segment under different playback parameters is calculated.

15. The video transmission device according to claim 11, characterized in that: The processor is further configured to: When transmitting the target video source, the starting time point, the ending time point and / or the segment duration of each transition segment are marked.

16. The video transmission device according to any one of claims 10 to 13, characterized in that: The processor is further configured to: When transmitting the target video source, the poor quality stable segment is identified.

17. A video receiving device, comprising a processor, characterized in that: The processor is used to: Obtain a target video source; wherein the video segment corresponding to each stable quality segment in the target video source includes a business analysis coefficient; the target video source includes a plurality of stable quality segments with stable quality parameters; the business analysis coefficient is obtained by fitting the QoE evaluation value of each stable quality segment under different playback parameters with the preset analysis parameters for video business perception; the QoE evaluation value of each stable quality segment under different playback parameters is calculated according to a complex model training method; the complex model training method includes an all-media artificial intelligence training method; According to the service analysis coefficient, the service quality of each of the poor-quality stable segments of the target video source is analyzed to obtain a service perception evaluation result of the target video source.

18. The video receiving device according to claim 17, characterized in that: The processor performs analytical analysis on the service quality of each of the poor quality stable segments of the target video source according to the service analytical coefficient, including: The business analysis coefficient corresponding to the poor quality stable segment is used to replace the calculation coefficient in the business quality simplification algorithm to obtain the business perception evaluation result of the target video source.

19. A video service perception device, characterized in that: include: An analysis module, configured to analyze the quality parameters of the target video source and obtain a plurality of stable quality-poor segments with stable quality parameters in the target video source; A calculation module, used to calculate the experience quality QoE evaluation value of each of the poor quality stable segments under different playback parameters according to a complex model training method; wherein the complex model training method includes an all-media artificial intelligence training method; A fitting module, used for fitting the QoE evaluation value of each of the poor quality stable segments under different playback parameters with the preset analysis parameters for video service perception, to obtain a service analysis coefficient for video service perception for each of the poor quality stable segments; A transmission module is used to transmit the target video source, wherein the video segment corresponding to each of the poor-quality stable segments in the target video source includes the business resolution coefficient.

20. A video service perception device, characterized in that: include: An acquisition module is used to acquire a target video source; wherein the video segment corresponding to each stable quality segment in the target video source includes a business analysis coefficient; the target video source includes a plurality of stable quality segments with stable quality parameters; the business analysis coefficient is obtained by fitting the QoE evaluation value of each stable quality segment under different playback parameters with a preset analysis parameter for video business perception; the QoE evaluation value of each stable quality segment under different playback parameters is calculated according to a complex model training method; the complex model training method includes an all-media artificial intelligence training method; The parsing module is used to perform parsing analysis on the service quality of each of the poor-quality stable segments of the target video source according to the service parsing coefficient, so as to obtain a service perception evaluation result of the target video source.

21. A network device, characterized in that: The method comprises a processor, a memory and a program stored in the memory and executable on the processor, wherein the program, when executed by the processor, implements the steps of the video service perception method as claimed in any one of claims 1 to 7, or implements the steps of the video service perception method as claimed in any one of claims 8 to 9.

22. A readable storage medium, characterized in that: The readable storage medium stores a program, and when the program is executed by the processor, the program implements the steps of the video service perception method according to any one of claims 1 to 7, or implements the steps of the video service perception method according to any one of claims 8 to 9.