Single-point matching method for OTT quality detection mode matching system

Abstract

The invention discloses a single-point matching method for an OTT quality detection mode matching system, which can more accurately detect the service quality problem and improve the matching precision by carrying out single-point matching on OTT service data, and can improve the matching accuracy due to the fact that the single-point matching method is relatively simple and small in calculation amount. A large amount of OTT service data can be rapidly detected, and the detection efficiency is improved; according to the method, the single-point matching model is constructed according to the specific OTT service quality characteristics of the code rate detection of the OTT video stream, the quality detection of the OTT audio stream and the quality detection of the OTT audio stream, so that different quality detection modes can be realized, and the flexibility is relatively high.

Description

Technical Field

[0001] The present invention relates to the technical field of information transmission, and in particular to a single-point matching method for an OTT quality detection pattern matching system. Background Art

[0002] The OTT quality detection pattern matching system is a tool or system for quality detection of OTT (Over The Top) services. The system uses specific algorithms and models to perform quality detection on each link in the OTT service to ensure the quality and stability of the service. The OTT quality detection pattern matching system is widely used in various OTT service scenarios, such as video streaming, audio streaming, VoIP communication, etc. Through this system, service providers can understand the quality status of services in real time, discover and solve potential problems in a timely manner, and improve user satisfaction and loyalty.

[0003] However, existing OTT quality detection pattern matching systems mainly use global matching or local matching methods for quality assessment. The global matching method usually requires converting the entire video stream into frame-level or segment-level feature data, and then comparing the features. This method has high computational complexity and is highly dependent on the overall structure of the video content, making it difficult to accurately reflect local quality issues. The local matching method mainly focuses on the matching of local segments in the video stream, but it often requires setting more parameters and is difficult to achieve universality across scenarios. Based on the above reasons, the present invention proposes a single-point matching method for an OTT quality detection pattern matching system. Summary of the invention

[0004] 1. Technical issues to be solved

[0005] In view of the shortcomings of the prior art, the present invention can more accurately detect service quality problems and improve matching accuracy by performing single-point matching on OTT service data, and because the single-point matching method is relatively simple and has a small amount of calculation, a large amount of OTT service data can be quickly detected to improve detection efficiency; the present invention constructs a single-point matching model for specific OTT service quality features of bit rate detection of OTT video streams, quality detection of OTT audio streams, and quality detection of OTT audio streams, thereby realizing different quality detection modes and having high flexibility.

[0006] (II) Technical solution

[0007] To achieve the above object, the present invention provides the following technical solution: a single-point matching method for an OTT quality detection pattern matching system, comprising the steps of:

[0008] S1: Data collection, collecting key data from the OTT business system, including but not limited to video stream data, audio stream data, and network data packets;

[0009] S2: Data preprocessing, preprocessing the collected data, including data cleaning, format conversion, and outlier processing to ensure the accuracy and consistency of the data and extract key features from the collected data;

[0010] S3: Construction of a single-point matching model. Based on the preprocessed data, a single-point matching model is constructed. The model can identify and match specific OTT service quality features, thereby achieving accurate assessment of service quality.

[0011] S4: Single-point matching: Use the constructed single-point matching model to perform matching analysis on the real-time collected data. By comparing the feature points in the data with the feature points in the model, determine whether there are any abnormalities in the service quality.

[0012] S5: Result evaluation: evaluate the quality of the OTT service based on the results of the single-point matching, and output a report containing the OTT service quality analysis results.

[0013] Preferably, the data collection in step S1 can be achieved by means of network monitoring tools, log analysis tools, and performance monitoring software to ensure the real-time and accuracy of the data.

[0014] Preferably, the construction of the single point matching model in step S3 includes the following steps:

[0015] S11: Define specific OTT service quality characteristics in the quality detection mode;

[0016] S12: extracting features of specific OTT service quality features to be detected;

[0017] S13: Perform single-point matching calculation.

[0018] Preferably, the specific OTT service quality features include bit rate detection of OTT video stream, quality detection of OTT audio stream and detection of OTT video picture quality.

[0019] Preferably, the step S4 uses the constructed single-point matching model to perform matching analysis on the real-time collected data, including the following steps:

[0020] S21: Real-time data collection: Use data collection tools to collect data from OTT business systems in real time;

[0021] S22: Feature point extraction: extract feature points from the real-time collected data and compare them with the feature points in the single-point matching model;

[0022] S23: Matching analysis: Based on the comparison results of the feature points, determine whether there is any abnormality in the service quality.

[0023] Preferably, extracting feature points from the real-time collected data in step S22 includes the following steps:

[0024] S221: Data processing, performing denoising and contrast enhancement operations on the data collected in real time to improve the accuracy and stability of feature points;

[0025] S222: feature point detection, using an algorithm to detect the processed data to further determine the accuracy of the feature points;

[0026] S223: Feature point description: describing the detected feature points and generating feature vectors.

[0027] Preferably, the feature point extraction algorithm in step S222 includes a SURF (Speeded-Up Robust Feature) algorithm or an ORB (Oriented FAST and Rotated BRIEF) algorithm.

[0028] Preferably, if the single-point matching result in step S5 meets the requirements of the quality detection mode, the OTT service quality is considered to be good; otherwise, the OTT service is considered to have quality problems.

[0029] (III) Beneficial effects

[0030] Compared with the prior art, the present invention provides a single-point matching method for an OTT quality detection pattern matching system, which has the following beneficial effects: the present invention can more accurately detect service quality problems and improve matching accuracy by performing single-point matching on OTT service data, and because the single-point matching method is relatively simple and has a small amount of calculation, a large amount of OTT service data can be quickly detected, thereby improving detection efficiency; the present invention constructs a single-point matching model for specific OTT service quality features of bit rate detection of OTT video streams, quality detection of OTT audio streams, and quality detection of OTT audio streams, thereby realizing different quality detection modes and having high flexibility. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 The figure is a flow chart of the single point matching method of the present invention. DETAILED DESCRIPTION

[0032] In order to better understand the purpose, structure and function of the present invention, and to improve the speed and versatility of OTT quality detection, the present invention can more accurately detect service quality problems and improve matching accuracy by performing single-point matching on OTT service data, and because the single-point matching method is relatively simple and has a small amount of calculation, a large amount of OTT service data can be quickly detected to improve detection efficiency; the present invention constructs a single-point matching model for the bit rate detection of OTT video streams, the quality detection of OTT audio streams, and the specific OTT service quality features of the quality detection of OTT audio streams, so that different quality detection modes can be achieved, which has high flexibility. A single-point matching method for an OTT quality detection pattern matching system of the present invention is further described in detail.

[0033] refer to Figure 1 The present invention: a single-point matching method for an OTT quality detection pattern matching system, comprising the steps of:

[0034] S1: Data collection: collect key data from the OTT business system, including but not limited to video stream data, audio stream data, and network data packets. Data collection can be achieved through network monitoring tools, log analysis tools, and performance monitoring software to ensure the real-time and accuracy of the data.

[0035] S2: Data preprocessing, preprocessing the collected data, including data cleaning, format conversion, and outlier processing to ensure the accuracy and consistency of the data and extract key features from the collected data;

[0036] S3: Construction of a single-point matching model. Based on the preprocessed data, a single-point matching model is constructed. The model can identify and match specific OTT service quality features, thereby achieving accurate assessment of service quality.

[0037] Specifically, when the specific OTT service quality feature is the bit rate detection of the OTT video stream, the construction of the single-point matching model includes the following steps:

[0038] S101: defining a specific OTT service quality feature in a quality detection mode as a bit rate detection of an OTT video stream, and setting a reasonable range of a bit rate and a reasonable range of a network delay in the quality detection mode;

[0039] S102: obtaining the actual bit rate and network delay values ​​of the current video stream from the collected OTT data by parsing the video stream data packets or the like;

[0040] S103: Perform single-point matching calculation. If all the numerical features to be detected meet the numerical range requirements in the corresponding quality detection mode, the single-point matching is successful, and it is determined that the OTT service quality meets the requirements at this point.

[0041] When the specific OTT service quality feature is the quality detection of the OTT audio stream, the construction of the single-point matching model includes the following steps:

[0042] S111: define a specific OTT service quality feature in the quality detection mode as quality detection of the OTT audio stream, and set the range of the audio sampling rate, the number of channels, and the audio code format;

[0043] S112: extracting the sampling rate, the number of channels and the audio encoding format from the audio stream data;

[0044] S113: Perform single-point matching calculation and logic judgment. Only when all the logic judgment results are true, the single-point matching of this combined feature is judged to be successful, that is, the audio OTT service is judged to meet the requirements at this quality detection point.

[0045] When the specific OTT service quality feature is the detection of OTT video picture quality, the construction of the single-point matching model includes the following steps:

[0046] S121: defining a specific OTT service quality feature in the quality detection mode as detection of OTT video picture quality, and constructing a reference feature vector by color distribution, contrast, and brightness features of the high-quality video picture;

[0047] S122: extracting features such as color distribution, contrast, brightness, etc. from the currently detected video frame through image analysis technology to form a feature vector to be detected;

[0048] S123: Perform single-point matching calculation, use the Euclidean distance formula to calculate the distance between two vectors to measure the similarity, and set a threshold. When the distance between the two vectors is less than the set threshold, it is considered that the single-point matching of the video picture quality is successful, that is, it is determined that the current video picture quality meets the requirements of the quality detection mode at this detection point.

[0049] S4: Single-point matching: Use the constructed single-point matching model to perform matching analysis on the real-time collected data. By comparing the feature points in the data with the feature points in the model, determine whether there are any abnormalities in the service quality.

[0050] Specifically, using the constructed single-point matching model, matching analysis is performed on the real-time collected data, including the following steps:

[0051] S21: Real-time data collection: Use data collection tools to collect data from OTT business systems in real time;

[0052] S22: Feature point extraction: extract feature points from the real-time collected data and compare them with the feature points in the single-point matching model;

[0053] Furthermore, extracting feature points from the real-time collected data includes the following steps:

[0054] S221: Data processing, performing denoising and contrast enhancement operations on the data collected in real time to improve the accuracy and stability of feature points;

[0055] S222: feature point detection, using an algorithm to detect the processed data to further determine the accuracy of the feature points. The feature point extraction algorithm includes a SURF (Speeded-Up Robust Feature) algorithm or an ORB (Oriented FAST and Rotated BRIEF) algorithm;

[0056] S223: Feature point description: describing the detected feature points and generating feature vectors.

[0057] S23: Matching analysis: Based on the comparison results of the feature points, determine whether there is any abnormality in the service quality.

[0058] S5: Result evaluation: evaluate the quality of the OTT service based on the results of the single-point matching, and output a report containing the results of the OTT service quality analysis. If the single-point matching results meet the requirements of the quality detection mode, the OTT service quality is considered to be good; otherwise, the OTT service is considered to have quality problems.

[0059] Compared with the prior art, the present invention provides a single-point matching method for an OTT quality detection pattern matching system, which has the following beneficial effects: the present invention can more accurately detect service quality problems and improve matching accuracy by performing single-point matching on OTT service data, and because the single-point matching method is relatively simple and has a small amount of calculation, a large amount of OTT service data can be quickly detected, thereby improving detection efficiency; the present invention constructs a single-point matching model for specific OTT service quality features of bit rate detection of OTT video streams, quality detection of OTT audio streams, and quality detection of OTT audio streams, thereby realizing different quality detection modes and having high flexibility.

[0060] It is to be understood that the present invention is described by some embodiments, and it is known to those skilled in the art that various changes or equivalent substitutions may be made to these features and embodiments without departing from the spirit and scope of the present invention. In addition, under the teachings of the present invention, these features and embodiments may be modified to adapt to specific circumstances and materials without departing from the spirit and scope of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed herein, and all embodiments falling within the scope of the claims of this application are within the scope of protection of the present invention.

Claims

1. A single-point matching method for an OTT quality detection pattern matching system, characterized in that: The steps include: S1: Data collection, collecting key data from the OTT business system, including but not limited to video stream data, audio stream data, and network data packets; S2: Data preprocessing, preprocessing the collected data, including data cleaning, format conversion, and outlier processing to ensure the accuracy and consistency of the data and extract key features from the collected data; S3: Construction of a single-point matching model. Based on the preprocessed data, a single-point matching model is constructed. The model can identify and match specific OTT service quality features, thereby achieving accurate assessment of service quality. S4: Single-point matching: Use the constructed single-point matching model to perform matching analysis on the real-time collected data. By comparing the feature points in the data with the feature points in the model, determine whether there are any abnormalities in the service quality. S5: Result evaluation: evaluate the quality of the OTT service based on the results of the single-point matching, and output a report containing the OTT service quality analysis results.

2. According to claim 1, a single-point matching method for an OTT quality detection pattern matching system is characterized in that: The data collection in step S1 can be achieved by means of network monitoring tools, log analysis tools, and performance monitoring software to ensure the real-time and accuracy of the data.

3. According to claim 1, a single-point matching method for an OTT quality detection pattern matching system is characterized in that: The construction of the single point matching model in step S3 includes the following steps: S11: Define specific OTT service quality characteristics in the quality detection mode; S12: extracting features of specific OTT service quality features to be detected; S13: Perform single-point matching calculation.

4. The single-point matching method for an OTT quality detection pattern matching system according to claim 3, characterized in that: The specific OTT service quality features include bit rate detection of OTT video stream, quality detection of OTT audio stream and detection of OTT video picture quality.

5. The single-point matching method for an OTT quality detection pattern matching system according to claim 1, characterized in that: In step S4, the constructed single-point matching model is used to perform matching analysis on the real-time collected data, including the following steps: S21: Real-time data collection: Use data collection tools to collect data from OTT business systems in real time; S22: Feature point extraction: extract feature points from the real-time collected data and compare them with the feature points in the single-point matching model; S23: Matching analysis: Based on the comparison results of the feature points, determine whether there is any abnormality in the service quality.

6. A single-point matching method for an OTT quality detection pattern matching system according to claim 5, characterized in that: The step S22 extracts feature points from the real-time collected data, including the following steps: S221: Data processing, performing denoising and contrast enhancement operations on the data collected in real time to improve the accuracy and stability of feature points; S222: feature point detection, using an algorithm to detect the processed data to further determine the accuracy of the feature points; S223: Feature point description: describing the detected feature points and generating feature vectors.

7. A single-point matching method for an OTT quality detection pattern matching system according to claim 6, characterized in that: The feature point extraction algorithm in step S222 includes a SURF (Speeded-Up Robust Feature) algorithm or an ORB (Oriented FAST and Rotated BRIEF) algorithm.

8. The single-point matching method for an OTT quality detection pattern matching system according to claim 1, characterized in that: In step S5, if the single-point matching result meets the requirements of the quality detection mode, it is considered that the OTT service quality is good; otherwise, it is considered that the OTT service has quality problems.

Images