Monitoring analysis method and analysis system for set top box application

Abstract

The invention relates to the technical field of television systems, in particular to a monitoring analysis method and analysis system for a set top box application. Comprising a signal acquisition module, a real-time video feature extraction module, a standby signal source storage module and a seamless switching module. According to the mapping relation between the video features and the watching quality indexes, on one hand, switching of the signal source is facilitated when the watching experience is not good, switching to the standby signal source of weighted sorting is achieved through the buffer area technology and the seamless switching algorithm, stable switching is achieved, the interruption feeling caused by signal switching is reduced, meanwhile, the signal source of the cloud end is called out, and the watching experience is improved. On the other hand, the signal quality index is verified through the video features, the quality index is reflected through double indexes, the accuracy of quality index identification is ensured, early warning reminding is carried out in time when the signal acquisition module determines that the signal quality index is inaccurate, and the sustainable use effect of the monitoring set top box is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of television systems, and in particular to a monitoring and analyzing method and an analyzing system for set-top box applications. Background Art

[0002] With the popularity of smart TVs and online video services, set-top boxes are important devices that connect users to digital content. In modern home entertainment, set-top boxes are important terminal devices that receive and play TV programs. However, signal quality monitoring and user experience optimization of set-top boxes have always been technical difficulties. Traditional signal quality monitoring methods usually only focus on a single indicator, such as signal strength or bit error rate, while ignoring other key factors such as frame loss rate and network delay, resulting in incomplete signal quality assessment. In actual use, due to signal interference, weather conditions or other factors, the set-top box may receive unstable signals, resulting in problems such as image freezes and audio and video asynchrony. Although the existing technology can solve some problems by manually switching channels or restarting the device, this method is inefficient and provides a poor user experience. In addition, existing systems often lack effective correlation and optimization mechanisms when processing multi-source signals, resulting in unstable user experience. The operating status and user experience quality become particularly important. In view of this, we propose a monitoring and analysis method and analysis system for set-top box applications. Summary of the invention

[0003] The purpose of the present invention is to provide a monitoring and analysis method and analysis system for set-top box applications to solve the problems of insufficient comprehensiveness of signal quality evaluation, weak correlation between signal sources and user experience, unintelligent management of backup signal sources, and low efficiency of real-time monitoring and early warning.

[0004] In order to solve the above technical problems, one of the objectives of the present invention is to provide a monitoring and analysis method for set-top box applications, comprising the following steps: S1. Deploy a signal acquisition module in the set-top box to collect quality data of the signal source, define a mapping relationship between the quality data and the signal quality index, and calculate the signal quality index of the current signal source through normalization processing and weighted calculation; S2, triggering the video feature extraction technology to identify the content of the currently playing TV program, and outputting the video features in the video feature extraction, and feeding back to S1 according to the mapping relationship between the video features and the viewing quality indicators, and comparing the signal quality indicators with the viewing quality indicators. If they do not match, a device failure warning signal is issued. If they match, the quality indicators are output, including poor viewing experience, good viewing experience, and excellent viewing experience; S3, search for signal sources that match the video features, use the technology in S1 to detect the quality of the signal sources, and perform weighted sorting and associate them to the cloud, defining them as backup signal sources.

[0005] S4, when sensing a quality indicator signal other than excellent viewing experience, switches to the backup signal source with the highest weighted ranking through buffer technology and seamless switching algorithm, and transmits it to S1 and S2 again to output the quality indicator. If a quality indicator signal other than excellent viewing experience is output again, the next weighted ranking backup signal source is called out until excellent viewing experience is output.

[0006] Preferably, the quality data in S1 includes signal strength, bit error rate, frame loss rate and network delay; The signal strength is obtained through an API of a modem or a network interface card, and indicates the strength of a received radio signal; The bit error rate is obtained through the statistical information of the network equipment, which indicates the ratio of the number of erroneous bits during data transmission to the total number of transmitted bits; the frame loss rate comes from the feedback of the video decoder, which indicates the ratio of the number of frames lost during video transmission to the total number of transmitted frames; the network delay is obtained through the ping command, which indicates the transmission time of data from the source to the destination.

[0007] Preferably, the video feature extraction technology in S2 comprises the following steps: Audio stream extraction: Real-time capture of the audio stream being played, extraction of audio amplitude, zero-crossing rate and other features, reflecting the dynamic changes of the audio, Fourier transform, analysis of spectrum, peak frequency, bandwidth features, reflecting the frequency distribution of the audio, MFCC extraction of speech features, generation of a unique audio identifier for subsequent matching of the same program content; Video stream extraction: Capture the playing video stream in real time, use the OpenCV library to extract the key frames of the video, use the pre-trained deep learning model to perform image recognition on the key frames, identify the video content, generate a unique identifier for the video content, and combine it with the audio fingerprint to more accurately identify the program content.

[0008] Preferably, the mapping relationship between the video features and the viewing quality indicators in S2 includes the following steps: By analyzing the changes in the zero-crossing rate of the audio stream and comparing the synchronization of audio and video, the stuttering result is output. If an audio stream breakpoint is detected and the audio and video are not synchronized, the viewing experience is poor. If no audio stream breakpoint is detected and the audio and video are not synchronized, or the audio stream breakpoint is detected and the audio and video are synchronized, the viewing experience is good. If no audio stream breakpoint is detected and the audio and video are synchronized, the viewing experience is excellent.

[0009] Preferably, searching for a signal source matching the video feature in S3 comprises the following steps: According to the signal source types supported by the set-top box, perform corresponding scanning to obtain a signal source list; And adopt the feature matching algorithm to verify the matching degree between the video features of the program played by the signal source and the video features in S2. If they match, the backup signal source is output; if not, the signal source is searched continuously; Perform quality checks on multiple backup signal sources, measure the strength of the RF signal (such as dBm), calculate the error rate in data transmission, detect frame loss in the video stream, measure the delay time of signal transmission, and again calculate the signal quality index of the current signal source through normalization and weighting.

[0010] Preferably, the weighted sorting in S3 comprises the following steps: The receiving quality index is normalized to multiple backup signal sources of the same range, and the comprehensive signal quality index is calculated according to the predefined weight allocation, and the signal quality index of each backup signal source is detected, and the signal sources are sorted from high to low according to the signal quality index, and the backup signal source with the highest priority is arranged at the front; The sorted backup signal source list is converted into a format for storage, a cloud storage service is used to store the signal source list, and the signal source list stored in the cloud is regularly updated.

[0011] Preferably, the step S4 uses the buffer technology and the seamless switching algorithm to switch to the backup signal source with the highest weighted ranking, including the following steps: Dynamically adjust the size of the buffer according to the network bandwidth and signal quality, preload a certain amount of video data in the buffer, and reduce the size of the buffer if the signal quality indicator in S1 is appropriate; If the signal quality indicator received in S1 is not appropriate, the size of the buffer is increased immediately. When a quality indicator signal other than excellent viewing experience is sensed, a command to switch the signal source is output. The switching logic will pause the playback of the current signal source and continue to play the pre-loaded content from the buffer, and start loading the video data of the backup signal source into the new buffer in the background; Extract the video stream features of the current signal source and the backup signal source, including frame rate, video resolution, and audio synchronization. Through the feature matching algorithm, verify the consistency of the backup signal source with the current playback content and find the natural switching point in the video stream.

[0012] Preferably, S4 also includes a re-output limiting algorithm, which is used to dynamically preset a threshold of the number of switching times of the backup signal source, and output a device warning signal when the number of switching times is equal to the threshold.

[0013] The second object of the present invention is to provide a monitoring and analysis system for set-top box applications, including any one of the monitoring and analysis methods for set-top box applications described above, including a signal acquisition module, a real-time video feature extraction module, a backup signal source storage module and a seamless switching module; The signal acquisition module is used to collect quality data of the signal source, define the mapping relationship between the quality data and the signal quality index, and calculate the signal quality index of the current signal source through normalization processing and weighted calculation; The real-time video feature extraction module is used to trigger the video feature extraction technology to identify the content of the currently playing TV program, and output the video features in the video feature extraction, and feed back to the signal acquisition module according to the mapping relationship between the video features and the viewing quality index, and compare the signal quality index and the viewing quality index. If they do not match, an equipment failure warning signal is issued, and if they match, the quality index is output; The backup signal source storage module is used to search for signal sources that match video features, detect the quality of the signal sources through the signal acquisition module, and perform weighted sorting and associate them to the cloud, defining them as backup signal sources, and storing the backup signal sources that match video features through the cloud.

[0014] The seamless switching module is used to switch to the backup signal source with the highest weighted ranking through buffer technology and seamless switching algorithm when sensing a quality indicator signal other than excellent viewing experience, and transmit it to the signal acquisition module and the real-time video feature extraction module again to output the quality indicator. If a quality indicator signal other than excellent viewing experience is output again, the next weighted ranking backup signal source is called out until excellent viewing experience is output.

[0015] Compared with the prior art, the present invention has the following beneficial effects: The video feature extraction technology is used to identify the content of the currently playing TV program, and the video features in the video feature extraction are output. According to the mapping relationship between the video features and the viewing quality indicators, on the one hand, it is convenient to subsequently match the signal source of the same program, which is conducive to switching the backup signal source stored in the cloud when the viewing experience is not good, saving time and avoiding the time occupied by searching for the signal source after the poor experience is perceived. At the same time, the signal source is weighted and sorted, which is convenient for the subsequent call-out of the backup signal source according to the cycle, giving priority to the call-up of the best quality, reducing the operating intensity caused by frequent replacement, and switching to the weighted sorted backup signal source through the buffer technology and seamless switching algorithm to achieve stable switching of the backup signal source and reduce the interruption caused by signal switching. On the other hand, the signal quality indicator is verified by video features, and the quality indicator is reflected by dual indicators, which not only ensures the accuracy of quality indicator identification, but also when the signal acquisition module determines that the signal quality indicator is inaccurate, timely early warning reminders are issued to improve the sustainable use effect of the monitoring set-top box. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 This is an overall flow chart of Example 1; Figure 2 This is a flow chart of the re-output restriction algorithm of Example 1. DETAILED DESCRIPTION

[0017] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0018] Example 1 like Figure 1-Figure 2 As shown, one of the purposes of the present invention is to provide a monitoring and analysis method for set-top box applications, comprising the following steps: S1. Deploy a signal acquisition module in the set-top box to collect quality data of the signal source, and define the mapping relationship between the quality data and the signal quality index. Calculate the signal quality index of the current signal source through normalization processing and weighted calculation. Because the quality of the collected signal source is multi-sourced, including quality indicators such as signal strength, bit error rate, frame loss rate and network delay, it is more accurate to obtain the final signal quality index by combining the above data. Specifically, the quality data in S1 includes signal strength, bit error rate, frame loss rate, and network delay; The signal strength is obtained through the API of the modem or network interface card and indicates the strength of the received radio signal; The bit error rate is obtained through the statistical information of the network device, which indicates the ratio of the number of bits that have errors during data transmission to the total number of transmitted bits; the frame loss rate comes from the feedback of the video decoder, which indicates the ratio of the number of frames lost during video transmission to the total number of transmitted frames; the network delay is obtained through the ping command or similar network delay measurement tools, which indicates the transmission time of data from source to destination in milliseconds; Each signal quality indicator is converted to a range of 0 to 1 for uniform weighted calculation. Assuming the signal strength ranges from -100 dBm to -50 dBm (typical value), the normalization formula is: , so the signal strength value is 0 at -100 dBm and 1 at -50 dBm; assuming the bit error rate ranges from 0 to 1%, the normalized formula is: , so that the bit error rate is 1 when it is 0, and 0 when the bit error rate is 1%; assuming that the frame loss rate ranges from 0 to 5%, the normalized formula is: , so that the frame loss rate is 1 when it is 0, and 0 when it is 5%; assuming that the network delay ranges from 0 to 200 milliseconds, the normalized formula is: , so that the network delay is 0 when it is 1, and the network delay is 200 milliseconds when it is 0. According to the influence of each signal quality indicator on the overall signal quality, the weight is assigned. Assume that the weights are as follows: signal strength is 0.3, bit error rate is 0.4, frame loss rate is 0.2, and network delay is 0.1. The normalized indicator value is multiplied by the corresponding weight, and then summed to obtain the comprehensive signal quality index (SQI). Finally, the range of SQI is 0 to 1. The closer the value is to 1, the better the signal quality. Normalization and weighted calculation ensure fair comparison of different indicators. The weight allocation reflects the influence of each indicator on the overall signal quality, which helps to improve the stability of set-top box applications and user experience.

[0019] S2, triggering the video feature extraction technology to identify the content of the currently playing TV program, and outputting the video features in the video feature extraction, and feeding back to S1 according to the mapping relationship between the video features and the viewing quality indicators, comparing the signal quality indicators and the viewing quality indicators, if they do not match, issuing a device failure warning signal, if they match, outputting the quality indicators, including poor viewing experience, good viewing experience and excellent viewing experience. The video features facilitate the subsequent matching of the signal source of the same program, which is conducive to switching the signal source when the viewing experience is not excellent, and is conducive to the user's experience of watching TV. On the other hand, the signal quality indicators are verified by the video features, and the dual indicators are realized to reflect the quality indicators, which not only ensures the accuracy of the quality indicator identification, but also when the signal acquisition module determines that the signal quality indicator is inaccurate, timely warning reminders are issued to improve the sustainable use effect of the monitoring set-top box; In order to monitor the effect of TV programs in real time when they are actually used, the video feature extraction technology in S2 includes the following steps: Audio stream extraction: Real-time capture of the audio stream being played, extraction of audio amplitude, zero-crossing rate and other features, reflecting the dynamic changes of the audio, Fourier transform, analysis of spectrum, peak frequency, bandwidth features, reflecting the frequency distribution of the audio, and use MFCC (Mel-Frequency Cepstral Coefficient) to extract voice features, which is suitable for identifying voice content and generating a unique audio identifier for subsequent matching of the same program content; Video stream extraction: Capture the playing video stream in real time, use the OpenCV library to extract the key frames of the video to reduce the amount of data for subsequent processing, use pre-trained deep learning models (such as ResNet, Yolo) to perform image recognition on the key frames, identify the video content, generate a unique identifier for the video content, and combine it with audio fingerprints to more accurately identify the program content.

[0020] Furthermore, the mapping relationship between the video features and the viewing quality indicators in S2 includes the following steps: By analyzing the changes in the zero-crossing rate of the audio stream and comparing the synchronization of the audio and video, the stuttering results are output. If an audio stream breakpoint is detected and the audio and video are not synchronized, the viewing experience is poor. If it is detected that the audio stream has no breakpoint and the audio and video are not synchronized, or the audio stream breakpoint and the audio and video are synchronized, the viewing experience is good. If it is detected that the audio stream has no breakpoint and the audio and video are synchronized, the viewing experience is excellent. The viewing experience can be verified by both the audio and video streams at the same time, thereby improving the accuracy of the verified viewing experience in actual conditions.

[0021] S3, search for signal sources that match video features, use the technology in S1 to detect the quality of signal sources, and perform weighted sorting and associate them to the cloud, defining them as backup signal sources. Backup signal sources that match video features (for the same program) are stored in the cloud, making it easy to call them up at any time, saving time and avoiding the time spent searching for signal sources after perceiving a poor experience. At the same time, weighted sorting of signal sources facilitates subsequent call-out of backup signal sources based on a cycle, giving priority to those with good quality, and reducing the operating intensity caused by frequent changes; The searching for the signal source matching the video feature in S3 includes the following steps: According to the signal source types supported by the set-top box, implement corresponding scanning to obtain a list of signal sources. Signal source types include cable TV signals, satellite TV signals, IPTV (Internet TV) and local media files (such as videos in USB storage devices). IPTV signal source scanning uses IGMP (Internet Group Management Protocol) or UPnP (Universal Plug and Play) protocols to scan IPTV signal sources within the local area network. By cooperating with IPTV service providers, a list of available signal sources can be obtained. Cable and satellite signal source scanning uses the set-top box's tuner to automatically scan available cable or satellite channels, detect signal strength and quality, obtain an electronic program guide (EPG), and identify available TV program content. Local media signal source scanning identifies available video files by scanning connected storage devices (such as USB, hard disk), and integrates third-party media library services to obtain video content stored locally or on the network. A feature matching algorithm is used to verify the matching degree between the video features of the program played by the signal source and the video features in S2. If they match, the backup signal source is output. If they do not match, the signal source is searched again to ensure that the content played by the backup signal source is consistent with the current program, avoiding the problem of inconsistent content after switching the signal source, and improving the user experience. Perform quality checks on multiple backup signal sources, measure the strength of the RF signal (such as dBm), calculate the error rate in data transmission, detect frame loss in the video stream, measure the delay time of signal transmission, and again calculate the signal quality index of the current signal source through normalization and weighted calculation, so that each quality index is standardized to the same range (such as 0-1) for subsequent calculations. For example, signal strength = (current value - minimum value) / (maximum value - minimum value), bit error rate = 1 - (current value / maximum value), which is the same as in S1 and will not be repeated here.

[0022] Next, weighted sorting is performed in S3, including the following steps: The receiving quality indicators are normalized to multiple backup signal sources in the same range. According to the predefined weight allocation, the comprehensive signal quality indicator is calculated, and the signal quality indicator of each backup signal source is detected. The signal sources are sorted from high to low according to the signal quality indicators. The backup signal source with the highest priority is ranked first. The signal source with higher quality is selected first to ensure the best viewing effect and reduce freezes and interruptions. The sorted list of backup signal sources is converted into a format for storage (such as JSON or CSV), and the list of signal sources is stored in a cloud storage service (such as AWS S3, Google Cloud Storage) to ensure data security and privacy (such as encrypted transmission and storage), and the list of signal sources stored in the cloud is updated regularly (such as once an hour). When a change in the quality of the signal source is detected, the cloud data is updated in a timely manner, and the list of backup signal sources is stored in the cloud, which is convenient for subsequent quick calls, saves search time, and improves switching efficiency. In particular, when the quality of the current signal source decreases, it is possible to quickly switch to the backup signal source to reduce the sense of interruption. At the same time, the cloud list is updated regularly to ensure that the quality information of the signal source is the latest, and the changes in the signal source are reflected in a timely manner to avoid viewing problems caused by fluctuations in the quality of the signal source; In summary, integrating third-party media library services, expanding the ability to identify and acquire local media, enriching the available signal sources, and improving the versatility of the set-top box, can ensure stable viewing by switching to backup signal sources in a timely manner when the signal source quality degrades, is unavailable, or fluctuates. This is suitable for a variety of usage scenarios and improves the reliability of the set-top box and user experience.

[0023] S4, when sensing a quality indicator signal other than excellent viewing experience, switches to the backup signal source with the highest weighted ranking through buffer technology and seamless switching algorithm, and transmits it to S1 and S2 again to output the quality indicator. If a quality indicator signal other than excellent viewing experience is output again, the next weighted ranking backup signal source is called out until the excellent viewing experience is output. This not only achieves stable switching of backup signal sources and reduces the sense of interruption caused by signal switching, but also the signal quality indicator does not represent the actual viewing experience when weighted ranking. Therefore, re-determining the quality indicator through the dual indicators of video features and quality data is conducive to a better experience of the program output by the set-top box. At the same time, there is no need to run the search again, only the cloud needs to be called out, which saves switching time and improves the experience. In set-top box applications, in order to achieve seamless switching of signal sources, when the signal quality decreases, it is possible to quickly and smoothly switch to the backup signal source to ensure the user's viewing experience. The buffer technology and seamless switching algorithm in S4 are used to switch to the backup signal source with the highest weighted ranking, including the following steps: Dynamically adjust the size of the buffer according to the network bandwidth and signal quality, preload a certain amount of video data (such as 10 seconds to 30 seconds) in the buffer, and if the signal quality indicator in the received S1 is appropriate, reduce the size of the buffer to save storage resources; If the signal quality indicator received in S1 is not appropriate, the size of the buffer is increased immediately to cope with possible delays. When a quality indicator signal other than excellent viewing experience is sensed, a command to switch the signal source is output. The switching logic will pause the playback of the current signal source and continue to play the pre-loaded content from the buffer, and start loading the video data of the backup signal source to the new buffer in the background; Extract the video stream features of the current signal source and the backup signal source, including frame rate, video resolution, and audio synchronization. Through the feature matching algorithm, verify the consistency of the backup signal source with the current playback content, and find natural switching points in the video stream, such as scene switching, still frames, or blackouts. The switching points are used as the opportunity for seamless switching to avoid switching in dynamic images, which may cause image freezes or tearing. Specifically, the switching points are found by analyzing the inter-frame differences of the video stream. When the frame difference exceeds the preset threshold, it is judged as a switching point. Through buffer technology and seamless switching algorithms, the continuity and smoothness of video playback are ensured, and freezes and interruptions are reduced. The buffer size and switching strategy are dynamically adjusted to adapt to different network bandwidths and signal qualities. When signal quality problems are detected, quickly switch to the backup signal source to reduce user waiting time. At the same time, the use of network resources is reduced by preloading and dynamically managing the buffer.

[0024] In addition, specific Figure 2As shown, S4 also includes a re-output restriction algorithm, which is used to dynamically preset the number threshold of switching backup signal sources. When the number of perceived switching is equal to the number threshold, a device warning signal is output. If the viewing experience of the program cannot be optimized after multiple switching of the backup signal source, it means that there may be a problem with the TV device itself, which is unrelated to the signal received by the set-top box. Therefore, a device warning signal is issued to facilitate users to repair the TV device in time. At the same time, the number threshold is dynamically adjusted. The switching number threshold can be adjusted according to the life of the TV device. As the life is shorter, that is, the longer the usage time (the greater the probability of damage to the TV device), the number threshold is smaller, so as to avoid multiple adjustments still having card segments, causing the set-top box to run larger. On the contrary, as the life is longer, the number threshold is larger, so as to avoid the backup signal source being used on this TV and the actual playback status being affected by external factors (such as distance, weather, etc.); On the contrary, if the number of perceived switching times is less than the number threshold, the switching command continues to be executed until the output viewing experience is excellent.

[0025] The second object of the present invention is to provide a monitoring and analysis system for set-top box applications, including any one of the above monitoring and analysis methods for set-top box applications, including a signal acquisition module, a real-time video feature extraction module, a backup signal source storage module and a seamless switching module; The signal acquisition module is used to collect the quality data of the signal source, define the mapping relationship between the quality data and the signal quality index, and calculate the signal quality index of the current signal source through normalization processing and weighted calculation; The real-time video feature extraction module is used to trigger the video feature extraction technology to identify the content of the currently playing TV program, and output the video features in the video feature extraction, and feed back to the signal acquisition module according to the mapping relationship between the video features and the viewing quality index, and compare the signal quality index with the viewing quality index. If they do not match, an equipment failure warning signal is issued, and if they match, the quality index is output; The backup signal source storage module is used to search for signal sources that match video features, detect the quality of the signal sources through the signal acquisition module, and perform weighted sorting and associate them with the cloud, defining them as backup signal sources. The backup signal sources that match video features are stored in the cloud.

[0026] When the seamless switching module senses a quality indicator signal other than excellent viewing experience, it switches to the backup signal source with the highest weighted ranking through buffer technology and seamless switching algorithm, and transmits it to the signal acquisition module and the real-time video feature extraction module to output the quality indicator again. If a quality indicator signal other than excellent viewing experience is output again, the next weighted ranking backup signal source is called out until excellent viewing experience is output.

[0027] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. The above embodiments and descriptions are only preferred examples of the present invention and are not intended to limit the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A monitoring and analysis method for set-top box applications, characterized in that: The following steps are involved: S1. Deploy a signal acquisition module in the set-top box to collect quality data of the signal source, define a mapping relationship between the quality data and the signal quality index, and calculate the signal quality index of the current signal source through normalization processing and weighted calculation; S2, triggering the video feature extraction technology to identify the content of the currently playing TV program, and outputting the video features in the video feature extraction, and feeding back to S1 according to the mapping relationship between the video features and the viewing quality indicators, and comparing the signal quality indicators with the viewing quality indicators. If they do not match, a device failure warning signal is issued. If they match, the quality indicators are output, including poor viewing experience, good viewing experience, and excellent viewing experience; S3, search for signal sources that match the video features, use the technology in S1 to detect the quality of the signal sources, and perform weighted sorting and associate them to the cloud, defining them as backup signal sources; S4, when sensing a quality indicator signal other than excellent viewing experience, switches to the backup signal source with the highest weighted ranking through buffer technology and seamless switching algorithm, and transmits it to S1 and S2 again to output the quality indicator. If a quality indicator signal other than excellent viewing experience is output again, the next weighted ranking backup signal source is called out until excellent viewing experience is output.

2. The monitoring and analysis method for set-top box applications according to claim 1, characterized in that: The quality data in S1 includes signal strength, bit error rate, frame loss rate and network delay; The signal strength is obtained through an API of a modem or a network interface card, and indicates the strength of a received radio signal; The bit error rate is obtained through the statistical information of the network equipment, which indicates the ratio of the number of erroneous bits during data transmission to the total number of transmitted bits; the frame loss rate comes from the feedback of the video decoder, which indicates the ratio of the number of frames lost during video transmission to the total number of transmitted frames; the network delay is obtained through the ping command, which indicates the transmission time of data from the source to the destination.

3. The monitoring and analysis method for set-top box application according to claim 2, characterized in that: The video feature extraction technology in S2 includes the following steps: Audio stream extraction: Real-time capture of the audio stream being played, extraction of audio amplitude, zero-crossing rate and other features, reflecting the dynamic changes of the audio, Fourier transform, analysis of spectrum, peak frequency, bandwidth features, reflecting the frequency distribution of the audio, MFCC extraction of speech features, generation of a unique audio identifier for subsequent matching of the same program content; Video stream extraction: Capture the playing video stream in real time, use the OpenCV library to extract the key frames of the video, use the pre-trained deep learning model to perform image recognition on the key frames, identify the video content, generate a unique identifier for the video content, and combine it with the audio fingerprint to more accurately identify the program content.

4. The monitoring and analysis method for set-top box application according to claim 3 is characterized in that: The mapping relationship between the video features and the viewing quality indicators in S2 includes the following steps: By analyzing the changes in the zero-crossing rate of the audio stream and comparing the synchronization of audio and video, the stuttering result is output. If an audio stream breakpoint is detected and the audio and video are not synchronized, the viewing experience is poor. If no audio stream breakpoint is detected and the audio and video are not synchronized, or the audio stream breakpoint is detected and the audio and video are synchronized, the viewing experience is good. If no audio stream breakpoint is detected and the audio and video are synchronized, the viewing experience is excellent.

5. The monitoring and analysis method for set-top box applications according to claim 1, characterized in that: The step of searching for a signal source matching the video feature in S3 includes the following steps: According to the signal source types supported by the set-top box, perform corresponding scanning to obtain a signal source list; And adopt the feature matching algorithm to verify the matching degree between the video features of the program played by the signal source and the video features in S2. If they match, the backup signal source is output; if not, the signal source is searched continuously; Perform quality checks on multiple backup signal sources, measure the strength of RF signals, calculate the error rate in data transmission, detect frame loss in video streams, measure the delay time of signal transmission, and again calculate the signal quality index of the current signal source through normalization and weighting.

6. The monitoring and analysis method for set-top box application according to claim 5, characterized in that: The weighted sorting in S3 includes the following steps: The receiving quality index is normalized to multiple backup signal sources of the same range, and the comprehensive signal quality index is calculated according to the predefined weight allocation, and the signal quality index of each backup signal source is detected, and the signal sources are sorted from high to low according to the signal quality index, and the backup signal source with the highest priority is arranged at the front; The sorted backup signal source list is converted into a format for storage, a cloud storage service is used to store the signal source list, and the signal source list stored in the cloud is regularly updated.

7. The monitoring and analysis method for set-top box applications according to claim 1, characterized in that: The S4 uses the buffer technology and the seamless switching algorithm to switch to the backup signal source with the highest weighted ranking, including the following steps: Dynamically adjust the size of the buffer according to the network bandwidth and signal quality, preload a certain amount of video data in the buffer, and reduce the size of the buffer if the signal quality indicator in S1 is appropriate; If the signal quality indicator received in S1 is not appropriate, the size of the buffer is increased immediately. When a quality indicator signal other than excellent viewing experience is sensed, a command to switch the signal source is output. The switching logic will pause the playback of the current signal source and continue to play the pre-loaded content from the buffer, and start loading the video data of the backup signal source into the new buffer in the background; Extract the video stream features of the current signal source and the backup signal source, including frame rate, video resolution, and audio synchronization. Through the feature matching algorithm, verify the consistency of the backup signal source with the current playback content and find the natural switching point in the video stream.

8. The monitoring and analysis method for set-top box applications according to claim 7, characterized in that: The S4 also includes a re-output limiting algorithm, which is used to dynamically preset a threshold of the number of switching backup signal sources, and output a device warning signal when the number of switching is equal to the threshold.

9. A monitoring and analysis system for set-top box applications, applied to the monitoring and analysis method for set-top box applications as claimed in any one of claims 1 to 8, characterized in that: It includes a signal acquisition module, a real-time video feature extraction module, a backup signal source storage module and a seamless switching module; The signal acquisition module is used to collect quality data of the signal source, define the mapping relationship between the quality data and the signal quality index, and calculate the signal quality index of the current signal source through normalization processing and weighted calculation; The real-time video feature extraction module is used to trigger the video feature extraction technology to identify the content of the currently playing TV program, and output the video features in the video feature extraction, and feed back to the signal acquisition module according to the mapping relationship between the video features and the viewing quality index, and compare the signal quality index and the viewing quality index. If they do not match, an equipment failure warning signal is issued, and if they match, the quality index is output; The backup signal source storage module is used to search for signal sources that match the video features, detect the quality of the signal sources through the signal acquisition module, and perform weighted sorting and associate them to the cloud, defining them as backup signal sources, and storing the backup signal sources that match the video features through the cloud; The seamless switching module is used to switch to the backup signal source with the highest weighted ranking through buffer technology and seamless switching algorithm when sensing a quality indicator signal other than excellent viewing experience, and transmit it to the signal acquisition module and the real-time video feature extraction module again to output the quality indicator. If a quality indicator signal other than excellent viewing experience is output again, the next weighted ranking backup signal source is called out until excellent viewing experience is output.

Images