Method, device, and program for managing quality of delivery service using video image quality evaluation function

Abstract

Provided is a playback service quality management method using a video quality evaluation function, characterized by comprising: a transmission step for transmitting, from a viewer receiver, all or part of data of a decoded frame of a currently viewed channel or VOD video to a video quality evaluator in a predetermined time interval; and an evaluation step in which a video image quality evaluator collects the frames according to a predetermined time interval and evaluates an image quality score; in order to evaluate image quality scores subjectively felt by human beings, the video image quality evaluator evaluates image quality by acquiring video data sets of multiple levels marked with image quality scores and learning a deep learning or machine learning model.

Description

Technical Field

[0001] The present invention relates to a method, device, and program for managing the quality of broadcast services using a video quality evaluation function. More specifically, it relates to a system capable of extracting / acquiring videos according to a predetermined or managed schedule, calculating a score using an artificial intelligence-based video quality evaluation function, and then reporting where quality problems occur along a video data stream path. Background Art

[0002] Video broadcasting services have evolved into various forms, including terrestrial television, cable television, satellite television, IPTV, and OTT (over the top) services. These services generally include live channel services and video on demand (VOD) services.

[0003] All delivery services are through Figure 6 This is achieved by following the basic steps shown.

[0004] The video content for each channel and VOD is typically provided by multiple content providers (CPs) and program providers (PPs). The video is integrated and delivered through the headend system of the system operator (SO). The network operator (NO) transmits the video stream to the viewer's receiver 17 through a transmission network consisting of its backbone network, regional network, and user network. The receiver can be a general set-top box or a variety of forms including mobile devices, televisions, etc. These CPs, PPs, SOs, and NOs can be combined in various forms and operated by multiple companies. This means that the operating entity at each stage can be an independent company.

[0005] The following patent documents disclose technologies for determining video quality by comparing with original videos.

[0006] [Prior art literature]

[0007] [Patent Document]

[0008] Patent Document 1: Korean Invention Patent No. 10-0944125 Summary of the Invention

[0009] Technical issues

[0010] The limitation of the above patent document technology is that the quality of the broadcast video can only be known when the original video is available. In reality, the original video cannot appear on the viewer's receiver 17, so there is a disadvantage that the actual video quality of the demander cannot be measured.

[0011] Typically, video streams are monitored at every stage of video transmission to SO delivery points to manage the quality of video broadcasting. However, any degradation in image quality that may occur at the delivery point is not monitored or evaluated.

[0012] Through evaluation Figure 6 The video quality at position ⑤ can be used to monitor the possible quality degradation at positions ①, ②, ③, and ④. However, it is not easy to determine the video stream at position ⑤.

[0013] The present invention aims to provide a technology capable of monitoring whether the image quality has degraded by using the video stream decoded (decompressed) at position ⑥.

[0014] In addition, the broadcasting service quality management system should have the following characteristics.

[0015] 1. Objective data can be used to evaluate the image quality experienced by the end-users of each broadcast service. This objective data can be achieved through deep learning or machine learning scoring models learned from various levels of video data.

[0016] 2. Since the video transmission of each broadcast service occurs in real time, the quality evaluation results should also be provided in real time. Real-time evaluation can be achieved by evaluating the video directly extracted from the viewer's receiver 17.

[0017] 3. Subjective evaluation of video quality as typically perceived by humans should be possible without human intervention. Subjective evaluation can be performed by comprehensively evaluating general characteristics, video production characteristics, compression characteristics, and transmission characteristics.

[0018] The present invention aims to provide a technology capable of managing the quality of the above-mentioned broadcasting services.

[0019] In addition, it is also better to collect decoded frames from the receiver to evaluate the image quality and store the results on the service quality management server.

[0020] 1. Video quality scores can be evaluated within the receiver, on external servers, on cloud servers, etc.

[0021] 2. Scores can be recorded separately by user, receiver, time, content and service.

[0022] 3. Through historical record management, receivers, user networks, regional networks and backbone networks that are below the average quality benchmark can be continuously identified and recorded.

[0023] In addition, an object is to provide a technology capable of distributing traffic load by controlling the video extraction operation of a video extractor according to a schedule managed by a quality management server.

[0024] Additionally, preferred incidents can use the distribution of recording scores to report statistics about vulnerable network points and problematic receivers. For example, if a particular NO backbone network is often associated with information from viewer receivers with low picture quality scores, it can be determined that the NO backbone network requires maintenance.

[0025] Technical Solution

[0026] The present invention solves the above-mentioned technical problems by providing a broadcast service quality management method utilizing a video quality evaluation function, characterized in that the method includes: a transmission step in which all or part of the decoded frames of a channel or VOD video currently being viewed are transmitted from a viewer receiver, which is a final receiver of the broadcast service, to a video quality evaluator in the broadcast service network via data communication at predetermined time intervals; and an evaluation step in which the video quality evaluator collects the frames and evaluates the quality scores; wherein, in order to evaluate the quality scores perceived by humans, the video quality evaluator evaluates the quality by acquiring a video dataset labeled with multiple levels of quality scores and learning a deep learning or machine learning model.

[0027] Here, video data can be collected by region and time.

[0028] In addition, it is preferable that the collected result data including the image quality score is transmitted to a service quality management server and recorded.

[0029] At this point, by performing statistical analysis on the data, the locations of the NO backbone network, NO regional network, NO user network, and viewer receivers can be identified.

[0030] In addition, if receivers connected to a specific NO backbone network, NO regional network, NO user network, or a network downstream of the viewer's receiver location consistently experience low image quality scores, the network at that location may be evaluated as having problems.

[0031] Additionally, it is possible to quantify the level of problems at each location of the NO backbone network, NO regional network, NO user network, or viewer receiver, and to periodically report statistical results by region and time.

[0032] On the other hand, to solve the above-mentioned technical problems, the apparatus of the present invention is a broadcast service quality management apparatus utilizing a video quality evaluation function, characterized in that it is implemented as including: a transmission module for transmitting all or part of the decoded frames of the channel or VOD video currently being watched from a viewer receiver, which is the final receiver of the broadcast service, to a video quality evaluator in the broadcast service network via data communication at predetermined time intervals; and an evaluation module for collecting the frames and evaluating the quality scores; wherein the video quality evaluator is implemented by acquiring a video data set labeled with a plurality of levels of quality scores and learning a deep learning or machine learning model in order to evaluate the quality scores perceived by humans subjectively.

[0033] On the other hand, in order to solve the above-mentioned technical problems, the program of the present invention is a program for an information device, recorded on a storage medium readable by the information device, and the storage medium records the program for the information device for executing the various steps of the method described in claim 1 on the information device.

[0034] Effects of the Invention

[0035] According to the present invention, there is provided a technology capable of monitoring whether image quality has deteriorated using a video stream decoded (decompressed) at position ⑥.

[0036] Furthermore, a broadcast service quality management system technology is provided that enables broadcast service quality management and has the following characteristics: 1. Objective data can be used to evaluate the image quality level experienced by the final receiver of each broadcast service; 2. Since video transmission for each broadcast service occurs in real time, quality evaluation results are also provided in real time; 3. Subjective evaluation of video quality, as typically perceived by humans, can be performed without human intervention.

[0037] Furthermore, decoded frames collected from receivers can be used to evaluate image quality, and the results can be stored on a service quality management server. This provides the following technology: 1. Video quality scores can be evaluated within the receiver, on an external server, or on a cloud server; 2. Scores can be recorded by user, receiver, time, content, and service; and 3. Through history management, technology can continuously identify and record receivers, user networks, regional networks, and backbone networks that fall below average quality standards.

[0038] In addition, there is provided a technology capable of distributing traffic load by controlling a video extracting operation of a video extractor according to a schedule managed by a quality management server.

[0039] Additionally, a technique is provided that can utilize the distribution of recording scores to report statistics on vulnerable network points and problematic receivers. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 is an exemplary block diagram of a hardware device for implementing a method according to an embodiment of the present invention.

[0041] Figure 2 is an example block diagram of a video extractor (module) installed in a viewer's receiver.

[0042] Figure 3 This is an example block diagram of an image quality evaluator.

[0043] Figure 4 is an example block diagram of a quality management server.

[0044] Figure 5 It is a time chart showing the operation flow.

[0045] Figure 6 This is an example block diagram of the basic structure of general broadcasting.

[0046] Description of Reference Numerals

[0047] 11: CP / PP delivery end

[0048] 12: SO headend

[0049] 13: SO delivery end

[0050] 14: NO backbone network

[0051] 15: NO regional network

[0052] 16: NO user network

[0053] 17: Viewer Receiver

[0054] 18: Video Player

[0055] 20: Video Extractor

[0056] 21: Video Acquisition Device

[0057] 22: Schedule Receiver

[0058] 23: Image quality evaluation interface

[0059] 30: Image Quality Evaluator

[0060] 31: Quality Management Server Interface

[0061] 32: Viewer Receiver API Server

[0062] 33: Video quality evaluation module

[0063] 40: Quality Management Server

[0064] 41: Management Module

[0065] 42: API Server

[0066] 43: Schedule Conveyor

[0067] 50: Database

[0068] 51: Operating Computer DETAILED DESCRIPTION

[0069] Hereinafter, the present invention will be described in detail with reference to the accompanying drawings. However, for components having the same configuration and the same function, detailed description may be omitted by retaining the same reference numerals even if the drawings are different.

[0070] Furthermore, when other components are arranged or connected to a component in front of, behind, to the left, to the right, above, or below, this includes the presence of another component interposed therebetween. Conversely, when a component is referred to as being "immediately in front of," to the right, to the left, above, or below, it means that no other component exists therebetween. Unless otherwise specifically stated, when a part "includes" other components, this means that the other components may also be included, not that the other components are excluded.

[0071] In addition, the names of the components are distinguished as first, second, etc. in order to distinguish the same relationship between them, but they are not necessarily limited in order. In addition, terms such as "unit", "device", "part", "component" and "module" recorded in the specification represent a comprehensive component that performs at least one function or operation. In addition, the information processors such as terminals and servers recorded in the specification basically mean a hardware system of hardware that implements specific functions or operations, but should not be understood as being limited to specific hardware, and it does not exclude the possibility of being composed of software driven on general hardware in order to implement specific functions or operations. That is, a terminal or server can be any device, or it can be software installed on any device, such as an application.

[0072] In addition, for the sake of convenience of explanation, the size and thickness of each component shown in the drawings are arbitrarily shown, and therefore the present invention is not necessarily limited to the contents shown in the drawings, and in order to clearly express multiple parts and areas of layers and regions, etc., thicknesses, etc. may be shown as exaggerated, enlarged or reduced.

[0073] <Basic structure-method>

[0074] The method of the present invention is a broadcast service quality management method using a video quality evaluation function. The method can be implemented by including a transmission step and an evaluation step.

[0075] The transmission step is a step of transmitting all or part of the decoded frame data of the channel or VOD video currently being viewed from the viewer receiver 17, which is the final receiver of the broadcast service, to the video quality evaluator 30 in the broadcast service network via data communication at a predetermined time interval.

[0076] The above evaluation step is a step in which the video quality evaluator 30 collects the frames and evaluates the quality scores.

[0077] The video quality evaluator 30 is characterized in that, in order to evaluate the image quality score perceived by humans subjectively, the image quality evaluation is achieved by acquiring a video data set of various levels marked with image quality scores and learning a deep learning or machine learning model.

[0078] Here, the video quality evaluator 30 is as follows: Figures 1 to 6 As shown, it is provided independently of the viewer receiver 17 and is connected via a network using a predetermined bandwidth (described later).

[0079] Video data can be collected (transmitted) by region and time. This is because it is difficult to collect all data in real time due to the increased traffic load.

[0080] In addition, it is preferable that the collected result data including the image quality score is transmitted to the service quality management server 40 and recorded.

[0081] At this time, by statistically analyzing the result data, the locations of the NO backbone network 14, the NO regional network 15, the NO user network 16, and the viewer receiver 17 can be identified.

[0082] In addition, if the viewer receiver 17 connected to a specific NO backbone network 14, NO regional network 15, NO user network 16 or a network downstream of the viewer receiver 17 location continuously shows low image quality scores, the network at that location can be evaluated as having a problem.

[0083] Additionally, it is preferred to quantify the level of problems at each location of the NO backbone network 14, NO regional network 15, NO user network 16, or viewer receiver 17, and to regularly report the resulting statistics by region and time.

[0084] <Video Extractor 20>

[0085] Video extractor 20 can be installed in a modular form within viewer receiver 17. Video extractor 20 extracts decompressed broadcast video information into video. The extracted video is then transmitted to video acquirer 21, which compresses the video and prepares it for transmission. The compressed video, ready for transmission, is then transmitted from video quality assessment interface 23 to image quality assessor 30. Video extractor 20 does not always perform video extraction and transmission, but rather receives and executes these operations on a schedule.

[0086] <Operation Order>

[0087] The schedule receiver 22 ( 22 , video extraction schedule manager) receives the extraction schedule from the quality management server 40 .

[0088] The schedule receiver 22 sends an extraction signal to the video acquirer 21 according to the schedule to start extraction and compression.

[0089] The compressed video is sent to the image quality evaluator 30 through the image quality evaluation interface 23 .

[0090] The image quality evaluator 30 performs image quality evaluation after decompressing the video.

[0091] At this time, when the compressed video is transmitted, the unique number of each extracted viewer receiver 17 is transmitted together.

[0092] The image quality evaluator 30 creates an image quality evaluation score table for each viewing period based on the unique number of the viewer receiver 17 .

[0093] <Main Modules of Video Extractor 20>

[0094] 1. Image quality evaluation interface 23

[0095] This module is responsible for communicating with the designated image quality evaluator 30. It calls the API provided by the image quality evaluator 30. When transmitting video data via HTTP, HTTP 2.0 can also be used to reduce data volume. Video is transmitted at the time specified in the schedule. The transmission speed is adjusted to use a bandwidth less than or equal to the specified bandwidth, rather than the maximum bandwidth.

[0096] 2. Video Acquisition 21

[0097] The viewer receiver 17 receives RAW video from a video player. Pixel formats supported include YUV, RGB, and other formats such as YUV420p and YUV422. Video is received at a specified time in a schedule. The RAW video is losslessly or lossily compressed and transmitted to the image quality evaluator 30. In order for the video to be received by the player built into the viewer receiver 17, integration is necessary.

[0098] 3. Schedule receiver 22

[0099] The system receives multicast data packets sent from the quality management server 40. It extracts the schedule information for the corresponding viewer receiver 17 from the data packets and transmits it to the image quality evaluation interface 23 and the video acquisition unit 21. It receives data at designated time intervals to confirm changes to the existing schedule. After confirming the changes, it retransmits them to the relevant modules.

[0100] <Method of extracting video from the viewer receiver 17>

[0101] All viewer receivers 17 use hardware acceleration to perform decoding and screen rendering. In the SW area, access to decoded RAW data and access to decoding time are limited. Therefore, cooperation with chip manufacturers may be required for video extraction.

[0102] <Image Quality Evaluator 30>

[0103] The image quality evaluator 30 evaluates the video received from the viewer receiver 17. The viewer receiver 17 is equipped with a video extractor 20 and receives the video from the video extractor 20. After the received video is decompressed, the image quality evaluation module 33 evaluates the segmented video. The actual transmitted video is scored based on the video evaluation and the viewer receiver 17 identification value (terminal information). The viewer receiver 17 identification value and the video evaluation score are stored in the quality management server 40 or database 50.

[0104] <Operation Order>

[0105] Video extractor 20 receives the schedule from quality management server 40 and extracts the video. Video extractor 20 sends the viewer receiver 17's identification value, the video extraction time, and the compressed video to image quality evaluator 30. Image quality evaluator 30 decompresses the compressed video and then uses image quality evaluation module 33 to score the video quality. Image quality evaluator 30 stores the video quality score, viewer receiver 17's identification value, and video extraction time in quality management server 40 or database 50.

[0106] <Configuration of Image Quality Evaluator 30>

[0107] 1. Quality Management Server Interface 31

[0108] This module is used to communicate with the quality management server 40. It calls the API provided by the quality management server 40. The analysis result of the video received from the viewer receiver 17 is sent together with the terminal information.

[0109] 2. Viewer Receiver API Server 32

[0110] The module acquires video data and terminal information sent from the viewer receiver 17. The received data is sent to the image quality evaluation module 33 to obtain an evaluation result. The evaluation result is sent to the quality management server interface 31, which then transmits it to the quality management server 40. A REST API or protocol is provided to build an environment and meet customer needs. When using HTTP, both HTTP 1.1 and HTTP 2.0 are supported.

[0111] 3. Image quality evaluation module 33

[0112] The obtained video is evaluated based on the data learned by machine learning. The evaluation data is provided as a score, for example, on a 100-point scale. Since the evaluation is based on machine learning, a dedicated hardware / software configuration is required.

[0113] The image quality evaluation module 33 may be equipped with a model learned using various video data sets based on machine learning technology, select video frames, and extract features of the input video frames at a predetermined time period to evaluate the quality score of the video.

[0114] The image quality evaluation module 33 may include: a frame selector; a frame feature extractor; and a video quality regressor.

[0115] The Effective Frame Selector is a device that selects representative frames of high importance from the video frames that make up a video. The Frame Feature Extractor is a device implemented as a deep learning model to extract the features of each video frame and generate a feature vector. The Video Quality Regressor is a device implemented as a deep learning or machine learning model that can learn the changing characteristics of the extracted video frame feature vectors and evaluate the final quality score of the video frame over a predetermined time period.

[0116] The frame feature extractor may include a feature aggregator that uses multiple models in parallel to obtain respective feature vectors, and then aggregating the results of each model to generate the feature vector, thereby improving the accuracy of video frame feature extraction. The multiple models may include at least one of the following: a general characteristic feature extractor; a manufacturing quality feature extractor; a compression quality feature extractor; and a transmission quality feature extractor.

[0117] The general characteristic feature extractor is a deep learning model device that extracts general features or meaningful features that distinguish a particular image from other images. The manufacturing quality feature extractor is a deep learning model device that is used to extract quality features that a video has had since its initial production, that is, features derived from production. The compression quality feature extractor is a deep learning model device that extracts quality features that vary depending on the degree of compression applied to the video, that is, features resulting from compression. The transmission quality feature extractor is a deep learning model device that extracts quality features that may be degraded due to errors that may occur during multi-stage transmission of the video.

[0118] <Quality Management Server 40>

[0119] The quality management server 40 scores videos received from the viewer receiver 17 and stores and manages the rating information in a database 50. For example, a function can be provided for viewing video rating information based on the viewer receiver 17. Furthermore, the installation information of the viewer receiver 17 can be managed by region / area. Furthermore, video quality can be continuously monitored by region / area. The quality management server 40 can register and manage a video extraction schedule in the video extractor 20 installed in the viewer receiver 17.

[0120] <Sequence-Video Quality Monitoring>

[0121] The video is transmitted from the video extractor 20 installed on the viewer / listener receiver 17 to the image quality evaluator 30. The image quality evaluator 30 scores the video quality based on the recognition value of the viewer / listener receiver 17. The image quality evaluator 30 stores the image quality evaluation result in the quality management server 40 or the database 50. The quality management server 40 includes information about the location where the viewer / listener receiver 17 is installed, and thus can provide video quality information by region / area.

[0122] <Sequence-Video Extraction Schedule Management>

[0123] The quality management server 40 registers a video extraction schedule for each region / area and transmits it to the video extractor 20. The video extractor 20 extracts the video on the corresponding schedule and transmits the video and the viewer receiver 17 identification value to the image quality evaluator 30.

[0124] <Configuration of Quality Control Server 40>

[0125] 1. Management Module 41

[0126] This system registers and manages terminals such as the viewer receiver 17. Since a large number of devices must be registered, a registration method that enables efficient management is required. Furthermore, statistical data is provided for analyzing video quality. This includes quality index changes for the NO backbone network 14, NO regional network 15, and NO user network 16, as well as changes in quality index for each terminal.

[0127] 2. API Server 42

[0128] Processes and stores the evaluation data reported by the image quality evaluator 30. Also controls the operation of the image quality evaluator 30. Provides a REST API or a protocol that builds an environment and meets customer needs.

[0129] 3. Schedule Conveyor 43

[0130] To prevent service network failures caused by video data traffic being transmitted from viewer receiver 17 through video extractor 20 to image quality evaluator 30, it is necessary to control video extractors 20 in viewer receiver 17 to transmit video at designated times. Schedule transmitter 43 periodically transmits video extraction operation schedule information to video extractors 20 in all viewer receivers 17. To reduce the burden of transmitting schedules to individual video extractors 20 in viewer receiver 17, schedule information for video extractors 20 in all viewer receivers 17 can be periodically transmitted as a multicast.

[0131] The video extraction schedule of the video extractor 20 may be adjusted according to the performance, operation status, speed, load, etc. of the image quality evaluator 30 . To this end, the quality management server 40 may receive a status report from the image quality evaluator 30 .

[0132] <Video upload bandwidth management>

[0133] A single frame of 1080p RAW data is 3MB (Yuv420p standard). Assuming a 10-second video, the total data size is 3 x 10 x 30 (frames / second) = 900MB. Using lossless compression can reduce this to an average of 600MB (2MB per frame). If the video extractor 20 of the viewer's receiver 17 sends data at maximum bandwidth, it could cause service network failures.

[0134] Therefore, the transmission bandwidth should be adjusted to be within 10 Mbps. 10 seconds of data transmission takes an average of 8 minutes.

[0135] <Video upload schedule>

[0136] Bandwidth limitations may make it difficult to prevent the risk of service network failure when each terminal transmits data randomly. Therefore, it is necessary to specify the video transmission schedule of all viewer receivers 17 and adjust it so as not to exceed the average bandwidth.

[0137] <device>

[0138] The device of the present invention is a broadcast service quality control device using a video quality evaluation function. The device can be implemented as comprising a transmission module and an evaluation module.

[0139] The transmission module is a module that transmits all or part of the decoded frame data of the channel or VOD video currently being watched from the viewer receiver 17, which is the final receiver of the broadcast service, to the video quality evaluator 30 in the broadcast service network through data communication at a predetermined time interval.

[0140] The evaluation module is a module of the video quality evaluator 30 that collects the frames and evaluates the quality scores.

[0141] The video quality evaluator 30 is characterized in that, in order to evaluate the image quality score perceived by humans subjectively, it is achieved by acquiring a video data set of various levels marked with image quality scores and learning a deep learning or machine learning model.

[0142] like Figures 1 to 6 As shown, the viewer receiver 17 includes a video extractor 20 according to the present invention, and is configured to send the video to the image quality evaluator 30 through the video extractor 20 .

[0143] <Procedure>

[0144] On the other hand, the program of the present invention is a program for information equipment for executing the steps of the above-mentioned method on the information equipment. The program is recorded on a storage medium that can be read by the information equipment.

[0145] Here, the information devices collectively refer to portable information processing devices such as computers, smart phones, and tablet computers.

[0146] Although the above content describes the preferred embodiments of the present invention, the present invention is not limited to the above-disclosed embodiments, but can be implemented in various forms different from each other within the scope of the patent claims, the specific embodiments of the present invention and the drawings. Other equivalent embodiments are possible, which also fall within the scope of the present invention. This is obvious to ordinary technicians in this field. The embodiments of the present invention are provided only to make the disclosure of the present invention complete and to fully disclose the scope of the present invention to ordinary technicians in the technical field to which the present invention belongs. The present invention is only defined by the scope of the claims.

[0147] Industrial Applicability

[0148] The present invention can be used in industries involving broadcast service quality management methods, devices, and programs utilizing a video quality evaluation function.

Claims

1. A method for managing broadcast service quality using a video quality evaluation function, characterized in that: The implementation includes: a transmitting step of transmitting, from a viewer receiver serving as a final receiver of the broadcast service, all or part of the decoded frame data of the channel or VOD video currently being viewed, via data communication, at predetermined time intervals, from the viewer receiver to a video quality evaluator in the broadcast service network; and In an evaluation step, a video quality evaluator collects the frames and evaluates the quality scores; Among them, in order to evaluate the image quality score perceived by humans subjectively, the video quality evaluator obtains video data sets of various levels marked with image quality scores and learns deep learning or machine learning models to achieve image quality evaluation.

2. The broadcast service quality management method using the video quality evaluation function according to claim 1, characterized in that: Video data is collected by region and time.

3. The broadcast service quality management method using the video quality evaluation function according to claim 1, characterized in that: The collected result data including the image quality score is transmitted to the service quality management server and recorded.

4. The broadcast service quality management method using the video quality evaluation function according to claim 3, characterized in that: By performing statistical analysis on the result data, the locations of the network operator backbone network, the network operator regional network, the network operator user network and the viewer's receiver are identified.

5. The broadcast service quality management method using the video quality evaluation function according to claim 4, characterized in that: If receivers connected to a specific network operator's backbone network, network operator's regional network, network operator's user network, or a network connected downstream from the viewer's receiver location consistently experience low image quality scores, the network at that location will be evaluated as having problems.

6. The broadcast service quality management method using the video quality evaluation function according to claim 3, characterized in that: Quantify the level of problems at each location of a network operator's backbone network, a network operator's regional network, a network operator's user network, or a viewer's receiver, and report regular statistics of the results by region and time.

7. A broadcast service quality management device using a video quality evaluation function, characterized in that: The implementation includes: a transmission module for transmitting, from a viewer receiver, which is a final receiver of the broadcast service, all or part of the decoded frame data of the channel or VOD video currently being viewed, to a video quality evaluator in the broadcast service network via data communication at predetermined time intervals; and An evaluation module, a video quality evaluator, collects the frames and evaluates the quality scores; The video quality evaluator is implemented by acquiring a video data set of various levels marked with quality scores and learning a deep learning or machine learning model in order to evaluate the quality scores perceived by humans subjectively.

8. A program for an information device, recorded on a storage medium readable by the information device, the storage medium recording the program for the information device for executing each step of the method recited in claim 1 on the information device.

Images