JP2013046113A - VIDEO QUALITY ESTIMATION DEVICE, METHOD AND PROGRAM USING BASIC GoP LENGTH - Google Patents

Abstract

PROBLEM TO BE SOLVED: To estimate the video quality evaluation value of a packetized video with high accuracy.SOLUTION: The amount of information for each video frame is derived from an input packet, and the basic GoP length is derived from the amount of information for each video frame. When the video frame type (I, B, P frame) of all video frames is derived from the basic GoP length, and the video frame deteriorates due to packet loss, a deterioration video frame is derived while taking account of the influence of propagation of the deterioration. A bit rate is calculated from the amount of information for each video frame, and the encoded video quality is derived. Finally, a video quality evaluation value is derived from the encoded video quality and the deterioration video frame.

Description

  The present invention relates to a video quality estimation apparatus, method, and program using a basic GoP (Group of Picture) length, and in particular, an IPTV service and video distribution service performed via an IP (Internet Protocol) network such as the Internet. The present invention relates to a video quality estimation apparatus, method, and program using a basic GoP length used for objectively evaluating quality.

  With the increase in speed and bandwidth of Internet access lines, the spread of video communication services that transfer video media including video and audio between terminals or between servers and terminals via the Internet is expected.

  Since the Internet is a network whose communication quality is not always guaranteed, when communication is performed using audio and video media, the bit rate is lowered and the line is congested due to the narrow line bandwidth of the network between user terminals. As a result, packet loss and packet transfer delay occur, and the quality perceived by the user for audio, video media, etc. (user experience quality: QoE (Quality of Experience)) deteriorates.

  Specifically, when the original video is encoded, the video signal in the frame is deteriorated by processing in units of blocks, or high-frequency components of the video signal are lost, so that the fineness of the entire video is lowered. Further, when the encoded video data is transmitted as a packet to the user terminal via the network, if a packet loss or discard occurs, the frame is deteriorated and the quality of the entire video is lowered.

  As a result, the user perceives blur, blur, and mosaic distortion in the received video.

  In order to confirm that the video communication service as described above is provided with good quality, the quality of the video provided to the user is measured by measuring the quality of the video experienced by the user during the service provision. It is important to monitor that is high.

  Therefore, there is a need for video quality objective evaluation technology that can appropriately express the quality of the video experienced by the user.

  Conventionally, as a method for evaluating video quality, there are a subjective quality evaluation method (for example, see Non-Patent Document 1) and an objective quality evaluation method (for example, Non-Patent Document 2).

  The subjective quality evaluation method is a quality scale (very good, good, normal, bad, very good) of the quality experienced by multiple users who actually watched the video and experienced the quality (may be 9 or 11 levels) Bad) and disturbance scale (no degradation is observed, degradation is observed but not bothered, slightly concerned about degradation, concerned about degradation, very concerned about degradation), etc. Average the quality evaluation value of each video (for example, video with a packet loss rate of 0% and a bit rate of 20 Mbps) by the number of users, and define the value as MOS (Mean Opinion Score) value or DMOS (Degradation Mean Opinion Score) doing.

  However, subjective quality evaluation not only requires special dedicated equipment (such as a monitor) and an evaluation facility that can adjust the evaluation environment (such as room illuminance and room noise), but many users actually evaluate the video. There is a need. For this reason, it takes time for the user to actually complete the evaluation, which is not suitable for evaluating the quality in real time.

  Therefore, it is desired to develop an objective quality evaluation method that outputs a video quality evaluation value using a feature amount (for example, bit rate, bit amount in frame units, packet loss information, etc.) that affects video quality.

  As one of the conventional objective quality evaluation methods, the input value of GoP (Group of Picture) obtained from the transmitted packet and the service provider is input, and the video frame type (I, B, There is a technique for deriving a video quality evaluation value by estimating a P frame) and considering how much deterioration is propagated in a video frame lost due to packet loss (see, for example, Non-Patent Document 2).

  Many of the conventional objective quality evaluation methods estimate a video quality evaluation value using a packet as described above.

ITU-T recommendation P.910 Ushiki, Masuda, Hayashi, Takahashi, "Packet layer video quality estimation using frame type estimation," IEICE Transactions B Vol.J94-B, No.1, pp.36-48.

  In order to confirm that the video communication service is provided with good quality, the quality of the video experienced by the user is measured while the service is provided, and the high quality of the video provided to the user is monitored. Therefore, there is a need for an objective video quality evaluation technique that can appropriately display the quality of the video experienced by the user.

  However, since the technology of Non-Patent Document 2 is based on the premise that the basic GoP length obtained from a service provider or the like (hereinafter referred to as the basic GoP length) is used, it cannot be applied to services where the basic GoP length is unknown. There is.

  Specifically, in Non-Patent Document 2, a value obtained by adding a certain amount (several video frames) to the basic GoP length is set as an I frame estimation section, and the video frame having the largest bit amount in the estimation section However, there is a problem that the estimated section of the I frame is not determined.

  The present invention has been made in view of the above points, and in order to solve the above-described problem, the bit amount of each video frame is input, and the basic GoP length is derived from the variation of the bit amount. As described in Document 2, a video quality estimation apparatus and method using a basic GoP length for deriving a video quality evaluation value in consideration of how much the video frame lost due to packet loss propagates, and The purpose is to provide a program.

  FIG. 1 is a principle configuration diagram of the present invention.

The present invention (Claim 1) is a video quality estimation apparatus using a basic GoP length for objectively evaluating video quality,
Video frame discriminating means for grouping input packets for each video frame;
Video frame information amount calculating means for deriving a video frame information amount for each video frame determined by the video frame determining means;
I frame interval calculation means for deriving an interval between video frames for each I frame from the video frame information amount for each video frame derived by the video frame information amount calculation means;
A basic GoP length calculating means for calculating, as a basic GoP (Group of Picture) length, an I frame interval having the highest occurrence frequency of the quality measurement interval from a set of I frame intervals derived by the I frame interval calculating means;
Based on the basic GoP length derived by the basic GoP length calculating means, video frame type calculating means for calculating video frame types (I, B, P frames) of all video frames;
Video bit rate calculating means for calculating a video bit rate from the video frame information amount for each video frame derived by the video frame information amount calculating means;
A lost video frame determining means for determining whether a packet loss is included in the video frame determined by the video frame determining means;
A deteriorated video frame calculating means for determining how much the influence of loss propagates from the video frame type calculated by the video frame type calculating means and the lost video frame determined by the lost video frame determining means;
Encoded video quality calculating means for calculating video quality related to video encoding using the bit rate of the video calculated by the video bit rate calculating means;
Video quality calculating means for calculating video quality from the encoded video quality calculated by the encoded video quality calculating means and the deteriorated video frame calculated by the deteriorated video frame calculating means.

  Conventionally, the basic GoP length (I-frame interval) used for video encoding has been implemented by the service provider giving the basic GoP length to the video quality evaluation device. Under the circumstances that cannot give the basic GoP length, it is not possible to evaluate the video quality using the above techniques.

  On the other hand, according to the present invention, the basic GoP length can be calculated from only received packet data based on the information amount for each video frame, and the video frame type of each video frame can be derived. Video quality can be evaluated even in situations where long information is not available.

  Therefore, the present invention makes it possible to monitor the video quality value for the video of the video communication service that is actually viewed by the user. Therefore, it is easy to determine whether the service being provided maintains a certain level of quality for the user. Judgment can be made.

  For this reason, it becomes possible to improve the point that the quality of the service being provided could not be evaluated with the prior art.

It is a block diagram of the video quality estimation apparatus in one embodiment of this invention. It is a block diagram of the I frame space | interval calculation part in one embodiment of this invention. It is an example of the video frame used as the discrimination | determination object in the video frame discrimination | determination part in one embodiment of this invention. It is an example of I frame interval calculation of the I frame interval calculation part in one embodiment of this invention. It is an example of the clustering of the video frame in one embodiment of this invention. It is an example of a structure of an encoding video quality and video quality calculation function. It is a flowchart (the 1) of an encoding video quality and video quality calculation process. It is a flowchart (the 2) of an encoding video quality and video quality calculation process. It is a flowchart of operation | movement of the video quality estimation apparatus in one Example of this invention.

  Embodiments of the present invention will be described below with reference to the drawings.

  The video quality estimation apparatus using the basic GoP length according to this embodiment of the present invention (hereinafter referred to as “video quality estimation apparatus”) is input based on video frame delimiter information indicated in the input packet. Each video frame included in the packet is determined, and the information amount (bit amount and number of packets) of the determined video frame is calculated. As the video frame delimiter information, in the case of RTP / UDP / IP, if the marker bit written in the RTP (Real-time Transport Protocol) header is 0x1 (that is, the last packet constituting the video frame) TS / RTP / UDP / IP or TS in the case of RTP timestamp (uses the same video frame consisting of multiple packets to have the same RTP timestamp to determine the video frame delimiter) In the case of / UDP / IP, TTS / RTP / UDP / IP and TTS / UDP / IP, when the Payload unit start indicator (PUSI) described in TS (Transport Stream) and TTS (Timestamped TS) is 0x1 ( Use the presence of a PES (Packetized Elementary Stream) header (uses the fact that one PES contains one video frame), etc. Is possible. Based on the calculated amount of information for each video frame, the video frames are clustered into several groups, and all the video frames in the group including the video frame with the largest amount of information are defined as I frames, and the frequency of occurrences occurs at I frame intervals Is the basic GoP length, and the video frame type (I, B, P frame) for each video frame is estimated. Here, as the clustering method, the shortest distance method and the ascending order method are targeted in the embodiment of the present application, but other methods can also be used. The video bit rate is calculated from the amount of information for each video frame, and the video quality immediately after encoding is derived from the bit rate. A video frame in which packet loss has occurred is discriminated, and a degraded video frame length is calculated from the lost video frame and the video frame type, taking into account the propagation state of degradation. Finally, the video quality is objectively evaluated from the encoded video quality and the degraded video frame length.

  For example, in the present embodiment, in order to realize objective video quality evaluation in video communication such as IPTV service and video distribution service performed via an IP network such as the Internet, packets are analyzed and these video communication are performed. Deriving a video quality value that quantitatively represents the feature quantity that affects the video quality involved.

  FIG. 1 shows the configuration of a video quality estimation apparatus according to an embodiment of the present invention.

  As shown in the figure, the video quality estimation apparatus 100 includes a video frame determination unit 1, a video frame information amount calculation unit 2, an I frame interval calculation unit 3, a basic GoP length calculation unit 4, and a video frame type calculation. 5, a video bit rate calculation unit 6, a lost video frame determination unit 7, a degraded video frame calculation unit 8, an encoded video quality calculation unit 9, and a video quality calculation unit 10.

  As shown in FIG. 2, the I-frame interval calculation unit 3 includes a shortest distance video frame calculation unit 31 (FIG. 2A) or an ascending video frame calculation unit 32 (FIG. 2B).

  In addition, each said calculation part shall have a memory (not shown) which preserve | saves a calculation result temporarily.

  The video frame discriminating unit 1 composes an input packet as shown in FIG. 3 based on marker bits having specific values, which are video frame delimiter information shown in the input packet, and PUSI information. Each video frame is discriminated by grouping each video frame. FIG. 3 shows an example when marker bits having a value of 0x1 are used as video frame delimiter information when the packet configuration is RTP / UDP / IP. Even if video frame delimiter information is lost due to packet loss, if the packet configuration is RTP / UDP / IP, the RTP time stamp increases for each video frame, so the loss can be detected from the amount of RTP time stamp increase. . Furthermore, since the frame rate of the video can be converted from the time stamp, even if a plurality of video frames are continuously lost, it is possible to confirm the presence of the video frame that has been lost from the time stamp. However, it cannot be determined how many packets each video frame has.

  The video frame information amount calculation unit 2 counts the number of packets grouped as packets constituting the video frame for each video frame discriminated by the video frame discrimination unit 1 and stores it as an information amount for each video frame. . At this time, instead of the number of packets, the number of packets may be counted as the number of bits or bytes.

  The I-frame interval calculation unit 3 clusters the video frames into several groups based on the information amount for each video frame calculated by the video frame information amount calculation unit 2, and is the same as the video frame having the largest amount of information. All the video frames classified into the group are determined as I frames, and the interval between each I frame is calculated as the I frame interval.

  The I-frame interval calculation unit 3 includes a shortest distance video frame calculation unit 31 and clusters video frames by applying the shortest distance method based on the information amount for each video frame calculated by the video frame information amount calculation unit 2. However, all the video frames classified into the same group as the video frame having the largest amount of information may be determined as I frames. Specifically, as shown in FIG. 4, when the video frame rate is 3 fps, the shortest distance method is applied with 6 video frames twice the frame rate of 3 being input to the shortest distance video frame calculation unit 31. Then, clustering into several groups (in this case, the input to the shortest distance method may be a value such as three times the video frame rate). In grouping by clustering, there are three types of video frame types, I, B, and P frames, in the video frame type discriminating unit 5, so processing continues until there are three groups, and it is the same as the video frame with the largest amount of information All the video frames classified into the group are determined as I frames, and the I frame interval calculation unit 3 calculates the intervals at which frame numbers recognized as I frames appear. In the example of FIG. 4, the video frame number of the video frame with the largest amount of information is “6”, so the video frame number “5” that is the same group as the “6” is the I frame. For example, if the video frame numbers “10” to “18” are the same group, “10” to “18” are all I frames.

  The I frame interval calculation unit 3 includes an ascending video frame calculation unit 32 instead of the shortest distance video frame calculation unit 31 described above, and calculates the information amount for each video frame calculated by the video frame information amount calculation unit 2. The video frames may be clustered by applying the ascending order method, and all the video frames classified into the same group as the video frame having the largest amount of information may be determined as I frames.

  Specifically, FIG. 5 shows an example of clustering by the ascending order method. In the figure, f indicates a video frame number, and VTSf indicates the TS number for each video frame. The derivation procedure in the figure is as follows.

  1) Input video frame data for (2 × frame rate).

  2) Store the number of video frames having the maximum number of TSs and the number of TSs every 10 video frames (however, if there are multiple maximum values, select the smaller video frame).

  3) The video frames selected in 2) are arranged in ascending order with respect to the TS number.

  4) Calculate the division value of the TS numbers arranged in order.

  5) A temporary I frame is determined with the division value being the maximum (outlined portion) as a threshold (however, if there are a plurality of maximum values, the first maximum value).

  6) The temporary I-frame interval is calculated from “the longer the distance”, and the most frequently occurring one is set as the basic GoP length (however, the distance of “10 or less” is not counted).

  In the example of FIG. 5, when the video frame rate is 30 fps, 60 video frames that are twice the frame rate of 30 are clustered into several groups as inputs to the ascending video frame calculation unit (in this case, the ascending order method). Can be a value such as three times the video frame rate). However, a bold line in the drawing indicates a delimiter for data input. Specifically, for the input 60 video frames, video frames with the maximum amount of information are extracted every 10 video frames, and ascending order is applied to rearrange the data based on the magnitude relationship of the information amount. (In this example, from the viewpoint of reducing the amount of calculation, a video frame having the maximum amount of information is extracted for every 10 video frames. Ascending order may be applied). Next, for the video frames rearranged in ascending order, the division method is used to divide the information amount of the adjacent video frame (in the example of FIG. 5, the information amount of the video frame number 34 is changed to the information amount of the video frame number 18). The same processing is performed for other video frames). Among the obtained division values, the maximum division value is set as a threshold, and a video frame having an information amount exceeding the threshold is set as a temporary I frame.

  The basic GoP length calculation unit 4 sets the GoP length having the highest occurrence frequency in each GoP length from the I frame interval determined by the I frame interval calculation unit 3 as the basic GoP length.

  The video frame type calculation unit 5 calculates the video frame type (I, P, B) from the basic GoP length calculated by the basic GoP length calculation unit 4 and the information amount for each video frame calculated by the video frame information amount calculation unit 2. ). As shown in Non-Patent Document 2, specifically, a value obtained by adding a value of 2 to 4 to the basic GoP length is set as a search range, and a video frame having the maximum amount of information in the search range is defined as an I frame. To do. The amount of information of video frames between I frames is averaged, and a value obtained by multiplying a predetermined coefficient α is used as a threshold value. Is determined to be a B frame (see non-patent document 2 FIG. 7). Note that the coefficient α is a value that varies depending on the implementation of the video codec, and generally has a value of 0.9 to 1.3.

  The video bit rate calculation unit 6 calculates the bit rate based on the information amount calculated by the video frame information amount calculation unit 2.

  The lost video frame discriminating unit 7 discriminates whether there is a packet loss in the packets grouped for each video frame by the video frame discriminating unit 1, and if there is a packet loss, determines that the video frame is a lost video frame. To do. Specifically, a lost packet is detected from the missing RTP time stamp of the RTP packet or the cyclic counter (CC) in the TS header of the TS packet, and the video frame including the lost packet is determined as a lost video frame. .

  As shown in Non-Patent Document 2, the degraded video frame calculation unit 8 is based on the lost video frame determined by the lost video frame determination unit 7 and the video frame type determined by the video frame type calculation unit 5. The number of deteriorated video frames is counted in consideration of the propagation state of loss deterioration (see Non-Patent Document 2 FIG. 3).

  The encoded video quality calculation unit 9 calculates the encoded video quality (VQC) based on the following equation based on the bit rate (BR) calculated by the video bit rate calculation unit 6. However, v1, v2, and v3 are constants stored in advance.

映像品質算出部１０は、符号化映像品質算出部９で算出された符号化映像品質（VQC）と劣化映像フレーム算出部８で算出された劣化映像フレーム数（DF）から以下の数式に基づき、映像品質を算出する。ただし、v4, v5, v6は、予め記憶されている定数とする。

The video quality calculation unit 10 is based on the following formula from the encoded video quality (VQC) calculated by the encoded video quality calculation unit 9 and the number of degraded video frames (DF) calculated by the degraded video frame calculation unit 8: Calculate video quality. However, v4, v5, and v6 are constants stored in advance.

また、IPネットワーク経由で行うIPTVサービス、映像配信サービスなどの映像通信に関わる映像品質に影響を与えるビットレートや映像フレーム毎のビット量と映像フレーム種別を利用し、符号化映像品質（VQC）、映像品質（VQ）を導出しても良い。この、ビットレート、映像フレーム毎のビット量と映像フレーム種別を利用して、符号化映像品質（VQC）、映像品質（VQ）を導出する、符号化映像品質・映像品質算出機能の一実施形態を以下に例示する。なお、詳細は、PCT/JP2010/055203に記載されている。

In addition, using the bit rate that affects video quality related to video communication such as IPTV service and video distribution service via IP network, the bit amount for each video frame and the video frame type, the encoded video quality (VQC), Video quality (VQ) may be derived. An embodiment of an encoded video quality / video quality calculation function for deriving encoded video quality (VQC) and video quality (VQ) using the bit rate, the bit amount per video frame and the video frame type. Is exemplified below. Details are described in PCT / JP2010 / 055203.

  FIG. 6 shows an example of the configuration of the encoded video quality / video quality calculation function. The encoded video quality / video quality calculation function performs encoding using the bit rate, the bit amount for each video frame, and the video frame type. Video quality (VQC) and video quality (VQ) are derived.

  The encoded video quality / video quality calculation function 200 shown in the figure analyzes the encoded video packet included in the input packet, and quantitatively represents the feature quantity affecting the video quality related to the video communication. Deriving a quality value. Specifically, by deriving a video quality evaluation value in consideration of the bit rate of the encoded video, the bit amount for each video frame type in the video frame of the encoded video, and the number of lost video frames, the encoded video is derived. The video quality is estimated. The encoded video quality / video quality calculation function 200 includes a packet analysis unit 20, a video aggregate frame characteristic estimation unit 21, an encoding quality estimation unit 22, and a packet loss quality estimation unit 23.

  The packet analysis unit 20 includes a bit rate calculation unit 201 that derives a bit rate of an encoded video packet included in an input packet, and a bit amount calculation unit 202 that derives a bit amount for each video frame type. The bit rate derived by the bit rate calculation unit 201 and the bit amount for each video frame type derived by the bit amount calculation unit 202 are output.

  The video aggregate frame characteristic estimation unit 21 receives the bit rate output from the bit rate calculation unit 201 of the packet analysis unit 20 and derives and outputs a frame characteristic indicating the bit amount characteristic for each frame type.

  The encoding quality estimation unit 22 includes the bit rate output from the bit amount calculation unit 202 of the packet analysis unit 20, the bit amount for each video type, and the frame for each video frame type output from the video aggregate frame characteristic estimation unit 21. The coded video quality evaluation value (VQC) is derived based on the characteristics.

  The packet loss video quality estimation unit 23 encodes the encoded video quality evaluation value (VQC) calculated by the encoding quality estimation unit 22, the bit amount and the number of video frame losses for each video frame type calculated by the packet analysis unit 20. A video that quantitatively represents the quality of the encoded video that is affected by the packet loss degradation, based on the number of lost video frames representing the frame quality and the frame characteristics for each video frame type derived by the video aggregate frame characteristic estimation unit 21 Deriving the quality evaluation value (VQ). 7 and 8 exemplify detailed flowcharts for deriving the encoded video quality (VQC) and video quality (VQ) using the bit amount and video frame type for each video frame.

  Next, the operation of the video quality estimation apparatus 100 according to the present embodiment will be specifically described below.

  FIG. 9 is a flowchart of the operation of the video quality estimation apparatus in one embodiment of the present invention.

  When a packet is input to the video frame determination unit 1 of the video quality estimation apparatus 100, the packets constituting each video frame are grouped based on information indicating the delimiter position of the video frame, and the video frame is determined. Is output to the video frame information amount calculation unit 2 and the lost video frame discrimination unit 7 (S101).

  The video frame information amount calculation unit 2 counts the number of packets constituting the video frame input from the video frame determination unit 1 as the information amount for each video frame, and sends it to the I frame interval calculation unit 3 and the video bit rate calculation unit 6. Output (S102).

Based on the input information amount for each video frame, the I-frame interval calculation unit 3 repeats clustering until it finally converges into three groups, and the video frame with the largest amount of information among the three converged groups is selected. All the video frames classified into the included group are set as I frames, the interval between each I frame is calculated, and output to the basic GoP length calculation unit 4 (S103).
In this embodiment, as a clustering method, the case where the shortest distance method is applied and the case where the ascending order method is applied are illustrated in FIGS. The basic GoP length calculation unit 4 sets, as the basic GoP length, the one having a high occurrence frequency among the I frame intervals calculated by the I frame interval calculation unit 3, and outputs the basic GoP length to the video frame type calculation unit 5 (S104).

  The video frame type calculation unit 5 calculates the video frame type (I, P, B frame) from the basic GoP length derived by the basic GoP length calculation unit 4 and the information amount for each video frame calculated by the video frame information amount calculation unit 2. ) Is derived and output to the degraded video frame calculation unit 8 (S105).

  The video bit rate calculation unit 6 calculates the bit rate from the information amount for each video frame input from the video frame information amount calculation unit 2, and outputs the bit rate to the encoded video quality calculation unit 9 (S106).

  The lost video frame discriminating unit 7 determines a lost video frame depending on whether or not each video frame input from the video frame discriminating unit 1 includes a packet loss, and the video frame determined to be a lost video frame is a degraded video. It is output to the frame calculation unit 8 (S107).

  The degraded video frame calculation unit 8 takes into account the propagation status of degradation from the lost video frame input from the lost video frame determination unit 7 and the video frame type input from the video frame type calculation unit 5, and determines the number of degraded video frames. Is output to the video quality calculation unit 10 (S108).

  The encoded video quality calculation unit 9 derives the encoded video quality immediately after encoding based on the bit rate input from the video bit rate calculation unit 6 and outputs it to the video quality calculation unit 10 (S109).

  The video quality calculation unit 10 derives the video quality from the encoded video quality input from the encoded video quality calculation unit 9 and the lost video frame input from the degraded video frame calculation unit 8 (S110), and ends the processing. To do.

  Thus, according to the present embodiment, it is possible to calculate the basic GoP length based on the amount of information for each video frame from only the received packet data and derive the video frame type of each video frame. The video quality can be evaluated even in situations where basic GoP length information is not available in advance.

  Therefore, it is possible to easily determine whether the service being provided maintains a certain level of quality for the user, and it becomes possible to grasp and manage the actual quality of the service being provided in real time. .

  The video quality estimation apparatus 100 in the above embodiment is realized by installing a computer program in a computer including a CPU (Central Processing Unit), a memory, and an interface, and the various functions of the video quality estimation apparatus 100 described above are performed. The various hardware resources of the computer and the computer program (software) are realized in cooperation.

  The operations of the constituent elements of the video quality estimation apparatus 100 shown in FIGS. 1 and 2 are constructed as a program and installed and executed on a computer used as the video quality estimation apparatus, or distributed via a network. Is possible.

  If the flag indicating I frame is not encrypted in the IPTV specification in Europe, the basic GoP size is obvious, but the flag indicating I frame is encrypted in the European IPTV specification. Since the basic GoP size is not obvious in the case of standard Japanese packet specifications, it is necessary to derive the GoP size using the packet, and the present invention is effective.

  Furthermore, the constructed program can be stored in a portable storage medium such as a hard disk, a flexible disk, or a CD-ROM, and installed in a computer or distributed.

  The present invention is not limited to the above-described embodiment, and various modifications and applications can be made within the scope of the claims.

  It can be used for a video quality estimation device that estimates a video quality evaluation value of video communication such as an IPTV service and a video distribution service performed via an IP network.

DESCRIPTION OF SYMBOLS 1 Video frame discrimination | determination part 2 Video frame information amount calculation part 3 I frame interval calculation part 4 Basic GoP length calculation part 5 Video frame type calculation part 6 Video bit rate calculation part 7 Loss video frame discrimination part 8 Degraded video frame calculation part 9 Code Video quality calculator
10 Video quality calculator
20 packet analysis unit 21 video aggregate frame characteristic estimation unit 22 encoding quality estimation unit 23 packet loss quality estimation unit 31 shortest distance video frame calculation unit
32 Ascending video frame calculation unit 100 Video quality estimation apparatus 200 Encoded video quality / video quality calculation function 201 Bit rate calculation unit 202 Bit amount calculation unit

Claims (7)

A video quality estimation device that uses the basic GoP length to objectively evaluate video quality,
Video frame discriminating means for grouping packets for each video frame based on the input packet information;
Video frame information amount calculating means for deriving a video frame information amount for each video frame determined by the video frame determining means;
I frame interval calculation means for deriving an interval between video frames for each I frame from the video frame information amount for each video frame derived by the video frame information amount calculation means;
A basic GoP length calculating means for calculating, as a basic GoP (Group of Picture) length, an I frame interval having the highest occurrence frequency of the quality measurement interval from a set of I frame intervals derived by the I frame interval calculating means;
Based on the basic GoP length derived by the basic GoP length calculating means, video frame type calculating means for calculating video frame types (I, B, P frames) of all video frames;
Video bit rate calculating means for calculating a video bit rate from the video frame information amount for each video frame derived by the video frame information amount calculating means;
A lost video frame determining means for determining whether a packet loss is included in the video frame determined by the video frame determining means;
A deteriorated video frame calculating means for determining how much the influence of loss propagates from the video frame type calculated by the video frame type calculating means and the lost video frame determined by the lost video frame determining means;
Encoded video quality calculating means for calculating video quality related to video encoding using the bit rate of the video calculated by the video bit rate calculating means;
Video quality calculating means for calculating video quality from the encoded video quality calculated by the encoded video quality calculating means and the degraded video frame calculated by the degraded video frame calculating means;
A video quality estimation device using a basic GoP length characterized by comprising: In the I frame interval calculation means,
2. The shortest distance video frame calculation means for clustering video frame information based on the shortest distance method and determining video frames grouped in the same group as video frames having the maximum video frame information amount as I frames. Video quality estimation device using the basic GoP length described. In the I frame interval calculation means,
2. The ascending order video frame calculating means according to claim 1, further comprising: ascending order picture frame calculating means for clustering the information amount of the picture frame based on the ascending order method, and judging the picture frame grouped with the picture frame having the maximum picture frame information quantity as the I frame. Video quality estimation device using basic GoP length. A video quality estimation method using a basic GoP length to objectively evaluate video quality,
A video frame discrimination step, wherein the video frame discrimination means groups the packets for each video frame based on the input packet information;
A video frame information amount calculating means for deriving a video frame information amount for each video frame determined by the video frame determining step;
An I frame interval calculating means for deriving an interval of video frames for each I frame from the video frame information amount for each of the video frames derived in the video frame information amount calculating step;
The basic GoP length calculation means calculates the basic GoP (Group of Picture) length as the basic GoP (Group of Picture) length from the set of I frame intervals derived in the I frame interval calculation step. A calculation step;
A video frame type calculating step for calculating a video frame type (I, B, P frame) of all video frames based on the basic GoP length derived in the basic GoP length calculating step; ,
A video bit rate calculating means for calculating a video bit rate from a video frame information amount for each video frame derived in the video frame information amount calculating step;
A lost video frame determining means for determining whether the video frame determined in the video frame determining step includes a packet loss;
Degraded video frame calculating means for determining how much the influence of loss propagates from the video frame type calculated in the video frame type calculating step and the lost video frame determined in the lost video frame determining step A calculation step;
An encoded video quality calculating step, wherein the encoded video quality calculating means calculates a video quality related to video encoding using the bit rate of the video calculated in the video bit rate calculating step;
A video quality calculating means for calculating a video quality from the encoded video quality calculated in the encoded video quality calculating step and the deteriorated video frame calculated in the deteriorated video frame calculating step;
A video quality estimation method using a basic GoP length characterized by In the I frame interval calculation step,
5. The basic GoP length according to claim 4, wherein the information amount of the video frame is clustered based on the shortest distance method, and the video frame that is grouped with the video frame having the maximum video frame information amount is determined as an I frame. Video quality estimation method. In the I frame interval calculation step,
The video using the basic GoP length according to claim 4, wherein the information amount of the video frame is clustered based on an ascending order method, and the video frame grouped with the video frame having the maximum video frame information amount is determined as an I frame. Quality estimation method. Computer
A video quality estimation program using a basic GoP length for functioning as each unit of the video quality estimation apparatus using the basic GoP length according to any one of claims 1 to 3.

Images