JP2008206042A - Video image quality evaluation method and apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a video quality evaluation apparatus capable of quantitatively evaluating the subjective quality of video images. <P>SOLUTION: A frame identification code generating unit 2 extracts a frame ID from a test video image and generates its QR code. A frame identification code embedding unit 3 embeds the QR code in a part of the test video image. An encoder 4 encodes the test video image with the QR code embedded therein. A decoder 8 decodes a received encoded video image to reproduce the test video image. A player 9 plays back the test video image on a display 9a thereof. A video image photographic unit 10 photographs the display 9a to produce DV video data. A frame identification code extracting unit 11 extracts a frame identification code (QR code in the present embodiment) from the DV video data. A quality evaluating unit 12 statistically evaluates the smoothness of the test video image, on the basis of the extracted QR code. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a video quality evaluation method and apparatus for evaluating the quality of a moving image transmitted on a network on a receiving side, and more particularly to a video quality evaluation method and apparatus capable of evaluating smoothness of a video.

  In addition to the subjective evaluation in which the evaluator subjectively evaluates the quality of the TV screen on which the decoded video is viewed, as a method for evaluating the quality of the encoded video, the content itself is physically There is an objective evaluation method that evaluates by a scale. In objective evaluation, various kinds of data such as a signal-to-noise power ratio (SNR) and a bit error rate (BER) of a video signal are measured, and video quality is quantitatively evaluated based on the numerical data.

In Patent Document 1, as an objective evaluation method, a test signal is input to a communication / broadcasting network to be evaluated, and an output signal that has deteriorated in the communication / broadcasting network is input to an objective quality evaluation unit. The objective quality evaluation unit discloses a technique for comparing quality of an input signal and an output signal to estimate quality degradation in a communication / broadcast network and evaluating the quality.
JP 2001-298428 A

  Although the subjective evaluation method can be expected to be evaluated in actual use, it requires a great amount of labor for collecting subjects and building an environment for conducting an evaluation test, and the burden on the subject is also great. In addition, since there are uncertain factors that greatly affect the evaluation result, such as the age or gender bias of the subject and the physical condition of the test day, it is difficult to always obtain a constant evaluation result. Furthermore, it is difficult to quantitatively evaluate video quality.

  On the other hand, the objective evaluation method does not require collecting subjects and can quantitatively evaluate the video quality without being influenced by external factors such as the test environment. However, since the evaluation result and the user's subjectivity do not always match, evaluation in actual use is difficult.

  An object of the present invention is to solve the above-described problems of the prior art and to provide a video quality evaluation method and apparatus capable of quantitatively evaluating the subjective quality of video.

  In order to achieve the above-described object, the present invention is characterized in that the subjective evaluation of the viewer with respect to the video quality can be quantitatively expressed, and the code information of the frame ID is displayed in each frame. Test video generating means for generating a test video, encoding means for encoding the test video, sending means for sending the encoded test video over the network, receiving the encoded test video, decoding and decoding Reproducing means, photographing means for photographing the reproduced video, code information extracting means for extracting code information from the photographed video, and quality evaluation means for evaluating the quality of the test video based on the extracted code information It is characterized by including.

  According to the present invention, since the video to be evaluated is displayed on the screen in an actual use environment, and the video obtained by photographing this screen is objectively evaluated, the quantitative evaluation result reflecting the subjective evaluation of the viewing user is reflected. Can be obtained.

  DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, the best embodiment of the present invention will be described in detail with reference to the drawings. FIG. 1 is a block diagram showing a configuration of a main part of a video quality evaluation system according to the present invention. Here, configurations unnecessary for explanation of the present invention are omitted.

  The test video database (DB) 1 stores moving images that facilitate visual evaluation of video smoothness. The frame identification code generation unit 2 extracts a frame ID that can uniquely identify each frame from the test video, and generates code information (frame identification code) of the frame ID.

  In the present embodiment, description will be made assuming that a frame number is adopted as a frame ID, a QR code is adopted as a frame identification code, and each frame number is converted to a QR code. As shown in FIG. 2, the frame identification code embedding unit 3 embeds the QR code 13 in a part of each frame image of the test video. The encoding unit 4 encodes the test video in which the QR code 13 is embedded. The transmission unit 5 sends the encoded test video to the video transmission path 6.

  The receiving unit 7 receives the encoded video addressed to the local station from the video transmission path 6, and the decoding unit 8 decodes the encoded video to reproduce the test video. The playback unit 9 plays back the test video on the display 9a. The video imaging unit 10 captures a test video displayed on the display 9a to generate digital video (DV) video data. The frame identification code extraction unit 11 extracts and decodes the frame identification code (in this embodiment, QR code 13) from the DV video data of the test video. The quality evaluation unit 12 statistically evaluates the smoothness of the video based on the time-series change of the frame number obtained by decoding the extracted QR code 13.

  FIG. 3 is a diagram showing an example of the evaluation procedure in the quality evaluation unit 12, and in this embodiment, the video quality is quantitatively evaluated based on the ratio of consecutive identical frames.

  In step S1, a variable i that specifies each frame number of a DV video shot by the video shooting unit 10, a variable j that counts the number of frames in which the same QR code continues, and a variable k that specifies a storage address of the variable j. An initial value “1” is set in.

  In step S2, the frame number Ni extracted from the i-th frame of the DV video and the frame number Ni + 1 extracted from the (i + 1) -th frame are read. In step S3, it is determined whether Ni and Ni + 1 match. If they match, the process proceeds to step S4, and the number of consecutive frames j is incremented.

  In step S5, the frame number variable i is updated. In step S6, it is determined whether or not to end the evaluation, and the process returns to step S2 until the above-described processes are completed for a predetermined number of frames or the evaluation is instructed separately. Is repeated. Therefore, as long as the frame numbers extracted from the frame images match, the continuous frame number j continues to increase.

  Thereafter, if it is determined in step S3 that Ni and Ni + 1 do not match, the process proceeds to step S7, and the number of consecutive frames j counted so far is registered in the storage area D (k). In step S8, the address variable is incremented. In step S9, an initial value “1” is set to the continuous frame number j.

  As the above process proceeds, the number of consecutive frames from the first frame is stored in time series in each of the storage areas D (1) to D (k). FIG. 4 is a diagram showing the relationship between the frame number indicated by the QR code read from each frame of the DV video and the number of consecutive frames j stored in each storage area D (k). ”Is repeated three times, frame number“ 4 ”is continued twice, and frame number“ 7 ”is continued three times, storage area D (1) has“ 3 ”and storage area D (2) has “2” is stored, and “4” is stored in the storage area D (3).

  In step S10, the data of each storage area D (k) is statistically processed. For example, if the average value of the number of consecutive frames is “3” and the DV video frame rate is 30 [fps], the reception is performed. It can be seen that the frame rate of the video displayed on the terminal is 30/3 = 10 [fps]. Since the frame rate Rtest [fps] of the test video is already known, the average value and variance value of the loss frames can be calculated based on both. Also, the frame stop time can be calculated based on the number of consecutive frames.

  In the above-described embodiment, the frame identification code (QR code) is described as being superimposed on a part of the test video. However, the frame identification code may be displayed on the entire surface.

  In the above-described embodiment, the QR code is used as the frame identification code and the frame number is converted into the QR code.However, the present invention is not limited to this, and the barcode, PDF417, It may be converted into other code information such as DataMatrix or Maxi Code.

  Furthermore, as shown in FIG. 5, as a frame identification code, a band-like binary code is arranged below the test video. As shown in FIG. 6, for example, white stripes increase with an increase in the frame identification code. If a code system that moves in one direction is adopted, the smoothness of the image can be recognized not only statistically but visually as a change in the stripe pattern.

It is the block diagram which showed the structure of the principal part of the video quality evaluation system which concerns on this invention. It is the figure which showed an example of the test image | video. It is the flowchart which showed an example of the evaluation procedure of video quality. It is the block diagram which showed the evaluation method of video quality. It is the figure which showed another example of the test image | video. It is the figure which showed an example of the binary encoding code.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 ... Test video database (DB), 2 ... Frame identification code production | generation part, 3 ... Frame identification code embedding part, 4 ... Encoding part, 5 ... Transmission part, 6 ... Video transmission path, 7 ... Reception part, 8 ... Decoding section, 9 ... reproducing section, 10 ... video shooting section, 11 ... frame identification code extracting section, 12 ... quality evaluating section

Claims (8)

Test video generation means for generating a test video in which code information of a frame ID is displayed in each frame of the video;
Encoding means for encoding the test video;
Sending means for sending the encoded test video onto a video transmission path;
Decoding means for receiving and decoding the encoded test video;
Playback means for playing back the decoded test video;
Photographing means for photographing the reproduced test video;
Code information extracting means for extracting code information from the photographed test video;
A video quality evaluation apparatus comprising: quality evaluation means for evaluating the quality of a test video based on the extracted code information.   2. The video quality according to claim 1, wherein the quality evaluation unit decodes each code information to reproduce a frame ID, and statistically evaluates the video quality based on a time series change of the frame ID. Evaluation device.   The video quality evaluation apparatus according to claim 1, wherein the test video generation unit superimposes the code information on a part of each frame image of the test video.   4. The video quality evaluation apparatus according to claim 1, wherein the code information is one of a QR code, a barcode, PDF417, DataMatrix, and Maxi code.   4. The video quality evaluation apparatus according to claim 1, wherein the code information is a binary code whose pattern regularly moves in accordance with an increase in frame ID. A procedure for generating a test video in which code information of a frame ID is displayed in each frame of the video;
A procedure for encoding the test video;
Sending the encoded test video onto a video transmission path;
Receiving and decoding the encoded test video;
Replaying the decoded test video;
A procedure for photographing the reproduced test video;
A procedure for extracting code information from the photographed test video;
And a procedure for evaluating the quality of the test video based on the extracted code information.   7. The procedure for evaluating the quality includes decoding each code information to reproduce a frame ID, and statistically evaluating the video quality based on a time series change of the frame ID. Video quality evaluation method.   The video quality evaluation method according to claim 6, wherein in the means for generating the test video, the code information is superimposed on a part of each frame image of the test video.