JP5142316B2 - VIDEO QUALITY OBJECTIVE EVALUATION METHOD, VIDEO QUALITY OBJECTIVE EVALUATION DEVICE, AND PROGRAM - Google Patents

Description

  The present invention relates to a technique for estimating the subjective quality of a video from the measurement of the feature quantity of the video signal without performing a subjective quality evaluation experiment in which a human views the video and evaluates the quality thereof, and particularly, block distortion, blur, etc. The present invention relates to a technique for estimating overall subjective quality in consideration of the influence of spatial degradation on subjective quality and the influence of temporal degradation on frame quality, freeze, and the like on subjective quality.

  Conventionally, there is a technique for estimating subjective quality from a video signal, but most of the techniques are intended only for spatial degradation or temporal degradation. For example, a technique for estimating subjective quality by effectively considering spatial degradation such as block distortion and blur (see Non-Patent Document 1), and temporal degradation such as a decrease in frame rate and freeze are effectively performed. There is a technique (see Non-Patent Document 2) that estimates the subjective quality in consideration.

Okamoto, Hayashi, Takahashi, Kurita, “Proposal of Objective Video Quality Evaluation Method Considering Spatiotemporal Features of Video”, The Institute of Electronics, Information and Communication Engineers, IEICE Transactions, Vo1.J88-B, No. 4, p.813-823, 2005 Watanabe, Okamoto, Kurita, “Proposal of Objective Video Quality Evaluation Method for Freeze Degradation of Arbitrary Length”, The Institute of Electronics, Information and Communication Engineers, IEICE Technical Report, Vo1.106, No.239, p.43-46, 2006

  When spatial degradation and temporal degradation occur simultaneously, the subjective quality can be estimated by the simple sum of the technique disclosed in Non-Patent Document 1 and the technique disclosed in Non-Patent Document 2. However, if spatial degradation and temporal degradation occur simultaneously, the subjective quality cannot be a simple sum of the spatial degradation effect and the temporal degradation effect due to the influence of the perceptual characteristics on human degradation. Therefore, even if the technique disclosed in Non-Patent Document 1 and the technique disclosed in Non-Patent Document 2 are simply combined, there is a problem that the overall subjective quality cannot be accurately estimated.

  The present invention has been made to solve the above-described problem, and an objective video quality evaluation method capable of appropriately estimating the overall subjective quality by taking into account the interaction between spatial degradation and temporal degradation. An object is to provide an objective video quality evaluation apparatus and program.

The video quality objective evaluation method of the present invention includes a first quality parameter derivation procedure for deriving a subjective quality evaluation value in consideration of spatial degradation of the evaluation video and a feature quantity of the spatial degradation of the evaluation video, and the evaluation Results of a second quality parameter derivation procedure for deriving a subjective quality evaluation value in consideration of temporal degradation of the video and a feature quantity of temporal degradation of the evaluation video, and results of the first and second quality parameter derivation procedures And an integrated procedure for estimating a total subjective quality of the evaluation video in consideration of the interaction between the spatial degradation and the temporal degradation, and the integration procedure includes the first quality parameter derivation procedure. From the result, the temporal influence is obtained from the result of the first influence evaluation value derivation procedure for obtaining the evaluation value S of the influence of the spatial degradation on the subjective quality of the evaluation video and the second quality parameter derivation procedure. Deterioration is said A second influence evaluation value derivation procedure for obtaining an evaluation value T of the influence on the subjective quality of the valence video, and a subjective quality estimation for obtaining an estimated value EV of the overall subjective quality of the evaluation video from the influence evaluation values S and T In the subjective quality estimation procedure, the estimated quality value EV of the evaluation video is expressed as EV = δ− from the influence evaluation values S, T and predetermined coefficients α, β, γ, δ. It is obtained by (α × S + β × T) + γ × S × T.

In one configuration example of the video quality objective evaluation method of the present invention, the coefficients α, β, γ, and δ are predetermined by subjective evaluation experiments .

Further, the video quality objective evaluation device of the present invention includes a first quality parameter deriving unit for deriving a subjective quality evaluation value considering spatial degradation of the evaluation video and a feature amount of the spatial degradation of the evaluation video, Second quality parameter deriving means for deriving a subjective quality evaluation value in consideration of temporal degradation of the evaluation video and a characteristic amount of temporal degradation of the evaluation video; and the first and second quality parameter deriving means. And integrating means for estimating a total subjective quality of the evaluation video in consideration of the interaction between the spatial degradation and the temporal degradation, and the integrating means derives the first quality parameter. From the result of the first influence evaluation value deriving means for obtaining the evaluation value S of the influence of the spatial degradation on the subjective quality of the evaluation video from the result of the means and the result of the second quality parameter deriving means, the time Degradation A second influence evaluation value deriving unit for obtaining an evaluation value T of the influence on the subjective quality of the evaluation video; and a subjective quality for obtaining an estimated value EV of the overall subjective quality of the evaluation video from the influence evaluation values S and T The subjective quality estimation means calculates the subjective quality estimation value EV of the evaluation video from the influence evaluation values S, T and predetermined coefficients α, β, γ, δ, EV = δ. It is obtained by-(α × S + β × T) + γ × S × T.

The video quality objective evaluation program of the present invention includes a first quality parameter derivation procedure for deriving a subjective quality evaluation value considering spatial degradation of the evaluation video and a feature amount of the spatial degradation of the evaluation video, A second quality parameter derivation procedure for deriving a subjective quality evaluation value considering temporal degradation of the evaluation video and a temporal deterioration feature quantity of the evaluation video; and the first and second quality parameter derivation procedures From the result of the above, the CPU is caused to execute an integration procedure for estimating the overall subjective quality of the evaluation video in consideration of the interaction between the spatial degradation and the temporal degradation . From the result of the quality parameter derivation procedure, the first influence evaluation value derivation procedure for obtaining the evaluation value S of the influence of the spatial degradation on the subjective quality of the evaluation video, and the result of the second quality parameter derivation procedure Or , A second influence evaluation value derivation procedure for obtaining an evaluation value T of the influence of the temporal deterioration on the subjective quality of the evaluation video, and the overall subjective quality of the evaluation video from the influence evaluation values S and T. A subjective quality estimation procedure for obtaining an estimated value EV. The subjective quality estimation procedure uses the influence evaluation values S and T and predetermined coefficients α, β, γ, and δ to determine the subjective quality of the evaluation video. The estimated value EV is obtained by EV = δ− (α × S + β × T) + γ × S × T.

  According to the present invention, by providing an integration procedure, it is possible to consider human perceptual characteristics with respect to spatial degradation and temporal degradation, and spatial degradation and temporal degradation occur simultaneously in the evaluation video. In this case, it is possible to appropriately estimate the overall subjective quality by taking into account the interaction between spatial degradation and temporal degradation, so that subjective assessment is performed by replacing the human subjective assessment method with the present invention. Requires a great deal of labor and time.

Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram showing a configuration of a video quality objective evaluation apparatus according to an embodiment of the present invention.
The video quality objective evaluation apparatus of FIG. 1 includes a first quality parameter derivation unit 11, a second quality parameter derivation unit 12, and an integration unit 14.

  The first quality parameter deriving unit 11 integrates the subjective quality evaluation value SD1 in consideration of the feature amount of spatial degradation generated in the degraded video to be evaluated (hereinafter referred to as evaluation video) and the spatial degradation. 14 to output. Note that the function of the first quality parameter deriving unit 11 can be realized by the technique disclosed in Non-Patent Document 1, but it can be replaced by other known techniques.

  The second quality parameter derivation unit 12 has a function of outputting to the integration unit 14 an evaluation value TD1 of subjective quality that takes into account the temporal deterioration feature amount generated in the evaluation video and the temporal deterioration. Note that the function of the second quality parameter deriving unit 12 can be realized by the technique disclosed in Non-Patent Document 2, but it can be replaced by other known techniques.

  The integration unit 14 has a function of deriving an objective evaluation value EV in consideration of spatial degradation and temporal degradation by integrating the input feature quantity and subjective quality evaluation values SD1 and TD1. Note that, for each of the operations described above, each unit has necessary calculation capability and memory.

Details of each unit will be described below. FIG. 2 is a flowchart showing the operation of the video quality objective evaluation device of this embodiment.
The first quality parameter deriving unit 11 derives the subjective quality evaluation value SD1 in consideration of the feature amount of the spatial degradation generated in the evaluation video and the spatial degradation, and outputs the subjective quality evaluation value SD1 to the integration unit 14 (step in FIG. 2). S1). FIG. 3 is a block diagram illustrating a configuration example of the first quality parameter deriving unit 11. The first quality parameter deriving unit 11 includes three deriving units 11-1, 11-2, and 11-3.

The derivation unit 11-1 uses a so-called full reference method, and when all information of the evaluation video and the original video that is the video before deterioration of the evaluation video can be used, the feature amount and subjective A quality evaluation value SD1 is derived.
The derivation unit 11-2 uses a so-called reduced reference method, and when a part of the original video and information of the evaluation video or a part of the evaluation video and the information of the original video can be used, the feature amount is obtained from the information. And the subjective quality evaluation value SD1 is derived.
The deriving unit 11-3 uses a so-called No reference method. When only the information of the evaluation video can be used, the deriving unit 11-3 derives the feature amount and the subjective quality evaluation value SD1 from this information.

  For example, the feature amount SD1 of spatial degradation that can be derived using all information of the original video and the evaluation video, or part of the original video and information of the evaluation video, or part of the evaluation video and the information of the original video Is the maximum lost edge energy and the maximum additional edge power defined in Reference 1 “ANSI T1.801.03-1996,“ Digital Transport of One-Way Video signals-Parameters for Objective Performance Assessment ””. (Maximum added edge energy) exists. Further, as the spatial degradation feature quantity SD1 that can be derived using only the information of the evaluation video, Reference Document 1 and Reference Document 2 “ITU-T Recommendation P.910,“ Subjective Video quality assessment methods for multimedia ” There is SI (Spatial Information) defined in “applications”, Aug. 1996 ”. The subjective quality evaluation value SD1 is obtained by using a method of expressing the evaluation value of the subjective quality of the evaluation video as an absolute value or a method of expressing the evaluation value of the subjective quality of the evaluation video as a difference value from the original video. There is a quality evaluation value. For example, there are ACR method as a method for expressing subjective quality as an absolute value, and DSIS method and DSCQS method as a method for expressing subjective quality as a difference value from an original video before deterioration. In addition to the above method, other known methods can be used to derive SD1.

  In the present embodiment, in particular, as the spatial degradation feature amount SD1 using all information of the original video and the evaluation video, Ave_EE (Average Edge Energy Difference), Min_HV (Minimum HV- (horizontal or vertical) to non-HV -edge energy difference) and Ave_MEB (Average moving energy of blocks). That is, the deriving unit 11-1 of the first quality parameter deriving unit 11 derives Ave_EE, Min_HV, and Ave_MEB as the feature quantity SD1.

  Ave_EE is a feature quantity that is quantified in consideration of degradation that occurs in the edge portion of the evaluation video. Min_HV is a feature value quantified in consideration of block distortion of the evaluation video. Ave_MEB is a feature amount quantified in consideration of the relationship between the magnitude of the motion of the evaluation video and the locality of degradation. A specific derivation formula for these feature amounts is described in Non-Patent Document 1. These feature quantities are output in units of 10 seconds, but the unit time can take a value other than 10 seconds.

  The second quality parameter deriving unit 12 derives a subjective quality evaluation value TD1 in consideration of the temporal deterioration feature quantity generated in the evaluation video and the temporal deterioration, and outputs the subjective quality evaluation value TD1 to the integration unit 14 (step in FIG. 2). S2). FIG. 4 is a block diagram illustrating a configuration example of the second quality parameter deriving unit 12. The second quality parameter derivation unit 12 includes three derivation units 12-1, 12-2, and 12-3.

The derivation unit 12-1 uses a so-called full reference method, and derives a feature amount and a subjective quality evaluation value TD1 from these information when all information of the original video and the evaluation video can be used.
The derivation unit 12-2 uses a so-called reduced reference method, and when a part of the original video and information of the evaluation video or a part of the evaluation video and the information of the original video can be used, the feature amount is obtained from these information. And the subjective quality evaluation value TD1 is derived.
The derivation unit 12-3 uses a so-called No reference method. When only the information of the evaluation video can be used, the derivation unit 12-3 derives the feature amount and the subjective quality evaluation value TD1.

  For example, as information TD1 of temporal degradation that can be derived using all information of the original video and the evaluation video, or part of the original video and information of the evaluation video, or part of the evaluation video and the information of the original video Is the maximum added motion energy defined in Reference Document 1. Further, as the temporal deterioration feature amount TD1 that can be derived using only the information of the evaluation video, there is TI (Temporal Information) defined in Reference Document 1 or Reference Document 2, and the like. Further, as the subjective quality evaluation value TD1, the subjective quality obtained by using a method of expressing the evaluation value of the subjective quality of the evaluation video as an absolute value or a method of expressing the evaluation value of the subjective quality of the evaluation video as a difference value from the original video. There is a quality evaluation value.

  In the present embodiment, an equivalent freeze length FL that integrates the effects of freeze degradation that has occurred in the evaluation video is used as the temporal deterioration feature amount TD1 that uses only the information of the evaluation video. That is, the deriving unit 12-3 of the second quality parameter deriving unit 12 derives the equivalent freeze length FL as the feature amount TD1. The equivalent freeze length FL is a feature value quantified in consideration of the motion of the evaluation video. A method for deriving the equivalent freeze length FL is described in Non-Patent Document 2. This feature value is output in units of 10 seconds, but the unit time can take a value other than 10 seconds.

  Next, the integration unit 14 uses the feature quantity and subjective quality evaluation value SD1 input from the first quality parameter derivation unit 11 and the feature quantity and subjective quality evaluation value TD1 input from the second quality parameter derivation unit 12. The total subjective quality is estimated in consideration of the interaction between the effects of spatial degradation and temporal degradation (step S3 in FIG. 2). FIG. 5 is a block diagram illustrating a configuration example of the integration unit 14. The integration unit 14 includes a first influence evaluation value derivation unit 14-1, a second influence evaluation value derivation unit 14-2, and a subjective quality estimation unit 14-3.

First, an objective evaluation value EV, which is an estimated value of overall subjective quality, is calculated as follows.
EV = δ− (α × S + β × T) + γ × S × T (1)
In Equation (1), S is an evaluation value of the influence of spatial degradation on subjective quality, T is an evaluation value of the influence of temporal degradation on subjective quality, and α, β, γ, and δ are coefficients.

The evaluation value S of the influence of the spatial deterioration on the subjective quality can be obtained from the feature quantity input from the first quality parameter deriving unit 11 and the subjective quality evaluation value SD1 as follows.
S = a ₁ × S ₁ + a ₂ × S ₂ +... + _An × S _n + a _{n + 1} (2)
In the formula _{(2), S 1, S} 2, ···, S n the feature and the subjective quality assessment value belonging to SD1, n is the number of feature quantity and subjective quality assessment value constituting the SD1, a _1, a ₂ ,..., A _n , a _{n + 1} are coefficients.

The evaluation value T of the influence of temporal deterioration on the subjective quality can be obtained from the feature amount input from the second quality parameter deriving unit 12 and the subjective quality evaluation value TD1 as follows.
T = b ₁ × T ₁ + b ₂ × T ₂ +... + B _m × T _m + b _{m + 1} (3)
In Equation (3), T ₁ , T ₂ ,..., T _m are feature quantities and subjective quality evaluation values belonging to TD1, m is the number of feature quantities and subjective quality evaluation values constituting TD1, b ₁ , b ₂ ,..., B _m , b _{m + 1} are coefficients.

In the present embodiment, n = 3, m = 1, evaluation values S ₁ = Ave_EE, S ₂ = Min_HV, S ₃ = Ave_MEB, T ₁ = FL, and Expression (2) is expressed as Expression (4). Then, equation (3) was set as equation (5).
S = a ₁ × [Ave_EE] + a ₂ × [Min_HV] + a ₃ × [Ave_MEB]
+ A ₄ (4)
T = b ₁ × [FL] + b ₂ (5)

Coefficients of equations (1) to (5) α, β, γ , δ, a 1, a 2, ···, a n, a n + 1, b 1, b 2, ···, b m, b _{m + 1} is determined in advance so as to be optimized by a subjective evaluation experiment. As a specific method of the optimization, an objective evaluation value EV obtained by the video quality objective evaluation apparatus of the present embodiment from an evaluation video and a comprehensive subjective evaluation value E determined by a human looking at the evaluation video are used. The square of the error is obtained for the evaluation images of the various deterioration conditions 1 to i in Table 1, and the coefficients α, β, γ, δ, a ₁ , a ₂ , .., A _n , a _{n + 1} , b ₁ , b ₂ ,..., B _m , b _{m + 1} may be optimized. The deterioration conditions 1 to i define differences in evaluation scenes, types of deterioration, and strength of the deterioration.

For example, according to Table 1, for an evaluation video under a certain degradation condition 1, the feature quantity S ₁ (= Ave_EE) obtained by the video quality objective evaluation apparatus is S ₁₁ , the feature quantity S ₂ (= Min_HV) is S ₁₂ , and the feature quantity. S ₃ (= Ave_MEB) is S ₁₃ , the characteristic amount T ₁ is T ₁₁ , the comprehensive subjective evaluation value determined by humans is E ₁ , and the objective evaluation value is EV ₁ . The subjective evaluation value is obtained in advance by performing a subjective evaluation experiment on the evaluation video of each deterioration condition 1 to i. Equations (1), (2), (3), and (3) so that the sum of the squares of the errors (F _{1 to} F _i ) between the subjective evaluation value and the objective evaluation value of each deterioration condition shown in Table 1 is minimized. Regression analysis was performed using equations (4) and (5) to determine the coefficients of each equation. Table 1 shows the case where n = 3 and m = 1, that is, the case where the evaluation values S and T are obtained by the equations (4) and (5).

As described above, regression analysis is performed so that the overall subjective quality can be accurately estimated from the subjective quality representing each feature amount and spatial / temporal degradation, and the coefficients α, β, γ, and δ in Equation (1) are calculated. In addition, the coefficients a ₁ , a ₂ ,..., A _n , a _{n + 1} , b ₁ , b ₂ ,..., B _m , b _{m + 1} ( When equations (4) and (5) are used, the coefficients a ₁ , a ₂ , a ₃ , a ₄ , b ₁ , b ₂ ) are determined. In this embodiment, an Excel solver program developed by Frontline Systems, Inc. was used as an algorithm for regression analysis. However, other regression analysis algorithms may be used.

Coefficient thus determined α, β, γ, δ, a 1, a 2, ···, a n, a n + 1, b 1, b 2, ···, b m, b m + 1 Is substituted into Equation (1), Equation (2), Equation (3), or similarly obtained coefficients α, β, γ, δ, a ₁ , a ₂ , a ₃ , a ₄ , b ₁ , Substituting b ₂ into Equation (1), Equation (4), and Equation (5) completes the overall subjective quality estimation equation.

  The first impact evaluation value deriving unit 14-1 of the integration unit 14 derives the impact evaluation value S using the formula (2) or the formula (4), and the second impact evaluation value deriving unit 14-2 The influence evaluation value T is derived using Expression (3) or Expression (5), and the subjective quality estimation unit 14-3 derives the objective evaluation value EV using Expression (1).

  Expressions (1) to (5) are expressions that take into account the interaction between spatial degradation and temporal degradation. Specifically, the evaluation value S in the equation (1) is a value considering spatial degradation, the evaluation value T is a value considering temporal degradation, and S × T is spatial degradation and temporal degradation. Is a value that takes into account. In addition, the degradation conditions 1 to i of the evaluation video shown in Table 1 are sufficiently degraded with respect to the scene type, spatial degradation, and temporal degradation so that an appropriate coefficient is optimized according to the application. To cover the range of types and their strengths. Since the range of each coefficient is greatly different for each coefficient, it is necessary to subjectively determine the range for each coefficient within a practical range. The subjective evaluation value E is obtained in advance by performing a subjective evaluation experiment on the evaluation video of each deterioration condition 1 to i.

  As described above, in the present embodiment, by providing the integration unit 14, human perceptual characteristics with respect to spatial degradation and temporal degradation can be considered, and spatial degradation and temporal degradation are considered. The overall subjective quality can be estimated appropriately by taking into account the interaction of. As a result, in the present embodiment, it is possible to appropriately estimate the subjective quality in a situation where spatial degradation and temporal degradation occur simultaneously. In this embodiment, for example, in a situation where the bit rate is low and degradation due to encoding has occurred, loss in the video transmission path is caused, so that estimation of subjective quality is effectively performed when freeze degradation has occurred. Can be done.

  Note that the video quality objective evaluation device of the present embodiment can be realized by a computer having a CPU, a storage device, and an external interface, and a program for controlling these hardware resources. In such a computer, the video quality objective evaluation program for realizing the video quality objective evaluation method of the present invention is provided in a state of being recorded on a recording medium such as a flexible disk, a CD-ROM, a DVD-ROM, or a memory card. Is done. The CPU writes the program read from the recording medium into the storage device, and executes the above-described processing according to the program.

  The present invention can be applied to a technique for estimating human subjective quality from measurement of a feature amount of a video signal without performing a subjective quality evaluation experiment in which a human views a video and evaluates its quality.

It is a block diagram which shows the structure of the video quality objective evaluation apparatus which concerns on embodiment of this invention. It is a flowchart which shows operation | movement of the video quality objective evaluation apparatus of FIG. It is a block diagram which shows the structural example of the 1st quality parameter derivation | leading-out part in the video quality objective evaluation apparatus of FIG. It is a block diagram which shows the structural example of the 2nd quality parameter derivation | leading-out part in the video quality objective evaluation apparatus of FIG. It is a block diagram which shows the structural example of the integration part in the video quality objective evaluation apparatus of FIG.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 11 ... 1st quality parameter deriving part, 12 ... 2nd quality parameter deriving part, 14 ... Integration part, 11-1 to 11-3, 12-1 to 12-3 ... Deriving part, 14-1 ... 1st 14-2... Second influence evaluation value derivation unit, 14-3... Subjective quality estimation unit.

Claims (4)

A video quality objective evaluation method for estimating the subjective quality of video,
A first quality parameter derivation procedure for deriving a subjective quality evaluation value in consideration of spatial degradation of the evaluation video and a feature quantity of the spatial degradation of the evaluation video;
A second quality parameter derivation procedure for deriving a subjective quality evaluation value in consideration of temporal degradation of the evaluation video and a characteristic amount of temporal degradation of the evaluation video;
An integrated procedure for estimating a total subjective quality of the evaluation video in consideration of the interaction between the spatial degradation and the temporal degradation from the results of the first and second quality parameter derivation procedures ;
The integration procedure is:
A first influence evaluation value derivation procedure for obtaining an evaluation value S of the influence of the spatial degradation on the subjective quality of the evaluation video from the result of the first quality parameter derivation procedure;
From the result of the second quality parameter derivation procedure, a second influence evaluation value derivation procedure for obtaining an evaluation value T of the influence of the temporal deterioration on the subjective quality of the evaluation video;
A subjective quality estimation procedure for obtaining an estimated value EV of the overall subjective quality of the evaluation video from the influence evaluation values S and T,
In the subjective quality estimation procedure, the estimated value EV of the subjective quality of the evaluation video is expressed as EV = δ− (α × S + β ×) from the influence evaluation values S and T and predetermined coefficients α, β, γ, and δ. T) A video quality objective evaluation method characterized by being obtained by + γ × S × T. The video quality objective evaluation method according to claim 1,
The video quality objective evaluation method , wherein the coefficients α, β, γ, and δ are determined in advance by a subjective evaluation experiment . An objective video quality evaluation device that estimates the subjective quality of video,
First quality parameter deriving means for deriving a subjective quality evaluation value considering spatial degradation of the evaluation video and a feature amount of the spatial degradation of the evaluation video;
Second quality parameter derivation means for deriving a subjective quality evaluation value considering temporal degradation of the evaluation video and a characteristic amount of temporal degradation of the evaluation video;
Integrating means for estimating a total subjective quality of the evaluation video in consideration of the interaction between the spatial degradation and the temporal degradation from the results of the first and second quality parameter derivation means;
The integration means includes
First influence evaluation value deriving means for obtaining an evaluation value S of the influence of the spatial degradation on the subjective quality of the evaluation video from the result of the first quality parameter deriving means;
Second influence evaluation value deriving means for obtaining an evaluation value T of the influence of the temporal deterioration on the subjective quality of the evaluation video from the result of the second quality parameter deriving means;
Subjective quality estimation means for obtaining an estimated value EV of the overall subjective quality of the evaluation video from the influence evaluation values S and T,
The subjective quality estimation means calculates the estimated value EV of the subjective quality of the evaluation video from the influence evaluation values S, T and predetermined coefficients α, β, γ, δ, EV = δ− (α × S + β × T) An objective video quality evaluation apparatus characterized by being obtained by + γ × S × T. A video quality objective evaluation program for operating a computer comprising a CPU and a storage device as a video quality objective evaluation device for estimating subjective quality of video,
A first quality parameter derivation procedure for deriving a subjective quality evaluation value in consideration of spatial degradation of the evaluation video and a feature quantity of the spatial degradation of the evaluation video;
A second quality parameter derivation procedure for deriving a subjective quality evaluation value in consideration of temporal degradation of the evaluation video and a characteristic amount of temporal degradation of the evaluation video;
An integrated procedure for estimating an overall subjective quality of the evaluation video in consideration of the interaction between the spatial degradation and the temporal degradation from the results of the first and second quality parameter derivation procedures; To run
The integration procedure is:
A first influence evaluation value derivation procedure for obtaining an evaluation value S of the influence of the spatial degradation on the subjective quality of the evaluation video from the result of the first quality parameter derivation procedure;
From the result of the second quality parameter derivation procedure, a second influence evaluation value derivation procedure for obtaining an evaluation value T of the influence of the temporal deterioration on the subjective quality of the evaluation video;
A subjective quality estimation procedure for obtaining an estimated value EV of the overall subjective quality of the evaluation video from the influence evaluation values S and T,
In the subjective quality estimation procedure, the estimated value EV of the subjective quality of the evaluation video is expressed as EV = δ− (α × S + β ×) from the influence evaluation values S and T and predetermined coefficients α, β, γ, and δ. T) An image quality objective evaluation program characterized by being obtained by + γ × S × T.

Images