DE102005040301A1 - A method for the automatic objective technical quality assessment of the content of audio and / or video material and for filtering quality-critical sequences from this material - Google Patents

Abstract

Method for the automatic objective technical quality evaluation of the content of audio and / or video material and for filtering up quality-critical sequences with the following steps: DOLLAR A a) recording a series of measurements of several measured values of a quality characteristic of the audio and / or video content and recording at least a further series of measurements of at least one further quality feature of the audio and / or video content, each measurement value having a temporal relationship to the examined audio and / or video content, DOLLAR A b) application of predetermined, for each individual quality feature -Measurement series of specific conditions on the measured values obtained in step a), in order to filter out those time segments of the analyzed material which conform to the stated conditions, one condition being a comparison of a measured value with an upper limit value defined for its measured variable and / or with a set accordingly represented lower limit value, DOLLAR A c) performing a logical combination of the filter results of all quality features and DOLLAR A d) indexing the quality-critical audio and / or video sequences depending on the result of the logical combination.

Description

The The invention relates generally to methods for evaluating the technical quality of audio / video content. In particular, the invention relates to a Method for automated objective quality measurement of audio / video content with the aim of an automated and objectified filtering out such excerpts from the audio / video content, which is a subjective Revaluation supplied Need to become. That is, the invention relates in its core to a method which is suitable for automatic objective quality technical evaluation of the content of Audio and / or video material and for filtering out quality-critical Sequences from this material is used. In addition, the invention relates a device for carrying out this procedure.

in the Prior art is the quality of audio / video content usually checked by subjective, manual full control. This is very time consuming and error prone. Lens and automated can be the quality multimedia content according to the state of the art measure satisfactory accuracy, since the existing for this in the prior art Procedures and devices are unable, in any case Quality defects of desired style agents such as e.g. False colors, intentional bucking and Like to distinguish.

in the State of the art there are various, partly standardized Quality metrics, the statements about Properties of audio / video content. These metrics are available from commercially available Software and hardware solutions determined and issued. Those known from the prior art Metrics can used to the effort of subjective quality measurement significantly reduce. In many cases, doing so as a result the measurement over the entire playing time of the content is averaged. These However, information is for an assessment of the technical quality of the content is not sufficient, because short excerpts with greatly reduced quality are in These values do not reflect, but quite an exclusion criterion for the Use of the content. Therefore, in the state of Technique a subsequent evaluation of these sections by human, i.e. subjective rating required.

The Man-perceived or perceived technical quality of audio / video content can be determine by subjective and objective measurement methods.

One Well-known subjective rating method is Double Stimulus Continuous Quality Scale (DSCQS). This is one of the International Telecommunication Union (ITU) recommended procedure (see Recommendation ITU-R BT.500-9: "Methodology for the Subjective Assessment of the Quality of Television Pictures ", 1998 and ITU-T Recommendation P.910: "Subjective video quality assessment methods for multimedia applications ", 1999) This method is a small context-dependent influence on the result. A disadvantage of this method, the high cost of its implementation and the height the associated costs.

On there is currently no standard in the field of objective evaluation, which bindingly prescribes a procedure to improve the quality of audio / video content to rate. However, several methods have been used in the prior art who are trying to use the Human Visual System (HVS) as possible perfect picture. The HVS is a model that visual Describes perception by humans. To the latter Procedures include among others the Webster approach (see A. A. Webster et al .: "An Objective Video Quality Assessment System Based on Human Perception ".- In: Human Vision, Visual Processing, and Digital Display IV, San Jose (CA), 1993, Pages 15 to 26.), the Moving Picture Quality Metric (MPQM) by Van den Branden Lambrecht (see C. J. van den Branden Lambrecht: "A working spatio-temporal model of the human visual system for image restoration and quality assessment applications ".- In: Proc. IEEE Int. Conf. Acoust., Speech, and Signal Processing, 1996, pages 2291 to 2294.) as well as the group of so-called JND (Just Noticable Difference) based Method.

Further Method for the objective quality assessment known from the prior art of audio / video content are the fully automatic video quality Model (VQM) of the National Telecommunications and Information Administration (NTIA) and the Objective Perceptual Video Quality Measurement Method Tektronix based on JND.

all above-mentioned method for the objective evaluation of the technical quality However, audio / video content has in common that they are not high reliability in the detection of technical defects, since they are not in the Are able, between actual technical defects and artistically different stylistic devices used in audio / video production.

The invention is therefore based on the object, a method for automatic objective provide technical quality assessment of the content of audio and / or video material and to filter out quality critical sequences which, better than the methods known in the art, are able to distinguish between actual technical defects and artistic stylistic devices of the audio / video Distinguish production. Furthermore, the invention has the object to provide a device for carrying out the method according to the invention.

According to the invention this Task solved by a method according to claim 1 and by an apparatus according to claim 8th.

The significantly improved performance of the inventive method across from the known from the prior art method results in essential in that not only one quality feature is subjected to a time series analysis, but that for several quality features Time series analyzes performed and finally even a logical link this separate time series takes place.

advantageous and preferred developments of the method according to the invention are the subject the claims 2 to 7.

advantageous and preferred embodiments The invention will be explained below with reference to figures. It shows:

1 schematically an extraction of significant scenes according to an embodiment of a method according to the invention and

2 schematically an embodiment of a working using the method according to the invention quality control station.

A method according to the invention uses the result of a performed objective quality assessment of audio / video material. This is a series of measurements, consisting of n discrete measured values of one or more quality features, each measured value being temporally related to the examined audio / video content (see 1 ). ( x 1 , t 1 ), ..., (x n , t n ) Ε A (Measurement series with a measured variable x and n measured values) (x 1 , Y 1 , z 1 , t 1 ), ..., (x n , y n , z n , t n ) Ε A (Measurement series with three measured quantities x, y, z and in each case n measured values)

Express is noted that the above-mentioned series of measurements with three measured quantities only a very special embodiment is. The invention may e.g. also be carried out with a series of measurements, which two measured quantities or four measured quantities or also includes even more measures, wherein every measurand a quality feature represents. Such a multi-dimensional series of measurements is natural nothing more than a total of two, three, four or more Measurement series, the each represent a separate measurand, where the measured values the individual measured variables time synchronous were recorded as shown in the previous paragraph three-dimensional series of measurements as an embodiment is shown.

• – PSNR (Peak Signal to Noise Ratio – Verhältnis des Signal-Spitzenwerts zum Rauschen),
• – MSE (Mean Square Error – mittlerer quadratischer Fehler),
• – Blurriness – Grad der „Verschwommenheit", „Unschärfe",
• – Blockiness – Grad der Ausbildung von Blockartefakten, hauptsächlich verursacht durch Kompressionsverfahren, und
• – Jerkiness – Maß für Ruckeln der Videodarstellung, nicht konstante Bildwiederholrate, hauptsächlich verursacht durch Übertragungsprobleme.

Possible measured variables (quality features) in the video area include:

• PSNR (Peak Signal to Noise Ratio - signal peak to noise ratio),
• MSE (mean square error),
• - Blurriness - degree of "blurriness", "blur",
• - Blockiness - Level of training of block artifacts, mainly caused by compression methods, and
• - Jerkiness - measure of juddering of video presentation, non-constant frame rate, mainly caused by transmission problems.

These and more video quality features as such are detailed described in: Winkler, Stefan: Digital Video Quality - Vision Models and Metrics.- John Wiley & Sons Ltd.- Chichester.- 2005.- ISBN 0-470-02404-6.

• – SNR (Signal to Noise Ratio) – Signal-Rausch-Abstand,
• – Max Peak – maximaler Audiopegel,
• – Analog over – Anzeige von Übersteuerung bei der Verarbeitung analoger Audiosignale,
• – Digital Clipping – Anzeige von Übersteuerung bei der Verarbeitung digitaler Audiodaten,
• – Digital Zero – digitale Nullstellen und
• – Block repeating – Blockwiederholungen, verursacht beim Einlesen von CDs.

Possible measured variables (quality features) in the audio domain include:

• SNR (Signal to Noise Ratio) Signal to Noise Ratio
• - Max Peak - maximum audio level,
• - Analog over - Display of overload during the processing of analog audio signals,
• - Digital Clipping - Display of clipping during the processing of digital audio data,
• - Digital Zero - Digital Zeros and
• - Block repeating - block repeats caused when reading in CDs.

Also these audio quality features as such, those skilled in the art from the prior art, so that yourself a more detailed Taking care of it at this point is unnecessary.

Determined conditions (rules R) are applied to the set of measured values to produce the temporal sections of the analyzed material filter out that conform to the conditions.

A Condition represents the comparison of a measured value with for set the measured variable Limit (s), the condition being met if a lower or exceeded the Limits is synonymous with lower audio / video quality.

lx_u

unterer Grenzwert für das Merkmal x

lx₀

oberer Grenzwert für das Merkmal x

Corresponding limit values are defined for each measured variable. R1: lx u ≤ x ≤ lx O R2: ly u ≤ y ≤ ly O R3: lz u ≤ z ≤ lz O

lx _u

lower limit for the characteristic x

lx ₀

upper limit for the characteristic x

tw

definiertes Zeitfenster für das Merkmal x

In order to suppress the effects of exceeding or exceeding limit values due to measurement errors or spontaneous outliers, which are not perceived by human observation, the conditions can be extended to several temporally successive measured values. R1: (lx u ≤ x m ≤ lx O ) ∧ (lx u ≤ x m + 1 ≤ lx O ) ∧ ... ∧ (lx u ≤ x m + tw ≤ lx O )

tw

defined time window for the characteristic x

In the case of several measured variables included in the evaluation, the respectively defined conditions are logically linked together. R = R1 v R2 v R3

After applying the conditions, a set of temporal coordinates is available, each representing the sections that are compliant with the conditions and thus considered to be technically poor in quality. (x a , t a ), ..., (x z , t z ) Ε B (Filter result with one measurand x) (x a , y a , z a , t a ), ..., (x z , y z , z z , t z ) Ε B (Filter result with three measured quantities x, y, z) B ⊆ A

Based this result set can the identified sections of the audio / video material directly from the time information be addressed and thus automatically a person for the purpose of subjective Follow-up be forwarded.

For example, the disclosed invention enables the simplification of a quality control process for audio / video content. Such a control process can, as in 2 represented, are largely automated. The system specifically presents the quality supervisor with scenes that have been identified as deficient. This considerably reduces the time required compared to a manual full control. In addition, the influence of subjective inaccuracies such as fatigue or fluctuating attention, which plays a significant role during a prior art manual full control, is greatly reduced.

An example of an application of a device according to the invention in the context of an automated quality control station is in 2 schematically illustrated and will be described below.

Of the on its technical quality towards audible audio / video content is in a directory monitored by a control system of the device stored. The analysis automatically starts with the help of the corresponding analysis tools integrated into the system. Is this Analysis ends, the system evaluates the results according to the further procedure described above. For this purpose, appropriate limits be defined in the system. These can be used as evaluation profiles manage within the application. The results are on the one hand visually as a chart in which limit violations are prominently displayed, prepared and presented on the other hand as a scene list. The scenes can be the Select quality controller by means of an integrated playback option (for example, media players) directly and on an online provided Rate and document the form.

at another embodiment a device according to the invention is an automatically monitored Input directory for the reception of video files ready. By the delivery A video file automatically becomes an objective quality analysis initiated, which as a result a file with the values of the individual measured variables - in concrete embodiment Blurriness, Blockiness and Jerkiness - delivers. Following the quality analysis becomes the measured value file automatically evaluated. The measured values of the individual measured variables are included the for defined the respective measured variable Thresholds compared, with additionally defined time window act as a filter criterion.

Subsequently, will for every Measured variable a diagram generated, which represents the time course of the measured variable. By means of the method according to the invention according to the pre-defined rules found out quality critical Scenes are clearly highlighted. In an automatic generated table are all classified as quality critical scenes with associated Time code listed. This marks what quality feature for the Classification as "quality critical" crucial was and how strong an undershooting or exceeding of a threshold value took place. Each listed scene can be started by a reference and be played. At the same time, for the selected scene provided a form which is a subjective rating Analysis allowed.

Already Manually post-controlled scenes are visually marked in the list. With the aid of a symbolism it can be recognized whether this scene is also in the manual, i.e. subjective, rating was rated as quality critical.

Claims (8)

Method for automatic objective technical quality rating content of audio and / or video material and for filtering out quality critical Sequences with the following steps: a) Recording a series of measurements of several measured values a quality characteristic of the audio and / or video content and recording at least one further series of measurements of measured values at least one further quality feature of the audio and / or Video content, with each reading a temporal reference to the examined audio and / or video content Has, b) Application of pre-defined, more specific for each quality characteristic series of measurements Conditions on the measured values obtained in step a), such temporal Filter out sections of the analyzed material that conform to the conditions mentioned, where a condition is a comparison of a measured value with a for set the measured variable upper limit and / or with a corresponding lower limit Limit value represents, c) To run a logical link the filter results of all quality features and d) indexing the quality critical audio and / or video sequences dependent on from the result of the logical link. Method according to claim 1, characterized in that that as a quality feature one of the following quality characteristics used for objective technical quality assessment: Blurriness, Blockiness, Jerkiness, MSE, PSNR, Digital Zeros, digital overload, Block repeats, analog override, Signal-to-noise ratio, maximum level. Method according to one of the preceding claims, characterized characterized in that at least one of the conditions according to step b) only then fulfilled is treated when the upper or lower limit is at least for one predetermined coherent Time is injured. Method according to one of the preceding claims, characterized characterized in that Step a) exactly three quality features measured and based on the filtering procedure according to steps b) and c) become. Method according to claim 2, characterized in that that two or more of the quality features mentioned in claim 2 for objective technical quality rating be used. Method according to one of the preceding claims, characterized characterized in that in step d) indexed quality critical Audio and / or video sequences directly addressed and on one Playback device are played. Method according to one of the preceding claims, characterized characterized in that in step d) indexed quality critical Audio and / or video sequences automatically as a chart and / or as a Scene list are visualized. Apparatus for carrying out the method according to one of the preceding claims, characterized in that - the device is designed so that they accept the audio and / or video content to be evaluated and can automatically process according to the above procedure and that - the device is further designed so that they a definition and technical management of different limit rates as Evaluation profiles for different Audio and / or video content.