Perceptual Evaluation of
Video Quality
PEVQ provides mean opinion score (MOS) estimates of the video quality degradation occurring through a network, e.g. in video telephony networks. PEVQ can be ideally applied to test video streaming, video conferencing and IPTV applications. The degraded video signal output from a network is analyzed by comparison to the undistorted reference video signal on a perceptual basis. PEVQ is based on the approach to model the human visual system with which anomaly perceived in the video signal can be detected and quantified by a multitude of key performance indicators (KPIs).
OPTICOM, the leading provider of signal based perceptual measurement technology for voice, audio and video, offers its new easy to use
PEVQ V 2.0 measurement as an OEM solution. PEVQ evaluates the quality of CIF, QCIF, VGA and Rec. 601 video formats based on perceptual measurement, reliably, objectively and fast.

PEVQ is based on the earlier PVQM technology developed by KPN Research and new developments of OPTICOM.

> Principle PEVQ is a full reference, intrusive measurement algorithm for video quality. Its basic structure is shown in the figure below.
click to [ enlarge picture ]

> Key Features
Analysis of video quality

Outputs MOS score that correlates well with subjective MOS

Additional KPIs for detailed analysis
Measurement with QCIF (real time), CIF, VGA and Rec. 601 video sizes
Perceptually based end-to-end quality analysis
Available as DLL/Library or Source
Applications: IPTV, streaming video, 3G, video telephony
The algorithm can be divided up into four separate blocks. The first block – pre-processing stage – is responsible for the spatial and temporal alignment of the reference and the impaired signal. This process ensures that only corresponding frames are compared to each other.

The second block calculates the perceptual difference of the aligned signals. Perceptual means that only those differences are taken into account which are actually perceived by a human viewer. Furthermore the activity of the motion in the reference signal provides another indicator representing the temporal information.

This indicator is important as it takes into account that in frame series with low activity the perception of details is much higher than in frame series with quick motion.

The third block in the figure classifies the previously calculated indicators and detects certain types of distortions.

Finally, in the fourth block all the appropriate indicators according to the detected distortions are aggregated, forming the final result - the mean opinion score (MOS). The MOS value describes the video quality on a range from 1 for very bad quality, to 5 for excellent quality.

> Specifications PEVQ - Perceptual Evaluation of Video Quality
Functionality Perceptual Evaluation of Video Quality- PEVQ, with mapping to MOS scale (1 bad, … 5 excellent quality)
Input data AVI with RGB24, YUV video data

QCIF, CIF, VGA and Rec. 601 frame sizes

  6 to 20 seconds in length
Complexity V2.0 is widely optimized and is capable of processing QCIF streams faster than realtime on a 3GHz Pentium 4 machine.
Platforms Windows
(others on request)
> Output
The PEVQ MOS value lies within a range from 1 (bad) to 5 (excellent). The PEVQ MOS is based on a multitude of perceptually motivated parameters.
Distortion indicators
For a more detailed analysis the perceptual level of distortion in the luminance, chrominance and temporal domain are provided.
The delay of each frame of the test signal related to the reference signal.
The brightness of the reference and degraded signal.
The contrast of the distorted and the reference sequence.
To allow for a coarse analysis of he distortions in different domains the PSNR is provided for the Y, Cb and Cr components separately.
describes the smoothness of a video playback which is often impaired by down-sampling, coding processes and perturbed transmissions.
is a distortion characterized by reduced sharpness of contour edges and spatial detail.
is often the result of a low bit rate coding that uses a block matching algorithm for the motion estimation and a coarse quantization for the image blocks.
Frame Skips and Freezes
are temporal artefacts occurring in video transmissions caused by e.g. overloaded networks.
Effective Frame Rate
Down-sampling of a video signal on a frame by frame basis often results in loss of information which often leads to the degradation of the video signal. The effective frame rate is an indicator quantifying the severeness of such a process.
Temporal and Spacial Activity
Temporal and spacial activity indicators quantify the amount of activity /movement in the video content. These indicators are derived from ITU-T recommendation P.910.
> Download more detailed information

Download PEVQ – Perceptual Evaluation of Video Quality
[as PDF / 534 kb]


Download OPTICOM PEVQ V2.0 Newsletter
[as PDF / 257 kb]