Available Online at www.ijcsmc.com
International Journal of Computer Science and Mobile Computing
A Monthly Journal of Computer Science and Information Technology
ISSN 2320–088X
IJCSMC, Vol. 2, Issue. 4, April 2013, pg.58 – 64
RESEARCH ARTICLE
© 2013, IJCSMC All Rights Reserved 58
Perceptual Video Quality Measurement
Based on Generalized Priority Model
S. Bharath
1
, S. Jaganath
2
, J. Prakash
3
1
Assistant Professor, Department of ECE, PSNA College of Engineering and Technology, Dindigul, India
2
Assistant Professor, Department of ECE, PSNA College of Engineering and Technology, Dindigul, India
3
Assistant Professor, Department of ECE, PSNA College of Engineering and Technology, Dindigul, India
1
bharath.psna@gmail.com;
2
s.jaganhsd@gmail.com;
3
jeyavelprakash@gmail.com
Abstract— We consider factors not only in a packet, but also in its locality, to account for possible temporal
and spatial masking effects. We apply our visibility model to packet priority for a video stream, when the
network gets jam-packed at an in-between router; the router is able to choose which packets to drop such that
visual quality of the video is minimally crashed. To show the effectiveness of our visibility model and its
corresponding packet priority method, experiments are done to compare our perceptual-quality-based packet
priority approach with existing Drop tail & hint track, Mean square error priority methods. The result shows
that our priority method produces videos of higher perceptual quality for different network conditions. Our
model was developed using data from high encoding-rate videos, and designed for high-quality video sent
over a mostly reliable network; however, the experiments show the model is valid to different encoding rates.
Key Terms: - Packet dropping policy; packet loss; perceptual video quality; video coding; visibility model
I. INTRODUCTION
The growing popularity of transmitting compressed video over the Internet increases the need for quality
assessment methods that can accurately characterize how the network is affecting the video quality seen by the
end-user. Transmitting video in digital form is the direct result of the benefits offered by digital compression.
The potential impact of multimedia information is currently restricted by the bandwidth of the existing
communication networks. Quality of Service (QoS) of Networks are of more importance, especially due to the
need of Synchronization (frame-rate of video must be maintained) of video streams. The Burst nature of
compressed video stream is the major hurdle for Network technologies to maintain QoS.
Video quality measurement in the network can be categorized into three different types based on the
accessibility of information about the original (reference) video. Full-reference (FR) methods evaluate the video
quality with access to the original video, providing the most precise measurements on the video quality
difference. Reduced-reference (RR) metrics extract partial information about the original video at the sender and
are sent reliably to the receiver to estimate the video quality. No-reference (NR) methods only use information
available in the bit stream or the decoded pixels without reference video information [1].
Packet losses in the network (for example, due to congestion) can significantly damage video quality during
transmission. Therefore, considerable research has been conducted to understand the relationship between
packet losses and visual quality degradation. Although PSNR (Peak Signal to Noise Ratio) and MSE do not
always reflect perceptual quality well, they are commonly used to measure video quality. The relation between
PSNR and perceptual quality scores is considered in [2]. It finds that packet losses are visible when the PSNR
drop is greater than a threshold, and the distance between dropped packets is crucial to perceptual quality.