Paper
4 February 2013 Performance evaluation of HD camcorders: measuring texture distortions using Gabor filters and spatio-velocity CSF
Kongfeng Zhu, Dietmar Saupe
Author Affiliations +
Proceedings Volume 8653, Image Quality and System Performance X; 86530A (2013) https://doi.org/10.1117/12.2002495
Event: IS&T/SPIE Electronic Imaging, 2013, Burlingame, California, United States
Abstract
This paper presents a new method of measuring physical texture distortions (PhTD) to evaluate the performance of high de_nition (HD) camcorders w.r.t. motion and lossy compression. It is extended to measure perceptual texture distortions (PeTD) by taking into account the spatio-velocity contrast sensitivity function of the human visual system. The PhTD gives an objective (physical) distortion of texture structures, while the PeTD measures the perceptual distortion of textures. The dead leaves chart, invariant to scaling, translation, rotation, and contrast, was selected as a target texture. The PhTD/PeTD metrics of the target distorted by camcorders were measured based on a bank of Gabor _lters with eight orientations and three scales. Experimental results for six HD camcorders from three vendors showed: 1) the PhTD value increases monotonically w.r.t. the motion speed, and decreases monotonically w.r.t. the lossy compression bitrate; 2) the PeTD value decreases monotonically w.r.t. the motion speed, but stays almost constant w.r.t. the lossy compression bitrate. The experiment gives a reasonable result even if the distortions are not radially symmetric. However, some subjective tests should be done in future work to validate the performance of the perceptual texture distortion metric.
© (2013) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Kongfeng Zhu and Dietmar Saupe "Performance evaluation of HD camcorders: measuring texture distortions using Gabor filters and spatio-velocity CSF ", Proc. SPIE 8653, Image Quality and System Performance X, 86530A (4 February 2013); https://doi.org/10.1117/12.2002495
Lens.org Logo
CITATIONS
Cited by 3 scholarly publications.
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Video

Image compression

Video compression

Imaging systems

Motion measurement

Contrast sensitivity

Cameras

Back to Top