Share Email Print

Proceedings Paper • Open Access

Visual Signal Analysis: Focus on Texture Similarity (Presentation Recording)

Paper Abstract

Texture is an important visual attribute both for human perception and image analysis systems. We present new structural texture similarity metrics and applications that critically depend on such metrics, with emphasis on image compression and content-based retrieval. The new metrics account for human visual perception and the stochastic nature of textures. They rely entirely on local image statistics and allow substantial point-by-point deviations between textures that according to human judgment are similar or essentially identical.

We also present new testing procedures for objective texture similarity metrics. We identify three operating domains for evaluating the performance of such similarity metrics: the top of the similarity scale, where a monotonic relationship between metric values and subjective scores is desired; the ability to distinguish between perceptually similar and dissimilar textures; and the ability to retrieve "identical" textures. Each domain has different performance goals and requires different testing procedures. Experimental results demonstrate both the performance of the proposed metrics and the effectiveness of the proposed subjective testing procedures.

Paper Details

Date Published: 5 October 2015
PDF: 1 pages
Proc. SPIE 9598, Optics and Photonics for Information Processing IX, 959802 (5 October 2015); doi: 10.1117/12.2193656
Show Author Affiliations
Thrasyvoulos N. Pappas, Northwestern Univ. (United States)

Published in SPIE Proceedings Vol. 9598:
Optics and Photonics for Information Processing IX
Abdul A. S. Awwal; Khan M. Iftekharuddin; Mohammad A. Matin; Mireya García Vázquez; Andrés Márquez, Editor(s)

© SPIE. Terms of Use
Back to Top