Perceptual indexing of visual information

Research output: Contribution to journalArticle

Abstract

The application of Human perceptual models in image and video coding is motivated by the fact that non-perceptual distortion metrics (mean square error) do not correlate well with the perceived quality at lower bit-rates despite their acceptable signal to noise ratio. In this paper, we propose a novel approach for indexing the visual content of images based on human perceptual thresholds employed for encoding. In other words, the thresholds that are employed in perceptual coding also serve as an index. These thresholds depend on the overall luminance, frequency/orientation, and the variety of patterns in an image and can serve as indexing features. These features therefore have the potential to retrieve perceptually similar images in response to a query image. Detailed simulations have been carried out using the proposed indexing concept in the DCT compressed domain. Here, the indices have been computed using the DCTune coding technique, which has been shown to provide a superior visual quality in encoding images. Simulation results demonstrate that superior retrieval performance can be achieved for specific classes of images while comparable performance is obtained for other image classes.

Original languageEnglish (US)
Pages (from-to)978-984
Number of pages7
JournalUnknown Journal
Volume4671 II
DOIs
StatePublished - 2002

Fingerprint

Image coding
Signal-To-Noise Ratio
coding
Mean square error
Luminance
Signal to noise ratio
thresholds
discrete cosine transform
luminance
retrieval
signal to noise ratios
simulation

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Condensed Matter Physics

Cite this

Perceptual indexing of visual information. / Fahmy, G. F.; Panchanathan, Sethuraman.

In: Unknown Journal, Vol. 4671 II, 2002, p. 978-984.

Research output: Contribution to journalArticle

Fahmy, G. F. ; Panchanathan, Sethuraman. / Perceptual indexing of visual information. In: Unknown Journal. 2002 ; Vol. 4671 II. pp. 978-984.
@article{96b8e754342b40d4b35397b481941fe5,
title = "Perceptual indexing of visual information",
abstract = "The application of Human perceptual models in image and video coding is motivated by the fact that non-perceptual distortion metrics (mean square error) do not correlate well with the perceived quality at lower bit-rates despite their acceptable signal to noise ratio. In this paper, we propose a novel approach for indexing the visual content of images based on human perceptual thresholds employed for encoding. In other words, the thresholds that are employed in perceptual coding also serve as an index. These thresholds depend on the overall luminance, frequency/orientation, and the variety of patterns in an image and can serve as indexing features. These features therefore have the potential to retrieve perceptually similar images in response to a query image. Detailed simulations have been carried out using the proposed indexing concept in the DCT compressed domain. Here, the indices have been computed using the DCTune coding technique, which has been shown to provide a superior visual quality in encoding images. Simulation results demonstrate that superior retrieval performance can be achieved for specific classes of images while comparable performance is obtained for other image classes.",
author = "Fahmy, {G. F.} and Sethuraman Panchanathan",
year = "2002",
doi = "10.1117/12.453020",
language = "English (US)",
volume = "4671 II",
pages = "978--984",
journal = "Scanning Electron Microscopy",
issn = "0586-5581",
publisher = "Scanning Microscopy International",

}

TY - JOUR

T1 - Perceptual indexing of visual information

AU - Fahmy, G. F.

AU - Panchanathan, Sethuraman

PY - 2002

Y1 - 2002

N2 - The application of Human perceptual models in image and video coding is motivated by the fact that non-perceptual distortion metrics (mean square error) do not correlate well with the perceived quality at lower bit-rates despite their acceptable signal to noise ratio. In this paper, we propose a novel approach for indexing the visual content of images based on human perceptual thresholds employed for encoding. In other words, the thresholds that are employed in perceptual coding also serve as an index. These thresholds depend on the overall luminance, frequency/orientation, and the variety of patterns in an image and can serve as indexing features. These features therefore have the potential to retrieve perceptually similar images in response to a query image. Detailed simulations have been carried out using the proposed indexing concept in the DCT compressed domain. Here, the indices have been computed using the DCTune coding technique, which has been shown to provide a superior visual quality in encoding images. Simulation results demonstrate that superior retrieval performance can be achieved for specific classes of images while comparable performance is obtained for other image classes.

AB - The application of Human perceptual models in image and video coding is motivated by the fact that non-perceptual distortion metrics (mean square error) do not correlate well with the perceived quality at lower bit-rates despite their acceptable signal to noise ratio. In this paper, we propose a novel approach for indexing the visual content of images based on human perceptual thresholds employed for encoding. In other words, the thresholds that are employed in perceptual coding also serve as an index. These thresholds depend on the overall luminance, frequency/orientation, and the variety of patterns in an image and can serve as indexing features. These features therefore have the potential to retrieve perceptually similar images in response to a query image. Detailed simulations have been carried out using the proposed indexing concept in the DCT compressed domain. Here, the indices have been computed using the DCTune coding technique, which has been shown to provide a superior visual quality in encoding images. Simulation results demonstrate that superior retrieval performance can be achieved for specific classes of images while comparable performance is obtained for other image classes.

UR - http://www.scopus.com/inward/record.url?scp=0036029272&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0036029272&partnerID=8YFLogxK

U2 - 10.1117/12.453020

DO - 10.1117/12.453020

M3 - Article

AN - SCOPUS:0036029272

VL - 4671 II

SP - 978

EP - 984

JO - Scanning Electron Microscopy

JF - Scanning Electron Microscopy

SN - 0586-5581

ER -