Improved scheme for object searching using moment invariants

K. L. Lau, W. C. Siu, Ngai Fong Law

Research output: Chapter in book / Conference proceedingConference article published in proceeding or bookAcademic researchpeer-review


For multimedia retrieval application, shape is always a conspicuous element of an object. Moment-based approaches are widely used for shape description due to its translation, scaling and rotation invariance. Moment invariants are defined in the continuous domain. However, when considering the digital images in practice, quantization errors are introduced. Thus, the moment invariants calculated might not be truly invariant. This paper presents an analysis of quantization effects on four moment-based approaches of both regular and irregular objects. From the analysis, the scaling errors for all approaches are large when the scaling factor is less than 0.5. Moreover, the rotational errors are big for the objects rotated other than the multiples of 90. Our experimental results show that Dudani moment invariants suffer the largest error for overall sensitivity, while Affine moment invariants show the smallest. Furthermore, this error analysis has also been applied successfully to object searching applications using a threshold selection scheme.
Original languageEnglish
Title of host publicationAdvances in Multimedia Information Processing - PCM 2002 - 3rd IEEE Pacific Rim Conference on Multimedia, Proceedings
PublisherSpringer Verlag
Number of pages8
ISBN (Print)3540002626, 9783540002628
Publication statusPublished - 1 Jan 2002
Event3rd IEEE Pacific Rim Conference on Multimedia, PCM 2002 - Hsinchu, Taiwan
Duration: 16 Dec 200218 Dec 2002

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349


Conference3rd IEEE Pacific Rim Conference on Multimedia, PCM 2002

ASJC Scopus subject areas

  • General Computer Science
  • Theoretical Computer Science


Dive into the research topics of 'Improved scheme for object searching using moment invariants'. Together they form a unique fingerprint.

Cite this