Image Retrieval Based on Wavelet Features

Details Image Retrieval Based on Wavelet Features Image Retrieval Based on Wavelet Features

Apr 2006

adsabs.harvard.edu/cgi-bin/nph-data_query?bibcode=2006vopc.conf..221m&link_type=abstract

Virtual Observatory: Plate Content Digitization, Archive Mining and Image Sequence Processing, iAstro workshop, Sofia, Bulgaria,

Astronomy and Astrophysics

Astronomy

Scientific paper

A dominant (additive, stationary) Gaussian noise component in image data will ensure that wavelet coefficients are of Gaussian distribution, and in such a case Shannon entropy quantifies the wavelet transformed data well. But we find that both Gaussian and long tailed distributions may well hold in practice for wavelet coefficients. We investigate entropy-related features based on different wavelet transforms and the newly developed curvelet transform. Using a materials grading case study, we find that second, third, fourth order moments allow 100% successful test set discrimination.

Affiliated with

Also associated with

No associations

Rating

Image Retrieval Based on Wavelet Features does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.

If you have personal experience with Image Retrieval Based on Wavelet Features, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Image Retrieval Based on Wavelet Features will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFWR-SCP-O-1185701