Fast Polar Cosine Transform for image description

Zhuo Yang*, Sei Ichiro Kamata

*この研究の対応する著者

研究成果: Conference contribution

3 被引用数 (Scopus)

抄録

Polar Cosine Transform (PCT) is one of the Polar Harmonic Transforms that those kernels are basic waves and harmonic in nature. They are proposed to represent invariant patterns for two dimensional image description and are demonstrated to show superiorities comparing with other methods on extracting rotation invariant patterns for images. However in order to increase the computation speed, fast algorithm for PCT is proposed for real world applications like limited computing environments, large image databases and realtime systems. Based on our previous work, this paper novelly employs relative prime number theory to develop Fast Polar Cosine Transform (FPCT). The proposed FPCT is averagely over 11 ∼ 12.5 times faster than PCT that significantly boost computation process. The experimental results are given to illustrate the effectiveness of the proposed method.

本文言語English
ホスト出版物のタイトルProceedings of the 12th IAPR Conference on Machine Vision Applications, MVA 2011
ページ320-323
ページ数4
出版ステータスPublished - 2011 12月 1
イベント12th IAPR Conference on Machine Vision Applications, MVA 2011 - Nara, Japan
継続期間: 2011 6月 132011 6月 15

出版物シリーズ

名前Proceedings of the 12th IAPR Conference on Machine Vision Applications, MVA 2011

Conference

Conference12th IAPR Conference on Machine Vision Applications, MVA 2011
国/地域Japan
CityNara
Period11/6/1311/6/15

ASJC Scopus subject areas

  • コンピュータ ビジョンおよびパターン認識

フィンガープリント

「Fast Polar Cosine Transform for image description」の研究トピックを掘り下げます。これらがまとまってユニークなフィンガープリントを構成します。

引用スタイル