An efficient region-based image representation using Legendre color distribution moments

  • Authors:
  • ZhiYong Zeng;XueMei Ding;ShiGang Liu

  • Affiliations:
  • Fujian Normal University, Minhou Section, Fuzhou, China;Fujian Normal University, Minhou Section, Fuzhou, China;Xi'an Jiaotong University, Xi'an, China

  • Venue:
  • Proceedings of the 2nd International Conference on Interaction Sciences: Information Technology, Culture and Human
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

By exploiting region Legendre color distribution moments, a region-based image representation in terms of storage and complexity is proposed. The representation consists of three steps: First, an image is segmented regions by a classical algorithm. Second, a compact, fixed-number and computation effective representaton is designed for the color contents of each region of an image, which takes not only the local color feature of a region into consideration, but also the correlation of the color distribution of the region. Third, we use Legendre color distribution moments as feature vector of the regions to match images. With the robustness to rotation and translation, the representation avoids shortcoming of losing the correlation of the color distribution and the spatial color distribution information in color histogram experimental results on a database of 1000 general-purposed images demonstrate the efficiency and effectiveness of the proposed representation for image retrieval.