• Journal of Infrared and Millimeter Waves
  • Vol. 21, Issue 5, 327 (2002)
[in Chinese]1, [in Chinese]1, and [in Chinese]2、3、4、5
Author Affiliations
  • 1[in Chinese]
  • 2Image Computing Group
  • 3Department of Computer Science
  • 4City University of Hong Kong
  • 5Hong Kong
  • show less
    DOI: Cite this Article
    [in Chinese], [in Chinese], [in Chinese]. ASimilarity Measurement for Affine Invariance Retrieval Based on Synergetic Neural Nets (SNN)[J]. Journal of Infrared and Millimeter Waves, 2002, 21(5): 327 Copy Citation Text show less

    Abstract

    A new concept of control parameter, order vector, based on SNN was proposed. More specifically, its associate properties for image retrieval was studied. Based on the properties, an efficient element based affine invariance similarity measurement was proposed for retrieval of trademark images. The experiments indicate that the retrieval method is good in against noisy and occluded images, and the method has affine invariance in translated, rotated and scaled images.
    [in Chinese], [in Chinese], [in Chinese]. ASimilarity Measurement for Affine Invariance Retrieval Based on Synergetic Neural Nets (SNN)[J]. Journal of Infrared and Millimeter Waves, 2002, 21(5): 327
    Download Citation