Retrieval of Satellite Cloud Imagery Based on Subjective Similarity

The goal of this paper is to present an image database method which allows users to retrieve satellite cloud imagery based on subjective similarity of images. First, we propose a technique for shape representation, a region-based deformable model that belongs to the class of shape decomposition methods. The next step is to convert the decomposition result into a mathematical model using a hierarchical attributed relational graph, where a similarity measure is defined as the graph matching cost of two graphs with several weighting factors each indicating its importance in similarity. The most important contribution of this paper is that we formulate a subjective similarity measure with a set of optimized weighting factors so that it maximally reflects personal subjective impression of satellite cloud imagery. Finally we evaluated the accuracy of the result by comparing the retrieved results to subjective similarity.

文献情報

Asanobu KITAMOTO, Mikio TAKAGI, "Retrieval of Satellite Cloud Imagery Based on Subjective Similarity", Proceedings of the 9th Scandinavian Conference on Image Analysis (SCIA'95), pp. 449-456, 1995年6月 (in English)

BibTeX フォーマット

      @InProceedings{ kt:scia95,
      author = {Asanobu KITAMOTO and Mikio TAKAGI},
      title = {Retrieval of Satellite Cloud Imagery Based on Subjective Similarity},
      booktitle = {Proceedings of the 9th Scandinavian Conference on Image Analysis (SCIA'95)},
      pages = {449-456},
      year = 1995,
      month = 6,
      doi = {},
      note = { (in English)},
      }
    

関連資料・関連ウェブサイト

サイト内関連ページ

| リンク 1 | リンク 2 | リンク 3 |