Topological and directional logo layout indexing using Hermitian spectra
Topological and directional logo layout indexing using Hermitian spectra
复制标题
DOI：
--
复制
发表时间：
2008-02
期刊：
影响因子：
--
通讯作者：
Reinier H. van Leuken;Olga Symonova;R. Veltkamp
中科院分区：
文献类型：
--
作者：
Reinier H. van Leuken;Olga Symonova;R. Veltkamp
为了评价两幅图像之间的相似性，除了几何形状相似性之外，形状的布局或形状也是一个重要的特征。特别是，商标图像检索是一个布局相似很重要的应用领域，但在许多情况下被忽视了。本文提出了一种基于图的布局编码方法，该编码同时存储了方向布局信息和拓扑布局信息。埃尔米特矩阵与每个图相关联，并且包含图中存在的所有信息。这些厄米特矩阵的谱用于索引目的。通过遵守构造厄米特矩阵的几个约束，我们可以模拟拉普拉斯矩阵的谱行为，这被证明是在检索环境中被成功表示的。实验表明，与使用拉普拉斯矩阵的谱方法相比，该方法具有更好的表征能力。
To evaluate similarity between two images, the layout or configuration of the shapes is an important feature besides geometrical shape similarity. In particular, trademark image retrieval is an application domain where layout similarity is important, and in many cases overlooked. In this paper, we present a graph-based encoding of layout, in which both directional and topological layout information is stored. A Hermitian matrix is associated to each graph, and contains all the information that is present in the graph. The spectra of these Hermitian matrices are used for indexing purposes. By obeying several constraints on the construction of the Hermitian matrices, we can mimic the spectral behaviour of Laplacian matrices, which are proven to be successful representations in retrieval environments. Experiments show the improved representational power of the proposed approach over spectral methods using Laplacian matrices.