Han Risheng, Shen Shigen, Yue Guangxue, Ding Hui
Nanhu College, Jiaxing University, Jiaxing 314001, China.
College of Mathematics, Physics and Information Engineering, Jiaxing University, Jiaxing 314001, China.
ScientificWorldJournal. 2014;2014:918453. doi: 10.1155/2014/918453. Epub 2014 Aug 19.
A novel BA complex network model of color space is proposed based on two fundamental rules of BA scale-free network model: growth and preferential attachment. The scale-free characteristic of color space is discovered by analyzing evolving process of template's color distribution. And then the template's BA complex network model can be used to select important color pixels which have much larger effects than other color pixels in matching process. The proposed BA complex network model of color space can be easily integrated into many traditional template matching algorithms, such as SSD based matching and SAD based matching. Experiments show the performance of color template matching results can be improved based on the proposed algorithm. To the best of our knowledge, this is the first study about how to model the color space of images using a proper complex network model and apply the complex network model to template matching.
基于BA无标度网络模型的两个基本规则：增长和优先连接，提出了一种新颖的颜色空间BA复杂网络模型。通过分析模板颜色分布的演化过程，发现了颜色空间的无标度特性。然后，模板的BA复杂网络模型可用于选择在匹配过程中比其他颜色像素具有更大影响的重要颜色像素。所提出的颜色空间BA复杂网络模型可以很容易地集成到许多传统的模板匹配算法中，如基于SSD的匹配和基于SAD的匹配。实验表明，基于所提出的算法可以提高颜色模板匹配结果的性能。据我们所知，这是关于如何使用适当的复杂网络模型对图像的颜色空间进行建模并将复杂网络模型应用于模板匹配的首次研究。
