一种二维极坐标傅里叶描述子在图像检索中的应用

2-dimensional Polar Fourier Descriptor Applying in Image Retrieval

在图像检索中,形状描述是图像内容描述的关键部分。目前已经存在的大多数形状描述符均是基于某一特定应用的,不具有通用性。文章提出了一种基于二维极坐标傅里叶变换的傅里叶描述子对目标形状的描述方法,分析并实验验证了该方法的各个性质,并讨论了它的实际计算问题。实验结果表明,该描述符通用性强,鲁棒性高。

Shape description is the key part of image content description for image retrieval.Most of the existing shape descriptors are usually either application dependent or non-robust.In this paper,a method for objects shape description-Fourier descriptor based on two-dimensional polar Fourier transform is proposed and its properties are explained and verified by experiments.Some problems concerned with practical computation are also discussed.The results show that the acquired shape descriptor is application independent and robust.