应用分数阶傅里叶变换的形状描述方法研究  

Study on shape description under fractional fourier transform

作　　者：张婉婷[1] 王斌[1,2] 

机构地区：[1]南京财经大学信息工程学院,南京210046 [2]南京财经大学电子商务省级重点实验室,南京210046

出　　处：《计算机工程与应用》2011年第27期185-188,198,共5页Computer Engineering and Applications

基　　金：江苏省自然科学基金(No.BK2009348);江苏省高校自然科学研究计划(No.08KJB510002)~~

摘　　要：傅里叶描述子是一种经典的形状描述方法。作为傅里叶变换的推广形式,分数阶傅里叶变换在数字信号处理工程领域已有相当广泛的应用,但在形状分析领域还很少有研究工作的报道。首次研究了基于分数阶傅里叶变换的形状描述方法,比较了不同阶数下的分数阶傅里叶描述子在图像检索中的性能。通过在MPEG-7的标准图像测试集的图像检索实验,得出:阶数ρ为0.1时,分数阶傅里叶描述子的检索效果最差,随ρ=0.1的增长,检索性能总体呈上升趋势,当ρ=0.5变化到1.0时,检索性能最高。同时,与Zernike矩进行比较:当阶数为0.1时,分数阶傅里叶描述子的检索性能较差;而阶数为0.5、1.0时分数阶傅里叶描述子的检索性能均较好。Fourier descriptor is an essential way for shape description.As a generation of Fourier transform,the fractional Fourier transform has been widely used in the field of digital signal processing projects,but there has been few research reports of applying it to shape analysis.In this paper,it＇s the first time to study the method of shape description under fractional Fourier transform,comparing the image retrieval performances of fractional Fourier descriptors with different fractional orders.Retrieval tests are conducted on MPEG-7 contour shape database CE-1.The retrieval results of fractional Fourier descriptors with different orders to different type of shape distortions are fully analyzed.When the order is 0.1,the fractional Fourier descriptor performs worst.With the growth of the order,retrieval results of the fractional Fourier descriptors become better and better.When the order changes to 1.0,the fractional Fourier descriptor performs best.Meanwhile,comparing with the Zernike moment：When the order is 0.1,the Zernike moment gets better result.When the order is 0.1 and 1.0,the fractional Fourier descriptors get better results.

关 键 词：图像检索 形状描述 傅里叶描述子 分数阶傅里叶变换 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]