基于分形理论的红外图像港岸背景抑制方法

Background Suppression in Littoral EnvironmentBased on Fractal Theory in IR Image

摘要:

为了实现港岸背景的抑制,提出了一种基于分形理论和小波变换的背景抑制方法。首先,在高斯去噪的基础上,使用Otsu方法对图像进行粗分割；然后,对粗分割得到的背景图像进行分形降维处理,选定序列图像的比较区域；最后,利用小波变换的奇异值检测特性,在比较区域寻找序列图像比较的基准点,从而确定序列图像的移动范围,进而实现序列图像背景抑制。通过对实际港岸背景红外图像的实验,验证了所提方法的可行性和有效性。

Abstract:

A background suppression method based on fractal theory and wavelet transform was proposed in order to suppress the littoral background. Firstly, IR image was segmented coarsely using the Otsu method on the basis of Gauss method. Secondly, the segmented background image was processed by a fractal method, and the compared region of the sequential images was selected. Finally, the compared datum mark of the sequential images was found by the wavelet transform method, and the movement between images was computed, and the background was suppressed by the movement between sequential images. The experimental result showed that the proposed method was feasible and effective.