适配模板滤波在医学图像中的应用

Adaptive Template Filter and Its Applications for Medical Images

适配模板滤波是一种既能提高信噪比又可保留边缘信息的滤波技术 ,当今 ,虽然已有多种滤波技术可以较大幅度的提高信噪比 ,但是同时也丢掉了许多有用的信息。而在医学图像中 ,有用信息的保留往往比简单的提高信噪比要重要的多 ,与传统滤波器不同的是 ,本文介绍的滤波技术的基本模板滤波系数可以根据输入图像的局部特性动态自适应调节。适配模板滤波技术根据图像上的每一个象素点及其周围邻近的八个临点的情况自动选出最佳匹配模板加以处理 ,识别及判断 ,以决定是给予保留还是滤除。这样可以最大限度地保留边缘的信息点 ,同时也可以尽可能大地抑制随机噪声。通过与现存的滤波器比较 ,无论在磁共振图像和仿真图像上 ,采用此种滤波器都可得到更清晰的图像边缘和更高的信噪比。

In this paper, an adaptive template filtering method is described which can be used to improve signal-to-noise ratio(SNR) at the same time retain important edge information of medical images. To date, various filtering approaches are reported. Most of them enhance SNR in certain degree, but along with loss of a lot of useful information. Unlike conventional filtering, where the template shape and coefficients are fixed, multiple adaptive templates are defined in this paper. For proposed algorithm, for each pixel, an optimal template is selected automatically depending on its 8 neighboring pixels. Simulation and MRI image tests, both 2D and 3D, showed that the adaptive template filter provides better SNR and sharper edges.