毫米波人体安检图像自适应滤波方法

为解决毫米波人体安检图像背景区域的大量毛刺、混叠等噪声对隐匿物体识别造成的干扰问题,提出一种自适应的二次模板匹配滤波方法。首先将原始图像进行水平集二值分割,然后将分割结果作为模板对原始图像进行背景滤除,最后将滤除背景的结果作为先验图像对原始图像进行改进的双边滤波。基于毫米波成像系统的实际图像进行对比实验,验证了此算法相比于传统滤波方法的改善,并证明此算法可以针对性地滤除毫米波安检图像的背景噪声,保留人体区域的图像细节,有利于毫米波安检图像中的隐匿物品识别与定位。

An Adaptive Filtering Method for Millimeter Wave Human Security Inspection Image

To decrease the influence of burrs, aliasing and other noises in the millimeter wave human security image background area on the identification in security inspection, an adaptive secondary template matching filtering method is proposed. First, the original figure is segmented by level set binary method. Then the segment result is used as a template to filter the original image background. Finally, the result is used as a priori image to do improving bilateral filtering of the original image. The experiments on the actual images of millimeter wave imaging system verify the improvement of this algorithm compared with the traditional filtering method, and prove that this algorithm can filter the background noises of the human body security image, retain the image details of the human body area, and is conducive to the object recognition and location in the millimeter wave security image.