基于高通滤波和顺序滤波的小目标检测

Small target detection based on high-pass filtering and order filtering

针对红外图像中弱小目标的检测问题,比较了几种高通模板的效果与性能,提出了一种基于高通和顺序滤波相结合的小目标检测方法。该方法是在采用高通滤波来抑制背景的基础上,利用噪声图像直方图呈高斯分布的特性,用门限判决对图像进行预检测。然后用顺序滤波去虚警点从而得到目标点。实验结果表明,在信噪比SNR>2时该方法能有效地检测出图像中的小目标。

Detection of weak and small targets in an infrared image is one of difficulties among image target detection and identification. This paper compares the effect and performance of several high-pass filters. A small target detection method based on high-pass filtering and order filtering is presented to solve the small target detection with low SNR in an infrared image. First high-pass filtering suppresses the background of the original image. Then some candidate dots are got by a threshold separation because the histgram of the noise image is distributed in Gaussian distribution. Finally the true target is filtraded by using an order filtering. The experiments show that this method can easily and effectively extract the small target in an infrared image with its SNR>2.\;