基于图像复原技术的红外小目标检测方法

A Method for Small Infrared Target Detection Based on the Technique of Image Restoration

提出了一种准确、快速的红外小目标单帧检测方法。该方法首先利用图像复原技术对红外小目标图像进行平滑预处理,达到去噪、提高图像对比度的效果;接着将滤波后的图像与原始图像做差分对消处理,抑制背景杂波;然后采用梯度法对残差图像进行锐化处理,凸显小目标;最后设置两个灰度级做阈值提取,从而检测到红外小目标。仿真实验结果表明本算法能快速、准确的检测出目标点,鲁棒性较好。

The paper presents an accurate and fast method for small infrared target detection in a single frame. The method firstly applies the technique of image restoration in the pre-processing of image smoothing, namely setting up a degradation model upon which proceeding with recovering filter to restrain the noise and enhance the image＇s contrast, then makes use of the background subtraction method to process the original images and filtered ones. Based upon that, a gradient sharpening method is applied to strengthen the target information and restrain the background information of the residual image. Finally, a binary technique is utilized to highlight the target. The simulation results show such method can find out the small target exactly and quickly besides good robustness.