基于运动模糊复原技术的红外弱小目标检测方法

发布时间：2018-05-04 02:38

本文选题：图像处理 + 红外弱小目标　；参考：《光学学报》2017年06期

【摘要】：在红外对空探测系统中,由于探测器时刻处于运动状态使得目标图像产生剧烈的运动模糊,给红外小目标检测造成困难。为了解决运动模糊条件下红外小目标检测的问题,提出将运动模糊复原技术和图像增强技术引入红外探测系统。先将探测器采集到的原始图像经过维纳滤波,对运动模糊进行处理并抑制噪声干扰,再利用梯度法对处理后的图像做锐化处理,增强目标边缘。实验验证和仿真分析结果都表明,该方法运动模糊复原效果明显,并在一定程度上抑制了噪声,提高了目标对比度,使目标在背景中更加凸显,并且能够显著提高目标图像质量。引入的评价参数峰值信噪比和均方差表现良好,该方法可以增强探测系统的使用性能。
[Abstract]:In the infrared space detection system, because the detector is in the state of motion at all times, the image of the target produces sharp motion blur, which makes it difficult to detect the small infrared target. In order to solve the problem of small infrared target detection under moving blur, the motion blur restoration technique and image enhancement technique are introduced into infrared detection system. Firstly, the original image collected by the detector is processed by Wiener filter, then the motion blur is processed and the noise interference is suppressed, then the processed image is sharpened by gradient method to enhance the edge of the target. The experimental results and simulation results show that the method has obvious effect on motion blur restoration, and to a certain extent, it suppresses the noise, improves the contrast of the target, and makes the target more prominent in the background. And it can improve the quality of the target image. The peak signal-to-noise ratio (PSNR) and mean square error (RMS) of the evaluation parameters are good, and this method can enhance the performance of the detection system.
【作者单位】：电子工程学院脉冲功率技术国家重点实验室;
【基金】：国家自然科学基金(61307025) 安徽省自然科学基金(1308085QF122)
【分类号】：TN215;TP391.41

【参考文献】