基于多结构元素灰度形态学的红外背景估计算法

背景估计技术是红外弱小目标检测中的关键技术.传统的形态学算法采用单一结构元素对红外图像进行处理,当背景存在强起伏时,残差图像存在大量的背景泄漏,导致大量虚假目标点的出现.该文在分析红外图像统计特征的基础上,提出了一种基于多结构元素灰度形态学的红外背景估计算法.采用实际红外图像进行了仿真实验,并与传统形态学算法的性能进行比较,结果表明该算法能够尽量保留图像细节,减小图像细节对背景估计的影响,大幅提升了图像的信噪比,提高目标单帧检测概率.算法易于实时并行处理,便于硬件实现.

Algorithm of IR background estimation based on grayscale morphologywith multiple structuring elements

Background estimation is a key technique of infrared(IR) dim point targets detection.Traditional morphology-based algorithms processed the image using a fixed structuring element.However,when the background is heavy cluttered, the residual image contains much background leak,which leads to false point targets.On the basis of the characteristics of the IR images,an algorithm of background estimation based on grayscale morphology with multiple structuring elements is presented.Through the simulations with the real IR images,performance comparisons with the traditional morphology-based algorithms show that the proposed algorithm can preserve the detail image to the greatest extent,reduce the influence on the background estimation,improve the signal-to-noise ratio(SNR) greatly and the detection probability in single frame.The algorithm is easy to use for real-time processing in parallel and realized by hardware.