基于高斯-拉普拉斯滤波的增强局部对比度红外小目标检测算法

A LOG Filter Based Enhanced Local Contrast Algorithm to Detect Infrared Small Targets

针对单帧图像红外小目标检测算法在低空复杂背景下虚警率高的问题,提出一种基于高斯-拉普拉斯(LOG)滤波的增强局部对比度算法来检测低空复杂背景下的红外小目标。通过LOG滤波运算快速提取候选目标像素,同时通过像素灰度指数运算增强目标。根据局部区域目标与背景的灰度特征计算目标显著图,通过自适应阈值分割提取目标。针对不同的低空复杂场景,构建了测试数据集,从信噪比增益、背景抑制因子、检测率、虚警率及算法计算效率方面将所提算法与Top-Hat算法、Max-median算法、RLCM算法、IPI算法及MPCM算法进行对比分析。实验结果表明,在不同场景中,所提算法相较于对比算法不仅具有较高的信噪比增益和背景抑制因子,而且具有较高的检测率、较低的虚警率和较高计算效率,验证了该算法的有效性和鲁棒性。

To address the problem of high false alarm rate of single-frame infrared small-target detection algorithm in low-altitude and complex backgrounds,a Laplacian of Gaussian(LOG)filter-based enhanced local contrast algorithm is proposed.First,the candidate target pixels are extracted quickly by LOG filtering,while the target is enhanced using pixel grayscale indexing.Then,the target saliency map is calculated based on the grayscale features of the target and the background in the local area.Finally,the target is extracted by adaptive threshold segmentation.Test datasets are constructed for different low-altitude complex scenarios,and the proposed algorithm is compared with the Top-Hat algorithm,Max-median algorithm,RLCM algorithm,IPI algorithm,and MPCM algorithm in terms of signal-to-noise ratio gain,background rejection factor,detection rate,false alarm rate,and computational efficiency.Results show that in different scenarios,the newly proposed algorithm not only has higher signal-to-noise ratio gain and background rejection factor,but also has higher detection rate,lower false alarm rate and higher computational efficiency than other algorithms,demonstrating the method’s effectiveness and robustness.