基于光流估计的红外弱小运动目标检测

Small Infrared Moving Target Detection Based on Optical Flow Estimation

提出了一种对复杂云层背景下红外图像序列中弱小运动目标分割和检测的方法。首先,利用Horn-Schunck算法计算序列图像的光流场,然后利用阈值分割和数学形态学滤波的方法进行目标检测,滤除噪声后提取出背景中的运动目标。实验结果表明,该算法对实时检测在复杂背景的红外图像中运动的弱小目标具有很好的效果。

An efficient approach for segmenting and detecting small moving target in infrared image sequences against complex cloud layer background is given. Horn-Schunck algorithm was used to calculate the optical flow of the image sequences. Then, segmentation of threshold and mathematical morphological filter were employed to detect the moving infrared target, and extract it from the background after filtering the noise. The experiment results showed that this method is effective for detecting the small infrared target under a complex scene in real time.