红外图像中快速运动目标的检测与跟踪方法

红外热成像图像具有分辨率较低,细节模糊,对于快速运动目标适应性较差的特点。本文提出了一种结合目标检测算法,目标跟踪算法的红外图像中快速运动目标的检测与跟踪方法。该方法根据红外图像特点,使用ViBE算法检测运动目标,检测出图像中显著运动目标后,触发跟踪器,使用fDSST目标跟踪算法对显著运动目标进行跟踪。测试结果表明,该方法对于快速运动的红外图像目标能够高效检测、快速跟踪。检测与跟踪效果相对传统方法具有检测率更高、鲁棒性更好、实时性更强的优势,对于红外图像中目标检测与跟踪具有很强应用价值。

Fast-moving Target Detection and Tracking Method in Infrared Image

An infrared thermal image exhibits characteristics of low resolution,fuzzy details,and poor adaptability to fast-moving targets.In this paper,a fast-moving target detection and tracking method for an infrared image is proposed,which combines infrared thermal image with machine vision.According to the characteristics of an infrared image,the ViBE algorithm is used to detect moving objects.After detecting the salient moving objects in the image,the tracker is triggered and the fast discriminative scale space tracking algorithm for target tracking is used to track the salient moving objects.The test results demonstrate that this method can detect and track the fast-moving infrared image target efficiently,and its detection and tracking efficiency is higher than in the traditional method.In addition,it is robust and has real-time advantages.Therefore,this method has a strong application value for infrared image target detection and tracking.