基于粒子滤波的红外弱小多运动目标检测

Dim Moving Multi-Target Detection Based on Particle Filter

针对低信噪比条件下红外图像序列中的多个弱小运动目标检测问题,提出了一种基于粒子滤波的多目标检测前跟踪方法。首先分析了红外图像的状态及观测模型,然后将弱小运动目标的检测转化为贝叶斯滤波框架下的检测前跟踪,并利用多个粒子滤波器予以实现。为了评估该算法的有效性,采用人工合成的红外图像序列对其进行测试,结果说明该算法可同时处理低信噪比条件下的多个弱小运动目标,检测和跟踪性能良好。

A multi - target track - before - detect algorithm based on particle filters is proposed for solving the problem of detecting dim moving targets in low SNR environment. The system dynamic model and meas- urement model of the infrared image are presented firstly. Then, a multi - target track - before - detect so- lution under Bayesian filtering framework is derived, and its actual implementation based on multi particle filters is discussed in detail. To test the effectiveness of the detection algorithm, the experimental results with synthetic IR image sequence show that the proposed algorithm can successfully detect and track multi dim moving targets with low SNR, while the performance of detection and tracking is satisfactory.