基于多特征信息融合粒子滤波的红外目标跟踪  

Infrared target tracking fusing particle filtering with multi-feature information

作　　者：孙艳丽[1] 周伟[2] 李建海[1] 

机构地区：[1]海军航空工程学院基础实验部,山东烟台264001 [2]海军航空工程学院信息融合研究所,山东烟台264001

出　　处：《现代电子技术》2017年第13期9-12,共4页Modern Electronics Technique

基　　金：国家自然科学基金(61471382):MIMO阵列雷达海上微弱目标检测技术研究

摘　　要：多特征信息有较好的检测性能和适应性,而粒子滤波则是一种处理目标跟踪模型的非线性和非高斯特点的有效方法,将两者优点结合并针对红外图像特点,提出一种基于多特征信息融合的跟踪算法,该方法按一定的权值系数利用目标颜色和纹理特征构建模型,并融合于粒子滤波框架中。实验表明该跟踪方法能准确地跟踪海上红外运动目标。Since the multi-feature information has perfect detection and adaptation performances,and the particle filtering is an effective method to process the nonlinear and non-Gaussian characteristics of the target tracking model. The advantages of the above items are combined to propose a tracking algorithm based on multi-feature information fusion for the characteristics of the infrared image. The target color and textural feature are adopted in the method to construct the model according to a certain weight coefficient. The model is fused to the frame of particle filtering. The experimental result shows that the tracking algorithm can track the maritime infrared moving target accurately.

关 键 词：粒子滤波 纹理特征 多特征融合 目标跟踪 

分 类 号：TP274[自动化与计算机技术—检测技术与自动化装置]