红外成像地面运动目标跟踪方法研究

发布时间：2018-02-27 03:28

本文关键词： 红外图像地面场景自适应模板选择目标跟踪自适应模板更新　出处：《华中科技大学》2016年硕士论文　论文类型：学位论文

【摘要】：红外成像地面场景运动目标的跟踪在军事等领域应用很普及。由于地面场景很复杂、干扰具有很强的随机性、目标先验信息较为缺乏及目标运动很复杂,使得红外图像中地面背景下运动目标的跟踪成为国内外学者的研究热点。本文研究了地面场景下的运动目标的跟踪方法,主要工作如下:首先介绍了红外成像地面运动目标跟踪系统组成结构,分析了红外图像的特性,研究了几种抑制红外图像中噪声的方法。在缺少目标先验信息的情况下,提出了一种基于混合比和大津阈值分割的自适应模板选择方法。其次,在跟踪阶段,本文采用一种特征关联与模板匹配结合的运动目标跟踪方法。根据自适应模板选择阶段获取的目标信息是否与边界相关而采取相应的跟踪方法。当自适应模板选择阶段获取的目标信息与边界无关时,通过自适应加权系数的多特征融合方法对目标进行跟踪。当自适应模板选择阶段获取的目标信息与边界相关时,通过Kalman滤波与模板匹配相组合的方法对目标进行跟踪。另外,对跟踪波门内与目标相似的物体的跟踪方法进行了研究,本文在基于Kalman滤波与模板匹配的方法的基础上实现了对目标及与目标相似的物体的跟踪。然后,本文对跟踪阶段匹配模板的更新方法进行了研究。为了减少跟踪过程中匹配模板的漂移,使匹配模板更好地适应目标的变化,本文采用了一种自适应模板更新策略,保证了跟踪过程的准确稳定。最后,本文对红外成像下地面运动目标跟踪系统组成模块进行了介绍,同时简要介绍了PC端的系统调测软件,以及算法移植所做的部分修改与优化。
[Abstract]:The tracking of moving targets in infrared imaging ground scenes is very popular in military and other fields. Because of the complexity of ground scenes and the strong randomness of interference, the prior information of targets is relatively short and the motion of targets is very complex. It makes the tracking of moving targets in the ground background in infrared images become the research hotspot of scholars at home and abroad. In this paper, the tracking methods of moving targets in ground scenes are studied. The main work is as follows: firstly, the structure of infrared imaging ground moving target tracking system is introduced, the characteristics of infrared image are analyzed, and several methods to suppress noise in infrared image are studied. An adaptive template selection method based on mixing ratio and threshold segmentation is proposed. Secondly, in the tracking stage, In this paper, a moving target tracking method based on feature association and template matching is proposed. According to whether the target information obtained in the adaptive template selection stage is related to the boundary, the corresponding tracking method is adopted when the adaptive template is selected. When the target information obtained in the alternative phase is not bound to the boundary, The multi-feature fusion method of adaptive weighting coefficient is used to track the target. When the target information obtained in the selection stage of the adaptive template is related to the boundary, the target is tracked by the combination of Kalman filter and template matching. In this paper, the tracking method of objects similar to targets in the wave door is studied. Based on the method of Kalman filtering and template matching, the tracking of targets and objects similar to targets is realized in this paper. In this paper, the updating method of matching template in tracking stage is studied. In order to reduce the drift of matching template in the tracking process and make the matching template better adapt to the change of target, an adaptive template updating strategy is adopted in this paper. The tracking process is guaranteed to be accurate and stable. Finally, the module of ground moving target tracking system under infrared imaging is introduced in this paper. At the same time, the system adjusting and measuring software of PC is introduced briefly. And the partial modification and optimization of algorithm migration.
【学位授予单位】：华中科技大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TP391.41

【参考文献】