| 摘要:目标交接是图像传感系统中的关键技术，针对基于特征融合目标交接算法中特征提取的不准确，提出基于决策级融合的多摄像机目标交接算法。将目标的每个特征作为一个证据，给出每个证据基本置信指派函数的构造方式。利用Dempster组合规则得到融合多个证据后的基本置信指派。利用融合决策规则来判断当前目标是否和辨识框架中的某个目标为同一个目标，完成目标交接，并给出目标在系统中唯一标识。多种场景仿真实验表明该算法可以准确地完成多摄像机目标交接。 |
| 关键词: 信息融合 目标交接 Dempster组合规则 决策级融合 基本置信指派 决策规则 |
|
| Image Perception System Object Handoff Using Decision-Level Fusion |
|
|
| Abstract:Object handoff is the key technique of the image perception system (IPS). Since the feature extraction is not accurate, a novel object handoff algorithm based on decision-level fusion is presented. This algorithm can be described as: firstly, every feature of the object is defined as an evidence, of which basic belief assignment (BBA) is computed; secondly, the Dempster combination rule is utilized to fuse multiple evidence to get the final BBA; lastly, decision rule is utilized to determine whether the current object is the same one with that in the identification framework, and the current object also acquires an unique identifier. After the above-mentioned procedures, the object handoff in the IPS is accomplished. The simulation results show that this object handoff algorithm can achieve better performance. |
| Key words: information fusion object handoff Dempster combination rule decision-level fusion basic belief assignment decision rule |