摘要
光电跟踪取证系统对运动目标进行跟踪拍摄时需要进行变焦和调焦,目标姿态和位置改变而产生调焦窗口偏离是导致传统自动调焦方法失效的主要原因。针对该问题,提出了一种针对跟踪目标动态建立调焦窗口的方法。该方法利用梯度能量评价函数提取目标特征,通过相关性处理和边界扩展建立调焦窗口,能克服调焦窗口偏离的缺陷。实验结果证明,对704×576像元不同离焦状态的图像都可以在20 ms内根据目标的位置和大小,构建准确有效的调焦窗口,满足光电跟踪取证系统对运动目标进行实时自动调焦的特殊要求。
Zoom and focus are required when photoelectric tracking system is used to track moving targets. The deviation of focusing window caused by the change of target posture and position is the main reason that leads to the failure of automatical focusing. In light of this, a method to establish dynamic focusing window for target is proposed. We used gradient energy evaluation function to extract target features and then estab- lish focusing window by correlation treatment method and boundary expansion, which can overcome the de- fects of original method. The experiment results show that the new method can establish effective focusing window for 704×576 pixels images during 20 ms. The results demonstrate that this method can meet the need of real-time operation.
出处
《红外技术》
CSCD
北大核心
2016年第3期197-202,共6页
Infrared Technology
关键词
光电跟踪
动态调焦
梯度能量评价函数
调焦窗口构建
特征提取
photoelectric tracking, dynamic focusing, gradient energy evaluation function, focusing window establishment, feature extraction