航空仪表图像自动监测系统研究

Study of Auto-monitoring System for Aviation Instrument Image

针对航空虚拟仪表系统的研发、生产过程中测试操作繁琐、重复劳动强度大等问题,设计了座舱仪表图像自动监测识别系统,系统包括图像采集、变化检测、仪表图检测和识别、检测结果归档4个部分。基于航空虚拟仪表图的结构特点,提出了一种分层模板图快速匹配的仪表图像检测算法,采用二值化的模板图与搜索区域的二值化图进行分级匹配方法实现快速检测。测试结果表明,提出的方法均能快速准确检测出屏显上出现的单仪表图和重复的多仪表图,系统整体上能达到准实时处理高清屏幕图的监测。

In order to reduce the manual operation and improve the efficiency during the design,manufacture and testing of virtual aviation instrument systems,an auto-monitoring and recognition system is designed for virtual instrument of aircraft cockpit,which consist of four modules,including image acquisition,frame change detection,location of meter diagram and storage of structured data.Furthermore,a hierarchical binary template fast matching algorithm is proposed to detect and locate the meter image and recognize the reading.Finally,the auto-monitoring system is implemented to verify the proposed algorithms.Experiment results show that the algorithms can accurately detect the single and multiple meter diagrams in screen image,and the system has a strong capability for quasi-real time processing HD images.