基于双目视觉的奖状训练器CJ1飞行操纵数据识别技术研究

Research on Recognition Technology of Flight Control Data of CJ1 Trainer Based on Binocular Vision

飞行操纵数据可以用于事故调查、飞行性能检测、飞机故障预判等,同时也是模拟设备开发的重要数据,国外生产的飞机,其操纵数据很难获取且价格昂贵。为此,文章提出了一种基于双目视觉的图像识别法,对国内奖状训练器飞行操纵数据的输入端数据进行获取。由于目标测量点处于黑舱的特殊环境中,其与周围环境在色度、饱和度和亮度上的区别越大,则所带来的检测误差越小,故文章将目标测量点应用红色标签进行标识提取,并将所获得的数据与奖状训练器自带的标注测试数据进行对比,数据误差范围在2%~3%,满足模拟设备鉴定时的操纵数据误差范围,表明该方法的有效性。该方法为获得飞行操纵数据提供了一种新方式,具有一定的参考价值。

Flight control data can be used in accident investigation,flight performance detection,aircraft failure prediction,etc.,and it is also an important data for the development of simulation equipment.For foreign-produced aircraft,its control data is difficult to obtain and expensive.Therefore,this paper proposes an image recognition method based on binocular vision to obtain the input data of the flight control data of the domestic award trainer.Since the target measuring point is in the special environment of the black cabin,the greater the difference in chroma,saturation and brightness between the target measuring point and the sur-rounding environment,the smaller the detection error will be.Therefore,this paper applies the red label to identify and extract the target measuring point,and compares the obtained data with the marked test data provided by the award trainer.The error range of the data is about 2%~3%,which meets the error range of the control data in the simulation equipment identification,which shows the effectiveness of the method.This method provides a new way to obtain flight control data and has certain reference value.