| Nowadays,high-speed railway has become a business card of China in the world.The rapid development of the railway requires more technology support to maintain the world’s leading level.In the traditional railway training,the principle and operation of railway equipment are taught through lectures.Currently,the VR technology has been applied to railway training,but this pure virtual training environment can not simulate all the situations in the operation,and the effectiveness of the training would be reduced.Therefore,the application of AR technology in railway assistance training has important impetus and practical significance for improving the training efficiency of railway staff.In this way,the theoretical knowledge mastered by the staff can be maximized in the practical application.Based on AR code as the research object,a series of studies on the AR code identification and the application of AR codes in railway training are carried out,the purpose is to improve the stability of the railway assistance training system.The main work is as follows:Firstly,in order to improve the recognition rate of the AR code,the AR code size with the highest identification rate is determined by comparative experiment with the illumination and distance as conditions,and the maximum identification distance for this size under various conditions is also tested.By adding a positioning flag to the classic AR code,the correct reading direction of the AR code is determined.Aiming at the problem that is difficult to identify the AR code due to the distortion of the AR code pasting on the cylinder,a correction algorithm is proposed to correct the AR code in plane and three-dimensional space.By analyzing the non-linear relationship between the number of rows and width of the each rows,the segmentation of the AR code implemented,and the actual application range of the AR code is expanded.Secondly,by analyzing the process of railway assistance training,taking the IBP as an example,the assistance training task description model and operation guidance model are constructed by introducing the Petri net modeling technology to the railway assistance training task network,and the synchronization of the training tasks content description and the multimedia information display during training is realized.According to the practical application of the railway assistance training system,an optical see-through head-mounted display(OST-HMD)is developed,the display is calibrated in order to eliminate the deviation between the human eye and the video camera on the OST-HMD.Finally,based on the relevant theoretical research,the assistance training system software is designed with open source library OpenFrameworks as the framework.The experiment is carried out using the OST-HMD in an actual training scenario attaching the AR code.The results verify that the improved AR code has a higher identification rate.The AR evaluation technology is used to evaluate this assistance training system and the results verify the superiority of this system in the field of railway training. |