Research And Implementation Of Helmet-following System Based On Binocular Vision

The helmet-following system has a core algorithm that is similar to the Helmet mounted sight currently used on military fighter jets.The difference is that the application scenario of the helmet-following system is not limited to fighter jets.It is applicable to tanks,armored vehicles,military vehicles and many other applications.With the development of computer technology,especially machine vision and image processing technology,the follow-up helmet system has undergone hardware and algorithm updating.In this paper,the binocular vision measurement method is adopted.Firstly,the currently used camera calibration algorithm is analyzed,and the basic principle of the Zhang Zhengyou checkerboard calibration method is combined.The calibration of the camera is achieved through the "two-step method" and the resulting parameter matrix is calibrated.The use of back-generation method verified the accuracy of the calibration results.In the feature extraction process,in order to increase the accuracy of the target extraction,the feature points set in this paper contain the feature flags of color and shape.At the time of feature extraction,all possible color mark regions are first extracted by color information,and then the shape of the extracted color mark regions is determined.And the accuracy of feature extraction was verified through experiments.In the three-dimensional reconstruction step,this paper first calculates the camera matrix,and then uses the method of three-dimensional point cloud computing to obtain the three-dimensional spatial coordinates of the center of gravity of the target feature region.Finally,using Euler angle calculation formula,the head movement angle is calculated by changing the three-dimensional coordinates of the feature points before and after the rotation.Further verification and error analysis of the corners obtained from the experiment.The above method is adopted to simulate the measurement of the follow-up helmet system in a laboratory environment and achieved the predetermined goal.It provides a simple and feasible measuring method for helmet-following system.Due to limitations of experimental hardware and algorithms,the helmet-following system ultimately produced in this paper is only a prototype.The system is relatively rough in all aspects.There are many tasks that need to be completed in future research work.