Research On Projectile Trajectory Testing Technology Based On Binocular Visio

Ballistic coordinates are important parameters for measuring weapon shooting accuracy and evaluating weapon system performance.Accurate measurement of them and completion of ballistic trajectory reconstruction are of great significance for weapon system optimization design and firing schedule preparation.A photogrammetric method based on high-speed imaging and binocular vision positioning technology is proposed,which can overcome complex environmental interference and is suitable for ballistic testing in field scenes.Based on the parallax method,a projectile space coordinate measurement model conforming to the muzzle coordinate system is established,and the main factors affecting the measurement accuracy of the parallax method are analyzed from the measurement system hardware and software.Combined with the measurement model,the calculation process of binocular vision system calibration under the traditional Zhang Zhengyou calibration method is researched and deduced.On this basis,the relevant theories and methods of binocular stereo correction are introduced.Combining with the test task of the external ballistic trajectory of the bullet,the overall plan of the test system is proposed.In view of the difficulty in calibration of the long-baseline binocular vision system under the traditional stereo calibration method,the binocular system calibration method is improved based on the idea of step-by-step calculation of internal and external parameters of the system,and the related technical requirements in the calibration process are analyzed.Combining the imaging characteristics of projectile targets in image sequences,a data processing algorithm framework is proposed.At the same time,two types of image point matching methods with the same name are analyzed and compared,and the feasibility verification is carried out by taking the typical semi-global dense stereo matching method SGBM as an example.According to the proposed data processing algorithm framework,detailed algorithms are given for image preprocessing,projectile target positioning and segmentation,and projectile center extraction in the process of projectile target image sequence processing.Taking into account the influence of illumination on image interpretation,Gamma correction algorithm is introduced.At the same time,in order to reduce tedious manual operations and meet the needs of fast and automatic processing of image sequences,an adaptive Gamma correction algorithm is improved,and only an ideal Gamma correction algorithm is required each time.Normalizing the average brightness value can complete the illuminance adjustment of all images and improve the image quality.In order to achieve continuous frame projectile target positioning,a TLD target tracking algorithm with high robustness and real-time performance is adopted,which can better adapt to the complex background,illumination changes,target scale scaling and appearance deformation during the tracking process to achieve stable tracking.The Grab Cut foreground target segmentation algorithm is used to achieve the secondary extraction of the target,and an image containing only the bottom information of the target is obtained,and the morphological closed operation is further used to optimize the bottom shape.On this basis,the Canny edge detection algorithm is used to automatically process the edge contour of the bottom of the simulated projectile target,and the least squares ellipse fitting method and the geometric moment method are respectively used to calculate the sub-pixel coordinates of the bottom center of the target.The experimental parameters were designed and the appropriate calibration board was selected,and the test system was built.The test system calibration experiment was carried out,the calibration error was calculated and the external factors that produced the calibration error were analyzed.Then the space coordinates of the simulated projectile target were measured,and the system measurement results were compared with the total station measurement results.The results show that the test system has a high degree of automation,fast measurement speed,and is not susceptible to environmental interference.The average static positioning error is14.3mm,and the trajectory measurement result is relatively accurate,which verifies the feasibility of the test method to achieve 100-meter trajectory measurement of bullets.