基于直线扫描的引信钢球分层成像检测算法
Layered imaging detection algorithm for fuze steel balls based on linear scanning
吉泓洁 1陈平 1赵耀霞 1张永伟 2朱勋2
作者信息
- 1. 中北大学 信息探测与处理山西省重点实验室,太原 030051
- 2. 重庆长安望江工业集团有限公司,重庆 401120
- 折叠
摘要
针对现有的火药引信装配正确性检测方法中存在的无法准确检测遮挡严重的零件的问题,提出了引信中钢球分层成像定位检测算法.将X射线源直线移动扫描待测物体形成多个虚拟的小孔径,结合光学合成孔径原理对所有小孔径进行合成,将合成的虚拟大孔径成像在不同距离深度的平面上,实现对目标的定位和检测.试验结果表明:所提算法能够在火药引信钢球相互重叠的情况下利用深度信息进行分层成像,提供待测物体的深度信息,能够对不同层次的目标进行检测,确定待检测物体的数量,解决了现有的检测方法在检测过程中出现的漏检问题.
Abstract
Aiming at the problem of inaccurate detection of severely obstructed parts in existing methods for detecting the correctness of gunpowder fuse assembly,a layered imaging positioning detection algorithm for steel balls in fuses is proposed.Firstly,move the X-ray source in a straight line to scan the object to be tested to form multiple virtual small apertures;Secondly,based on the principle of optical synthetic aperture,all small apertures are synthesized;Finally,the synthesized virtual large aperture is imaged on planes at different distances and depths to achieve target localization and detection.The experimental results show that the proposed algorithm can use depth information for layered imaging in the case of overlapping steel balls in the powder fuse,providing depth information of the object to be tested,and then detecting targets at different levels to determine the number of objects to be tested,solving the problem of missed detection in existing detection methods during the detection process.
关键词
直线扫描/X射线/钢球/分层成像/深度信息Key words
linear scanning/X-ray/steel ball/hierarchical imaging/depth information引用本文复制引用
基金项目
国家自然科学基金(62122070)
国家自然科学基金(61971381)
山西省自然科学基金(20210302124190)
山西省自然科学基金(20210302124191)
出版年
2024
NETL
NSTL
万方数据