一种红外成像导引头探测视场实时扩展算法

Improved Method for IR-guided Weapons to Enlarge Field of View

  • 摘要: 由于红外探测器灵敏阈相对较低的特点,红外探测体制的导引头多用于末端防空领域,或者与其他制导模式相组合完成打击任务。惯导红外复合制导是近几年在国内外流行的制导模式,低目标指示精度下的红外交班是复合制导体制一项重要的关键技术,框架扫描与视场实时拼接技术可以有效完成低目标指示精度下的红外交班。本文根据小口径两轴框架位标器特点,设计了框架闭环扫描搜索算法。同时分别提出了基于位标器惯性信息和图像特征信息的图像实时拼接方法,并针对惯性信息拼接积累误差大和图像特征拼接不适用于极端复杂场景和极端简单场景的缺点,设计了将惯性信息与图像特征偏移信息相融合的图像拼接方法,最后通过仿真验证了该方法不仅可以有效消除积累误差,在极端特殊场景下也可以达到品质较好的图像拼接效果,获得了更大的红外探测视场。

     

    Abstract: Owing to comparatively small working distances, IR-guided weapons are usually used in either terminal air defense or complex guided weapons. Radar/IR and INS/IR composites have recently gained popularity as tactical weapons. Reliable switching to the infrared mode with low guide precision is an important technique in composite seekers, making scan-technique and view-compound obligatory approaches for reliable handover. Based on the characteristics of the diminutive coordinator, this study designs a quick-scan technique, integrating into the INS assembly arithmetic-based image characteristics. To address the anterior error of complex images, compound arithmetic based on the Kalman filter is designed, combining INS communication with image characteristics. Finally, through simulations, the method was demonstrated to effectively eliminate accumulated errors. High-quality image stitching can also be achieved in extremely specific scenarios. This method yields a larger infrared detection field of view.

     

/

返回文章
返回