摘要
复杂环境下的红外弱小目标的检测是光学制导的研究热点。如果红外目标距离探测器较远,其光学成像显示为无纹理的弱点源。噪声干扰使得目标与噪声难以区分。基于此,本文提出基于侧抑制网络的红外背景抑制算法,然后研究高信噪比下的红外弱小目标检测技术,首先抑制红外背景,然后分割红外图像,确定红外目标,基于目标运动的时间相关性进行单帧检测,过滤虚警点,实现弱小目标检测,最后,结合实际数据验证检测技术的可行性,结果显示该技术对于复杂环境下红外弱小目标的检测率达到96%,实际应用效果较好。
关键词: 复杂环境;空间背景;红外检测;弱小目标;目标检测
Abstract
Infrared dim small target detection in complex environment is a research hotspot in optical guidance. If the infrared target is far away from the detector, its optical imaging appears as an untextured source of weakness. Noise interference makes the target indistinguishable from noise. Based on this, this paper proposes the infrared background suppression algorithm based on side suppression network, and then studies the infrared dim dim target detection technology under high signal-to-noise ratio. Firstly, the infrared background is suppressed, and then the infrared image is segmtioned to determine the infrared target. Based on the time correlation of the target motion, the single frame detection is carried out to filter the false alarm points and realize the dim dim target detection. Combined with the actual data to verify the feasibility of the detection technology, the results show that the detection rate of the technology for infrared dim small targets in complex environment reaches 96%, and the practical application effect is good.
Key words: Complex environment; Spatial background; Infrared detection; Weak target; Object detection
参考文献 References
[1] 樊华,武文波,焦智,等. 基于三维滤波的红外弱小目标检测技术研究[J]. 电子技术应用,2021,47(3):106-110.
[2] 杨德振,喻松林,冯进军,等. 机载复杂场景下的低虚警红外目标检测[J]. 光学精密工程,2022,30(1):96-107.
[3] 钮赛赛,周华伟,朱婧文,等. 基于YOLO智能网络的红外弱小多目标检测技术[J]. 上海航天,2019,36(5):28-34
[4] 孙熊伟. 复杂背景下海面红外小目标快速检测技术研究[D]. 安徽:中国科学技术大学,2019.
[5] 杨其利,周炳红,郑伟,等. 基于全卷积网络的红外弱小目标检测算法[J]. 红外技术,2021,43(4):349-356.
[6] 蔡云泽,张彦军. 基于双通道特征增强集成注意力网络的红外弱小目标检测方法[J]. 空天防御,2021,4(4):14-22.
[7] 杨海静. 基于视觉特征融合的红外弱小目标检测方法研究[D]. 重庆:重庆邮电大学,2020.
[8] 王宇翔,韩振铎,王宏敏. 基于多向差异度的红外弱小目标检测算法[J]. 红外技术,2012,34(6):351-355.
[9] 周慧鑫,赵营,秦翰林,等. 多尺度各向异性扩散方程的红外弱小目标检测算法[J]. 光子学报,2015,44(9):146-150.
[10] 黄苏琦. 时空谱多特征联合红外弱小目标检测方法研究[D]. 四川:电子科技大学,2020.
[11] 寇志强,艾斯卡尔•艾木都拉. 局部最大熵的红外小目标快速检测方法[J]. 激光杂志,2020,41(7):18-22.
[12] 张秋实. 红外和可见光图像的融合分类及红外目标检测[D]. 北京:北京化工大学,2018.