-
在激光近场探测中引入光谱识别的功能,从系统设计角度出发,主要解决三个问题,即光源的选择、信息处理、光谱分光的方法。文中提出一种适用于激光近场探测的一体式光谱分析传感器,将在第2章专门介绍,即解决在激光近场探测应用环境限制下的光谱分光问题,此章将主要针对光源选择与信息处理进行分析和建议。
-
激光近场探测中的光谱系统的目的是分析目标的材料属性,近而确定被探测物的材料是否为目标,因此在大多数情况下,需要对反射光谱进行采集,利用反射光谱分析物质属性,而反射光谱在获得的时候需要照明光源。
一般有两种方式,即依靠环境光源(主要是太阳光)的被动光源和主动制造的光源。被动光源依靠外界环境光,主要是太阳光进行照明,弹载平台本身不需要携带额外的照明设备,增加了可靠性,且太阳光光谱范围广,可以增加目标识别的种类与范围,对激光近场探测的抗干扰能力提升有明显优势,但其使用环境受限,只能在光照充足的时候使用。主动光照式需要弹载平台自身携带照明光源,分为宽谱光源和窄带激光两种,宽谱光源可以以照明弹或者固定光源的形式实现,实现丰富物质种类的判断与识别,窄带激光可以与目前的测距激光共用激光器,再设计和安装的成本都很低,缺点是可识别的种类较少,这几种照明方式的优劣比较如表1所示。
表 1 激光近场探测中的光谱照明方案对比
Table 1. Comparison of spectral illumination schemes in laser near field detection
Lighting method Passive Initiative Sunlight Broad spectrum
light sourceLaser Identification type Numerous Numerous Less Application Environment Limited Unlimited Unlimited Complexity Low High Low Concealment High Low Moderate -
对于光谱数据典型的处理方法如图1所示,在传统的数据处理流程中,需要先通过暗噪声和入射光的光谱校正来得到目标的反射光谱,由于获得的光谱数据一般维数较多,随后进行包含特征提取与识别、分类器设计、分类决策的标准模式识别流程,然而由于激光近场探测在工作阶段要处于高速判别的状态,传统的数据决策体系很难去适应高速的判别机制,需要在整个流程中加以简化,甚至联合硬件进行优化设计。而笔者在第2章所提出的光谱分析传感器也具有再设计简单的优势,适合针对不同的目标和应用环境进行快速、低成本的设计。
图2展示了基于光谱数据库分析的光谱特征提取算法开发闭环流程,由于在激光近场探测的目标判别中,所需判别的种类与导引阶段相比大大降低,因此需要在弹载平台总体方面给出总体设计,判断清激光近场探测所需解决的目标识别问题的范围,而这一类目标范围是传统依靠距离判断不出来的目标,而光谱作为专业的目标识别手段,在定义清需求后,有很大的概率不需要极其复杂的数据处理算法就可以进行判别,文中第3章就给出了一个这样的例子。
Research on high recognition accuracy laser near field detection sensor based on spectrum characteristics
-
摘要: 激光近场探测是激光测量重要的应用方向之一,尤其是基于距离信息的近场探测系统有着巨大的应用前景。传统的激光近场探测系统只依靠距离信息给出信号,而不判别反射距离信号的物体是否为目标,可能导致误报。基于光谱特性进行目标识别会大大减少激光近场测量系统的误报率,极大提高判别准确度。设计实现了一种单芯片的光谱分析器件,利用半导体技术将光谱元件和光电元件进行单芯片的集成,从而减小了传统光谱分析系统的体积和质量。实验结果表明,该微系统级的光谱分析芯片具有识别既定伪装物体的能力,使得该技术具有在激光近场探测中应用的前景。Abstract: Laser near field detection plays a crucial role in the laser measurement products series and has been successfully and widely applied in lots of field. Traditional laser near field detection system transmits the triggering signal according to the distance information, without being able to judge whether the distance signal is obtained from the target or not. As a result, a false positive could happen and the performance of the system would be weakened. However, recognition based spectral information can improve accuracy. In this paper, a target recognition system, which was compatible for the classical laser fuse scheme, based on spectrum analysis was proposed to increase the detection accuracy of the laser fuse system. Some suggestions about the general design of the improved laser fuse system with spectrum analysis sensor were also offered including active and passive detection with imaging function as an option. Moreover, based on the CMOS technology, a monolithic spectrum analysis sensor was designed and fabricated, which integrated the spectral element on top of the opto-electronical sensor in one chip. Thanks to the greatly reduced size and the weight, this spectrum analysis sensor had potential application in practice. The experiment results show that the capability of such system on chip (SOC) sensor in distinguishing the target from the camouflage.
-
Key words:
- laser near field detection /
- system on chip /
- CMOS technology /
- spectrum analysis /
- classification
-
表 1 激光近场探测中的光谱照明方案对比
Table 1. Comparison of spectral illumination schemes in laser near field detection
Lighting method Passive Initiative Sunlight Broad spectrum
light sourceLaser Identification type Numerous Numerous Less Application Environment Limited Unlimited Unlimited Complexity Low High Low Concealment High Low Moderate -
[1] 陈慧敏, 贾晓东, 蔡克荣. 激光引信技术[M]. 北京: 国防工业出版社, 2016. Chen Huimin, Jia Xiaodong, Cai Kerong. Laser Fuze Technology[M]. Beijng: National Defence Industry Press, 2016. (in Chinese) [2] 吕华, 姚宏宝, 陈平. 大视场激光引信探测目标识别方法研究[J]. 红外与激光工程, 2005, 34(4): 52−54. Lv Hua, Yao Hongbao, Chen Ping. Target identifying of large visual angle to the laser fuse [J]. Infrared and Laser Engineering, 2005, 34(4): 52−54. (in Chinese) [3] Gowen A A, O"Donnell C P, Cullen P J, et al. Hyperspectral imaging – an emerging process analytical tool for food quality and safety control [J]. Trends in Food Science & Technology, 2007, 18(12): 590−598. [4] Heikki Saari, Ismo Pellikka, Liisa Pesonen, et al. Unmanned Aerial Vehicle (UAV) operated spectral camera system for forest and agriculture applications[C]// Proc SPIE, Remote Sensing for Agriculture, Ecosystems, and Hydrology XIII, 2011, 8174: 81740H. [5] 许洪, 王向军. 多光谱、超光谱成像技术在军事上的应用[J]. 红外与激光工程, 2007, 36(1): 13−17. Xu Hong, Wang Xiangjun. Applications of multispectral/hyperspectral imaging technologies in military [J]. Infrared and Laser Engineering, 2007, 36(1): 13−17. (in Chinese) [6] Eismann M T, Seldin J H, Schwartz C R, et al. Target detection in desert backgrounds: infrared hyperspectral measurements and analysis[C]// Proc SPIE-The International Society for Optical Engineering, 1995, 2561: 80-97. [7] Liu Shuyang, Zhang Chen, Zhang Yunhao, et al. Miniaturized spectral imaging for environment surveillance based on UAV platform[C]// AOPC 2017, Proceedings of Optical Spectroscopy and Imaging, 2018: 10461. [8] Nicolaas Tack, Andy Lambrechts, Philippe Soussan, et al. A compact, high-speed, and low-cost hyperspectral imager[C]// Proc SPIE, Silicon Photonics VII, 2012, 8266: 82660Q. [9] Bert Geelen, Nicolaas Tack, Andy Lambrechts. A compact snapshot multispectral imager with a monolithically integrated per-pixel filter mosaic[C]//Proc SPIE, Advanced Fabrication Technologies for Micro/Nano Optics and Photonics VII, 2014, 8974: 89740L. [10] Bert Geelen, Nicolaas Tack, Andy Lambrechts. A snapshot multispectral imager with integrated tiled filters and optical duplication[C]// SPIE, Advanced Fabrication Technologies for Micro/Nano Optics and Photonics VI, 2013, 8613: 861314.