摘要
报道了高光谱红外时间空间混合调制型傅里叶干涉光谱成像方法,可应用于环境气体的二维空间监测,有助于提高化学防护的检测效率,迅速定位化学污染的源头。针对气体云团特征探测的技术要求和高光谱分辨要求,技术采用时空混合调制红外干涉成像被动探测方案,利用超长波红外焦平面器件接收场景的干涉图像,采用高速热成像电路采集干涉图像序列,生成干涉数据立方体,采用了并行高速数据处理服务器进行数据处理,以及智能图像鉴别算法,实现了云团气体种类和分布的快速识别。自研的CHIPED-1长波红外高光谱成像原理实验装置的探测灵敏度指标噪声等效辐射通量密度NESR在单次采样时达到了5.6×10-8 W·(cm-1·sr·cm2)-1,与商品化时间调制干涉高光谱成像仪相当;反映了技术的先进性,并留有较大的改进空间。文章以野外探测实验为例,研究了二维分布化学气体VOC的高光谱成像探测方法。在复杂背景和低试验浓度情况下实现红外光谱信息的光谱特征选择、提取及优化技术,快变背景下的信号预处理和基于辐射的无背景亮温鉴别技术,可以清楚看到氨气蒸气的空间分布。
In this paper,the method of hyperspectral infrared time-space hybrid modulation Fourier interference spectral imaging is put forward,which can be applied to the two-dimensional space monitoring of environmental gas,improve the detection efficiency of chemical protection,and quickly locate the source of chemical pollution.According to the technical requirements of gas cloud feature detection and hyperspectral resolution,the technology adopts the time-space hybrid modulation infrared interference imaging passive detection scheme,uses the ultra long wave infrared focal plane device to receive the interference image of the scene,uses the high-speed thermal imaging circuit to collect the interference image sequence,generates the interference data cube,and uses the parallel high-speed data processing server for data processing.In addition,the intelligent image identification algorithm can realize the fast recognition of cloud gas species and distribution.The detection sensitivity index of the self-developed chip-1 long wave infrared hyperspectral imaging principle experimental device noise equivalent radiation flux density NESR reached 5.6×10-8 w·(cm-1·sr·cm2)-1.It is equivalent to the commercialized time modulated interferometric hyperspectral imager,reflecting the advanced technology and leaving much room for improvement.In this paper,the hyperspectral imaging detection method of two-dimensional distribution of chemical gas VOC is studied,taking the field detection experiment as an example.The spatial distribution of ammonia vapor and SF6 can be clearly seen by the technology of feature selection,extraction and optimization of infrared spectrum information in complex background and low test concentration,signal preprocessing in fast changing background and light temperature identification without background based on radiation.
作者
杨智雄
雷正刚
余春超
郑为建
严敏
张培仲
胡耀航
曾邦泽
向麟槐
段绍丽
汪凌志
YANG Zhi-xiong;LEI Zheng-gang;YU Chun-chao;ZHENG Wei-jian;YAN Min;ZHANG Pei-zhong;HU Yao-hang;ZENG Bang-ze;XIANG Lin-huai;DUAN Shao-li;WANG Ling-zhi(Kunming Institute of Physics,Kunming 650223,China)
出处
《光谱学与光谱分析》
SCIE
EI
CAS
CSCD
北大核心
2020年第S01期67-68,共2页
Spectroscopy and Spectral Analysis
关键词
相关系数
亮温差谱
红外高光谱成像
气体探测
Correlation coefficient
Bright Temperature difference spectrum
Infrared hyperspectral imaging
Gas detection
