-
NETD是用来描述毫米波探测器可区分最小温差的物理量,与探测器噪声特性、灵敏度、响应度等指标息息相关,是毫米波探测器最重要的参数之一[14]。
NETD的定义如下:
$$ NETD=\frac{\overline{{{V}_{rms}}}}{{R}_{slope}}=\frac{\overline{{{V}_{rms}}}}{\Delta V/\Delta T} $$ (1) 式中:
$\overline{{V}_{rms}}$ 是毫米波探测器输出电压信号的噪声均方根值 (Root Mean Square,RMS);$ {R}_{slope}=\Delta V/\Delta T $ 是信号的传递函数。实际测量时,通常测量毫米波探测器对不同温度黑体目标响应信号,取所有温度下信号均方根值的均值作为$\overline{{V}_{rms}}$ ,计算VT曲线中线性部分的斜率为$ {R}_{slope} $ [14]。从NETD的定义很明显看出,积分时间增长,$\overline{{V}_{rms}}$ 会减小,并进一步导致NETD降低,因此在测量微弱信号时,可以通过延长积分时间以获得更好的NETD。采用文献[14]所述NETD测试方法,搭建了自动化测试系统,系统如图8所示。毫米波探测器垂直悬挂在水上方,探测器喇叭口距离水面为5 cm,水自身高度不少于2 cm,水面面积足够大以保证覆盖全部毫米波探测器接收角,温度计探头用于水温测量。探测器输出电压与温度计两端电压均通过数据采集卡(Data Acquisition,DAQ)进行采集,采集到的数据送入电脑中进行处理并实时显示结果。
测量了双极化探测器两个通道四个极化方向的NETD值,测试过程中水温由72 ℃逐渐降低到34 ℃,整个过程约耗时27.2 min,在408个温度点采集数据,相邻温度点间间隔时间约4 s,每个温度点通过2 s积分时间采集获得,每个温度点内采集探测器输出电压200个点(10 ms积分时间),空余时间用于程序运算与结果实时显示。最终测量得到VT曲线如图9(a)所示,进行线性拟合得到直线方程标注在图中,可见四个极化方向偏置电压相差较大,但是斜率
$ {R}_{slope} $ 较为接近。图9(b)所示为探测器在不同温度下$ {V}_{rms} $ 曲线,图中虚线标注为所有温度下$ {V}_{rms} $ 均值$ \overline{{V}_{rms}} $ 。图 9 (a)不同水温下探测器响应电压曲线;(b)不同水温下探测器响应均方根值
Figure 9. (a) Detector response voltage to the hot water at different temperatures; (b) RMS noise of measured response voltage to the hot water at different temperatures
所有探测器通道的
$ {R}_{slope} $ 、$ \overline{{V}_{rms}} $ 以及NETD值汇总在表1中,可见所有探测器通道的NETD均在0.3 K(10 ms积分时间)以下,证明探测器具有很高的灵敏度,满足被动式毫米波遥感成像使用需求。表 1 NETD测量结果
Table 1. NETD measurement results
Detector Rslope /V·K−1 $\overline{ {V}_{rms} }$ /V NETD/K Ch1-V 0.002 533 29 0.000 670 578 0.265 Ch1-H 0.002 177 11 0.000 571 628 0.263 Ch2-V 0.001 889 29 0.000 509 749 0.270 Ch2-H 0.001 839 05 0.000 570 989 0.310 -
为了验证双极化探测器在被动式毫米波遥感成像中的作用,搭建了一套被动式多极化毫米波遥感探测雷达,在城市环境下进行了双极化被动式毫米波遥感成像实验。
搭建的多极化毫米波遥感探测雷达如图10(a)所示,雷达采用主镜口径0.6 m的卡塞哥伦天线聚焦,使用双极化探测器探测W波段毫米波信号,将收集到的W波段毫米波信号功率转换为输出电压并由DAQ采集,整个结构安装在一套三维电动转台上,可实现水平和俯仰方向机械扫描以及探测器绕轴旋转。详细雷达介绍见参考文献[20],此处不再重复叙述。利用该雷达,对图10(b)所示区域进行了双极化被动式毫米波遥感成像,成像区域覆盖水平方向−30°~30°,竖直方向覆盖−50°~10°,雷达安置在5楼上,向下倾斜成像,目标成像距离约为50~300 m,每个像素点积分时间为12.5 ms,扫描角度间隔为0.05°,测试时的气温约为30 ℃。
图 10 (a)被动式多极化毫米波遥感探测雷达示意图;(b)成像区域光学照片
Figure 10. (a) Schematic diagram of multi-polarization passive millimeter-wave remote radar; (b) Visible image of imaging scenes
全极化信息可以采用TI、TQ、TU、TV四个Stokes参量表述[14,17],其计算公式如下,其中TI、TQ、TU、TV表述四个Stokes参量,TBh、TBv、
$ T{B}_{+45°} $ 、$ T{B}_{−45°} $ 、TBcl、TBcr分表表示水平极化信号、竖直极化信号、+45°极化信号、−45°极化信号、左旋极化信号、右旋极化信号,此处角度以水平线为参考平面。$$ \left[\begin{array}{c}T I\\ T Q\\ T U\\ T V\end{array}\right]=\left[\begin{array}{c}(T Bh+T Bv)/2\\ T Bv-T Bh\\ T {B}_{+45^{\circ}}-T {B}_{-45^{\circ}}\\ T Bcl-T Bcr\end{array}\right] $$ (2) 双极化测量时,将双极化毫米波探测器水平放置,同时完成TBh和TBv的探测,并可根据式(2)计算得到TI和TQ两个Stokes参量。进一步的,根据文献[14,17],还可以计算得到线性极化差异率(linear polarization difference ratio, LPDR)、单一极化率(the degree of sole polarization, DoSP)、线性极化率(linear polarization ratio,LPR)、线性极化率1(modified LPR 1,mLPR1)、极化率(polarization percentage, PP),同时提出了一种新的计算参量、线性极化率4(modified LPR 4,mLPR4),上述参量计算方法汇总如下。
$$ LPDR=\frac{Tob j-T Bv}{T Bv-TBh} $$ (3) $$ DoS P=\frac{Tob j-T Bv}{2T ob j-T Bv-T Bh} $$ (4) $$ LP R=\frac{T Bh-T ob j}{T Bv-T ob j} $$ (5) $$ mLPR1=\frac{T Bh-E\left(T I\right)}{T Bv-E\left(T I\right)} $$ (6) $$ mLPR4=\frac{T Bh}{T Bv} $$ (7) $$ P P=\frac{\left|T Q\right|}{T I} $$ (8) 式中:Tobj为物体的实际温度(文中全部按照303 ℃计算);E(TI)为所有像素点TI的平均值;M(TI)为所有像素点TI的中间值。
探测得到的TBh和TBv图像如图11所示。图中红色表示毫米波亮温高,蓝色表示毫米波亮温低。为了方便描述,在成像范围内选取了10个区域进行编号,对应的区域编号在图10(b)中标出。区域1表示远处的高层建筑,区域3表示视场范围两侧的高层建筑,主要材料为混凝土,区域6表示沥青道路和石板路,毫米波吸收系数较高,在太阳照射下本身温度也较高,因此表现出较高的毫米波亮温(深红色)。区域2表示远处的天空,呈现出较低的亮温。区域5表示植被,富含水分,辐射系数高,但是本身温度明显低于沥青道路,因此呈现出浅红色。区域7~10分别表示停在近处的汽车、停在远处的汽车、下水道井盖、空调外机,这些都是金属制品,本身不向外辐射毫米波信号,主要反射外界毫米波信号,由于总有一些角度反射的是来自天空的极冷信号,因此这些金属制品也表现出很低的毫米波亮温(蓝色)。区域4表示的是对面建筑窗户外侧向外倾斜45°的挡板,这些挡板同样主要反射天空信号,因此毫米波亮温较低。
TI图像表示毫米波强度分布图,无极化信息,得出的结论与TBh和TBv基本相同。TQ图像表示水平和竖直两个极化方向信号强度的差,很明显天空、混凝土墙面、沥青道路、石板路和窗户挡板等目标表现出较小的极化差别,而汽车则呈现出一定的区别,这是由于汽车为金属材质且角度众多,某些角度反射了环境中的极化信号,导致汽车呈现出一定的极化差异。
LPR图像与mLPR1
图像中汽车等金属制品呈现出剪影效果,其边缘轮廓可以方便的提取出来,这与文献[14,17]的描述是一致的。 LPDR和DoSP图像中,不同材质的物品,如汽车(金属)、绿色植被与地面(包括石板路、沥青路和干草土路)之间呈现出较为明显的区别,尤其是在LPDR图片中,汽车的前挡板(金属)和挡风玻璃(玻璃)呈现出明显的区别。这表明LPDR和DoSP图像有助于金属制品的识别和检出,并且具备一定的物质材质区分能力。
mLPR4
和PP表示水平与竖直两个极化方向之间的差异,与TQ图像类似,混凝土、石板路、植被、天空以及窗户挡板等非极化材质物品由于信号很小而变得无法辨认,这就使得汽车等极化差异物品被突出显示,很容易识别出。相比之下,mLPR4的突出显示效果更加明显。这有助于从自然环境中发现人造金属制品。
Dual polarization millimeter-wave detector for millimeter-wave remote sensing imaging
-
摘要: 针对多极化毫米波遥感成像重大需求,报道了一种高集成度双极化毫米波探测器。探测器采用固态电子学方案,通过喇叭天线接收毫米波信号,利用正交模耦合器进行极化信号分离,采用两级低噪放芯片放大信号,通过肖特基二极管实现功率检测。每支探测器集成两路接收通道,每个接收通道可同时探测水平与竖直两个极化方向信号。经测量,探测器探测频段为W波段(75~110 GHz),平均等效噪声温差达到0.3 K(@10 ms),尺寸不足155 mm×15 mm×20 mm,质量<0.3 kg。利用该探测器实现城市环境双极化毫米波遥感成像,成像结果表明,相比于单极化遥感成像,双极化遥感成像可以获得更多极化信息,可对城市环境下常见物品进行突出显示,这有助于今后进行自动物品识别、轮廓提取与远程物质成分识别。Abstract: To meet the great demand of multi polarization millimeter-wave remote sensing imaging, a highly integrated dual polarization millimeter-wave detector is reported. The detector adopts the solid-state electronics scheme. A horn antenna is used to collect the millimeter-wave radiation signal. An ortho-mode transducer structure (OMT) is utilized to orthogonal polarization states of the incident signal into the vertical and horizontal polarization channels. Dual polarization signal power detection is realized by Schottky diode detector after two stage low noise amplifier. The detector, which shows a high level of integration, has two same channels while each channel can receive two orthogonal polarization states of the incident millimeter-wave radiation simultaneously. The measurement results display that the detection band is W-band (75-110 GHz), the average Noise Equivalent Temperature Difference(NETD) reaches 0.3 K(@10 ms), the detector’s size is less than 155 mm×15 mm×20 mm, and the detector’s weight is less than 0.3 kg. Based on the dual polarization detector, an outdoor area in the city has been imaged and various polarization parameter images are generated to analyze the polarization characteristics. The results show that, compared with single polarization remote sensing, dual polarization remote sensing can obtain more polarization information, common objects in urban environment can be highlighted, which is helpful for automatic object recognition, contour extraction in the future, and remote material component recognition.
-
Key words:
- millimeter /
- detect /
- polarization /
- remote sensing /
- imaging
-
表 1 NETD测量结果
Table 1. NETD measurement results
Detector Rslope /V·K−1 $\overline{ {V}_{rms} }$ /VNETD/K Ch1-V 0.002 533 29 0.000 670 578 0.265 Ch1-H 0.002 177 11 0.000 571 628 0.263 Ch2-V 0.001 889 29 0.000 509 749 0.270 Ch2-H 0.001 839 05 0.000 570 989 0.310 -
[1] 常庆瑞. 遥感技术导论[M]. 北京: 科学出版社, 2004. [2] Zhang T R, Wang G L, Liang J, et al. Status and trends of space microwave remote sensing technology development [J]. Spacecraft Engineering, 2021, 30(6): 52-61. (in Chinese) doi: 10.3969/j.issn.1673-8748.2021.06.006 [3] Luo Haibo, Zhang Junchao, Gai Xingqin, et al. Development status and prospects of polarization imaging technology [J]. Infrared and Laser Engineering, 2022, 51(1): 20210987. (in Chinese) doi: 10.3788/IRLA20210987 [4] Wang Jun, Jiang Yadong. Research development about room temperature terahertz detector array technology with microbolometer structure (invited) [J]. Infrared and Laser Engineering, 2019, 48(1): 0102001. (in Chinese) doi: 10.3788/IRLA201948.0102001 [5] Wikner D A, Samples G. Polarimetric passive millimeter-wave sensing[C]//Passive Millimeter-Wave Imaging Technology V, SPIE, 2001, 4373: 86-93. [6] Samluk J P, Schuetz C A, Martin R D, et al. 94 GHz millimetre-wave imaging system implementing optical upconversion[C]//Millimetre Wave and Terahertz Sensors and Technology, SPIE, 2008, 7117: 238-244. [7] Wilson J P, Eng D L K, Kozacik S T, et al. Display of polarization information for passive millimeter-wave imagery [J]. Optical Engineering, 2012, 51(9): 091607. doi: 10.1117/1.OE.51.9.091607 [8] Duric A, Magun A, Murk A, et al. The fully polarimetric imaging radiometer SPIRA at 91 GHz [J]. IEEE Transactions on Geoscience and Remote Sensing, 2008, 46(8): 2323-2336. doi: 10.1109/TGRS.2008.917212 [9] Duric A, Mallepell M, Wuetrich M, et al. Fully-polarimetric passive scanning imager at millimeter wavelengths[C]//Passive Millimeter-Wave Imaging Technology VIII, International Society for Optics and Photonics, 2005, 5789: 35-43. [10] Murk A, Stähli O, Mätzler C, et al. Polarimetric imaging with the 91 GHz radiometer SPIRA[C]//Millimetre Wave and Terahertz Sensors and Technology III, International Society for Optics and Photonics, 2010, 7837: 78370F. [11] Tedeschi J, Bernacki B, Sheen D, et al. Fully polarimetric passive w-band millimeter wave imager for wide area search[C]//Polarization Science and Remote Sensing VI, SPIE, 2013, 8873: 243-256. [12] Su Jinlong, Tian Yan, Hu Fei, et al. Material clustering using passive millimeter-wave polarimetric imagery [J]. IEEE Photonics Journal, IEEE, 2019, 11(1): 5500109. doi: 10.1109/JPHOT.2018.2881287 [13] Hu Yan, Hu Fei, Yang Zhengwu, et al. Complex permittivity estimation from millimeter-wave radiometry [J]. IEEE Geoscience and Remote Sensing Letters, 2021, 18(7): 1254-1258. doi: 10.1109/LGRS.2020.2996380 [14] Cheng Yayun, Hu Fei, Wu Hongfei, et al. Multi-polarization passive millimeter-wave imager and outdoor scene imaging analysis for remote sensing applications [J]. Optics Express, 2018, 26(16): 20145. doi: 10.1364/OE.26.020145 [15] Hu Yan, Su Jinlong, Hu Fei, et al. Information extraction from polarized MMW radiation based on the weak correlation between emission and reflection [J]. Optical Materials Express, 2021, 11(7): 2059. doi: 10.1364/OME.430138 [16] Su Jinlong, Wu Hongfei, Li Pengfei, et al. Detection for ship by dual-polarization imaging radiometer [J]. Optics Express, 2021, 29(17): 27830. doi: 10.1364/OE.432432 [17] Tang Fangzhou, Gui Liangqi, Liu Jinbang, et al. Metal target detection method using passive millimeter-wave polarimetric imagery [J]. Optics Express, 2020, 28(9): 13336. doi: 10.1364/OE.390385 [18] Cheng Yayun, Hu Fei, Gui Liangqi, et al. Polarization-based method for object surface orientation information in passive millimeter-wave imaging [J]. IEEE Photonics Journal, 2016, 8(1): 5500112. doi: 10.1109/JPHOT.2015.2507359 [19] Hu Fei, Cheng Yayun, Gui Liangqi, et al. Polarization-based material classification technique using passive millimeter-wave polarimetric imagery [J]. Applied Optics, 2016, 55(31): 8690. doi: 10.1364/AO.55.008690 [20] An Deyue, Tu Hao, Qi Denggang, et al. Outdoor multi-polarization passive millimeter–wave imaging via integrated W-band dual polarization direct detection radiometer [C]//46th International Conference on Infrared, Millimeter and Terahertz Waves (IRMMW-THz). IEEE, 2021: 1-2.