E-mail Alert
RSS
Design and implement of a space-borne sun glint polarization parameter computing system
-
摘要:
太阳耀光影响光学遥感成像质量，可采用在探测器前加装偏振片的方式进行抑制，抑制效果取决于太阳和遥感器的相对位置以及偏振片的偏振方向。为实时准确地获取太阳耀光偏振信息，本文在星载大气校正仪上设计了一套星上太阳耀光偏振参数计算系统，利用大气校正仪670 nm波段的0°，60°，120°三个通道偏振图像实时计算耀光参数，并使用基于6S大气辐射传输模型的耀光参数建立晴空海洋偏振双向反射分布函数查找表，排除受云干扰的图像像元，最后利用实时探测数据进行高精度太阳耀光偏振方位角计算。系统以V5系列的现场可编程门阵列为计算平台，使用高层次综合工具进行算法的硬件实现，并在实验室内进行了实验验证。实验结果表明，系统计算偏振角误差与真实值相比在0.5°以内，在100 MHz主频时钟下一组25×25像元的数据计算时间消耗为19.47281 ms，FPGA资源使用率为41%。
Abstract:Sun glint is a significant confounding factor in passive optical remote sensing images. To mitigate this issue, a polarizer is typically incorporated in front of the remote sensor, leveraging the linear polarization characteristics of sun glint. The suppression effects depend on the relative position of the sun and the remote sensor, as well as the directions of polarizers. In this paper, we introduce a novel onboard system for the real-time computation of Sun glint polarization parameters, devised specifically for a spaceborne atmospheric correction instrument. Utilizing three channel polarization images (at 0°, 60°, and 120°) in the 670 band of the spaceborne atmospheric correction, we calculate the sun glint parameters and compared them against the 6S radiation transfer model, excluding image pixels heavily influenced by the could. The system is implemented using the V5 series Field Programmable Gate Array (FPGA) as the hardware platform, and the High-Level Synthesis Tool (HLS) as the software platform. The performance of the system is verified through a simple laboratory experiment, which demonstrates a calculation deviation within 0.5°. In terms of computational efficiency, the system processes a 25x25 pixel dataset in 19.47281 ms using a 100 MHz clock, with the highest resource utilization rate reaching 41%.
-
Overview: In passive optical remote sensing, the phenomenon of sun glint presents a substantial challenge in the acquisition and processing of high-quality images. Sun glint is the specular reflection from surfaces like water. Water bodies are characterized by low reflectivity, which classifies them as dark targets within the context of remote sensing. The radiation of sun glint is usually dozens of times higher than the target's radiation, and is easy to cause sensor saturation, leading to serious interference with the detection target. The current methods for suppressing solar glint in remote sensing imagery are mainly conducted on the ground. However, these approaches are often reactive rather than preventive and may not be suitable for real-time applications. According to Fresnel's law, the vertical component of sun glint is usually greater than the parallel component. In space, to mitigate this issue, a polarizer is typically incorporated in front of the remote sensor, leveraging the linear polarization characteristics of sun glint. The suppression effects depend on the relative position of the sun and the remote sensor, as well as the directions of polarizers. With the rapid development of satellite technology, the traditional method of installing parallel linear polarizers is difficult to meet our requirements. So, to suppress sun glint accurately and timely, we introduce a novel onboard system for the real-time computation of Sun glint polarization parameters, devised specifically for a spaceborne atmospheric correction instrument. Utilizing three channel polarization images (at 0°, 60°, and 120°) in the 670 nm band of the space-borne atmospheric correction, we calculate the sun glint parameters and compare them against the 6S radiation transfer model, excluding image pixels heavily influenced by the cloud. The system is implemented using the V5 series Field Programmable Gate Array (FPGA) as the hardware platform, and the High-Level Synthesis Tool (HLS) as the software platform. By utilizing the Cordic algorithm, converting data to appropriate datatypes, and implementing pipeline unrolling methods, we achieve a balanced trade-off between speed and resource allocation. A simple experiment was built to verify the system in the laboratory. The experiments performed that the calculation deviation is within 0.5°, calculating the 25 pixels×25 pixels data costs 19.47281 ms in 100 MHz clock, and the highest resource utilization rate accounts for 41%, meeting the requirements of the accuracy, real-time performance, and resource consumption.
-
-
图 2 在不同太阳天顶角下，太阳耀光偏振度和偏振角与观测天顶角的关系
Figure 2. The relationship between the degree of polarization and angle of polarization and observed zenith angle at different solar zenith angles
图 3 平行偏振片和特定角度偏振片抑制效果对比
Figure 3. Comparison of suppression effects between parallel polarizers and specific angle polarizers
图 6 线偏光源和大气校正仪探头整机
Figure 6. Line biased light source and atmosphere calibration instrument probe
表 1 查找表规则
Table 1. Ruler of the lookup table
Start End Step Average Solar zenith 0° 80° 8° View zenith 0° 80° 5° Differentials azimuth 0° 360° 10° Wind speed 4.1 m/s Wind azimuth 90° 
下载: 导出CSV
表 2 两种数据类型运算资源消耗对比
Table 2. Performance comparison of two datatype
Latency DSP48E FF LUT FLOAT 6 6 840 1182 AP_FIXED<32,16> 0 3 0 76
下载: 导出CSV
表 3 CORDIC算法迭代因子
Table 3. CORDIC algorithm iteration factor
i tanθ θ Iteration factor 0 1 45 0.70711 1 0.5 26.56505 0.63246 2 0.25 14.03624 0.61357 3 0.125 7.12502 0.60883 4 0.0625 3.57633 0.60765 5 0.03125 1.78991 0.607352 6 0.015625 0.89517 0.607278 ··· ··· ··· ···
下载: 导出CSV
表 4 三种算法性能与资源消耗比较
Table 4. Performance and resource comparison of three algorithms
Latency Interval DSP48E FF LUT CORDIC 4 1 0 442 3640 sin 39 39 3 226 1119 Taylor 36 36 6 2942 4678
下载: 导出CSV
表 5 优化前后性能与资源消耗比较
Table 5. Performance and resource comparison before and after optimization
Latency FF DSP48E Before After Before After Before After GHT 181 109 3317 4883 12 4 CDOLP 104 69 1852 3430 0 0 MDOLP 1397 368 8191 8729 14 20
下载: 导出CSV
表 6 0°实验测试结果
Table 6. 0° experiment results
Parameter Value Mean DN1 3717 Mean DN2 5562 Mean DN3 740 DOLP 0.967 Standard deviation of DOLP 0.023 AOP 4.377° Standard deviation of AOP 0.112
下载: 导出CSV
表 7 实验测试结果
Table 7. Experiment results
Angle of light MEAN DN1 MEAN DN2 MEAN DN3 DOLP Standard deviation of DOLP AOP Standard deviation of AOP 30° 1127 5937 3002 0.971 0.003 30.10 0.098 60° 728 3687 5612 0.976 0.002 60.45 0.126 90° 2963 1102 5982 0.974 0.003 90.35 0.100 120° 5569 734 3708 0.970 0.002 119.94 0.120 150° 5972 2971 1104 0.975 0.003 150.20 0.107 180° 3727 5559 735 0.973 0.002 180.30 0.110
下载: 导出CSV
-
[1] Kay S, Hedley J D, Lavender S. Sun glint correction of high and low spatial resolution images of aquatic scenes: a review of methods for visible and near-infrared wavelengths[J]. Remote Sens, 2009, 1(4): 697−730. doi: 10.3390/rs1040697
[2] Wang M H, Bailey S W. Correction of sun glint contamination on the SeaWiFS ocean and atmosphere products[J]. Appl Opt, 2001, 40(27): 4790−4798. doi: 10.1364/AO.40.004790
[3] Mobley C D. Polarized reflectance and transmittance properties of windblown sea surfaces[J]. Appl Opt, 2015, 54(15): 4828−4849. doi: 10.1364/AO.54.004828
[4] Hieronymi M. Polarized reflectance and transmittance distribution functions of the ocean surface[J]. Opt Express, 2016, 24(14): A1045−A1068. doi: 10.1364/OE.24.0A1045
[5] Fougnie B, Frouin R, Lecomte P, et al. Reduction of skylight reflection effects in the above-water measurement of diffuse marine reflectance[J]. Appl Opt, 1999, 38(18): 3844−3856. doi: 10.1364/AO.38.003844
[6] 刘志刚, 周冠华. 太阳耀光的偏振分析[J]. 红外与毫米波学报, 2007, 26(5): 362−365. doi: 10.3321/j.issn:1001-9014.2007.05.011
Liu Z G, Zhou G H. Polarization of sun glint[J]. J Infrared Millimeter Waves, 2007, 26(5): 362−365. doi: 10.3321/j.issn:1001-9014.2007.05.011
[7] Zhao H J, Ji Z, Zhang Y, et al. Mid-infrared imaging system based on polarizers for detecting marine targets covered in sun glint[J]. Opt Express, 2016, 24(15): 16396−16409. doi: 10.1364/OE.24.016396
[8] Avrahamy R, Milgrom B, Zohar M, et al. Improving object imaging with sea glinted background using polarization method: analysis and operator survey[J]. IEEE Trans Geosci Remote Sens, 2019, 57(11): 8764−8774. doi: 10.1109/TGRS.2019.2922827
[9] 张卫国. 海面太阳耀光背景下的偏振探测技术[J]. 中国光学, 2018, 11(2): 231−236. doi: 10.3788/co.20181102.0231
Zhang W G. Application of polarization detection technology under the background of sun flare on sea surface[J]. Chin Opt, 2018, 11(2): 231−236. doi: 10.3788/co.20181102.0231
[10] 朱鹤骞, 曲宏松. 海面背景耀光的自适应抑制系统[J]. 光学学报, 2022, 42(12): 1201006. doi: 10.3788/AOS202242.1201006
Zhu H Q, Qu H S. Adaptive suppression system of sea background flare[J]. Acta Opt Sin, 2022, 42(12): 1201006. doi: 10.3788/AOS202242.1201006
[11] Wang M S, Qiu S, Jin W Q, et al. Automatic suppression method for water surface glints using a division of focal plane visible polarimeter[J]. Sensors, 2023, 23(17): 7446. doi: 10.3390/s23177446
[12] 于新宇. 星载偏振光谱相机sCMOS成像系统设计[D]. 桂林: 桂林电子科技大学, 2023: 1–67.
Yu X Y. Design of sCMOS imaging system for space-borne polarization spectral camera[D]. Guilin: Guilin University of Electronic Technology, 2023: 1–67.
[13] Born M, Wolf E. Principles of Optics[M]. 5th ed. New York: Pergamon Press, 1975.
[14] Reda I, Andreas A. Corrigendum to “Solar position algorithm for solar radiation applications” [Solar Energy 76 (2004) 577–589][J]. Sol Energy, 2007, 81(6): 838. doi: 10.1016/j.solener.2007.01.003
[15] Cox C, Munk W. Measurement of the roughness of the sea surface from photographs of the Sun’s glitter[J]. J Opt Soc Am, 1954, 44(11): 838−850. doi: 10.1364/JOSA.44.000838
[16] Kotchenova S Y, Vermote E F. Validation of a vector version of the 6S radiative transfer code for atmospheric correction of satellite data. Part II. Homogeneous Lambertian and anisotropic surfaces[J]. Appl Opt, 2007, 46(20): 4455−4464. doi: 10.1364/AO.46.004455
[17] 陈震霆, 孙晓兵, 汪俊锋, 等. 近红外偏振辐射卫星数据的海洋耀光动态检测[J]. 遥感学报, 2019, 23(2): 215−229. doi: 10.11834/jrs.20197072
Chen Z T, Sun X B, Wang J F, et al. Dynamic detection of ocean glint from near-infrared polarized radiation satellite data[J]. J Remote Sens, 2019, 23(2): 215−229. doi: 10.11834/jrs.20197072
[18] Remer L A, Kaufman Y J, Tanré D, et al. The MODIS aerosol algorithm, products, and validation[J]. J Atmos Sci, 2005, 62(4): 947−973. doi: 10.1175/JAS3385.1
[19] Toubbe B, Bailleul T, Deuze J L, et al. In-flight calibration of the POLDER polarized channels using the Sun's glitter[J]. IEEE Trans Geosci Remote Sens, 1999, 37(1): 513−524. doi: 10.1109/36.739104
[20] Bicheron P, Leroy M, Hautecoeur O, et al. Enhanced discrimination of boreal forest covers with directional reflectances from the airborne polarization and directionality of Earth reflectances (POLDER) instrument[J]. J Geophys Res Atmos, 1997, 102(D24): 29517−29528. doi: 10.1029/97JD01330
[21] 钱鸿鹄. 多角度偏振成像仪实验室全视场偏振定标[D]. 合肥: 中国科学技术大学, 2017: 1–100.
Qian H H. Laboratory full field of view polarization calibration of directional polarimetric camera[D]. Hefei: University of Science and Technology of China, 2017: 1–100.
[22] 侯梦雨, 李正强, 谢一凇, 等. 国产卫星多角度偏振传感器的光谱特征云检测方法研究[J]. 大气与环境光学学报, 2022, 17(6): 598−612. doi: 10.3969/j.issn.1673-6141.2022.06.002
Hou M Y, Li Z Q, Xie Y S, et al. Research on spectral feature cloud detection method of directional polarimetric camera on Chinese satellite[J]. J Atmos Environ Opt, 2022, 17(6): 598−612. doi: 10.3969/j.issn.1673-6141.2022.06.002
[23] 程前, 高欣健, 高隽, 等. 基于邻域约束的大气偏振模式生成网络[J]. 光电工程, 2022, 49(6): 210423. doi: 10.12086/oee.2022.210423
Cheng Q, Gao X J, Gao J, et al. A generative method for atmospheric polarization modelling based on neighborhood constraint[J]. Opto-Electron Eng, 2022, 49(6): 210423. doi: 10.12086/oee.2022.210423
[24] Kastner R, Matai J, Neuendorffer S. Parallel programming for FPGAs[Z]. arXiv: 1805.03648, 2018. https://doi.org/10.48550/arXiv.1805.03648.
[25] Volder J E. The CORDIC trigonometric computing technique[J]. IRE Trans Electron Comput, 1959, EC-8(3): 330−334. doi: 10.1109/TEC.1959.5222693
-
点击扫一扫
图(7)
表(7)
计量
- 文章访问数: 328
- PDF下载数: 132
- 施引文献: 0
