E-mail Alert
RSS
-
摘要:
基于CMOS传感器的高集成小型相机在航天器上得到了越来越广泛的应用。空间相机能够完成航天器关键动作记录、对地遥感、近地天体观测等任务，并且具有体积小、重量轻、智能化的特点。为了实现良好的成像效果，适应于空间环境特点的自动曝光技术不可或缺。本文针对空间环境的特殊性及空间任务的不可逆性，提出了一种快速自适应曝光算法。该算法以能量分析为基础，进行目标与背景分离，针对目标进行加权统计，根据图像的统计结果，采用最速查表法，计算获得最佳曝光时间。设置双重目标调整范围，使得自动曝光算法收敛性较好。实验结果表明，该算法能够快速稳定地获得最佳曝光时间，曝光收敛速度快，稳定性高，资源占用少，非常适合空间场景探测。相关算法已成功应用于多个在轨型号。
Abstract:Highly integrated miniature cameras based on CMOS sensors have been used more and more widely in spacecraft. Space camera can accomplish the key operation record, remote sensing and near celestial observation of spacecraft, and has the characteristics of small volume, light weight and intelligent. In order to achieve a good imaging effect, automatic exposure technology which is suitable for the space environment is indispensable. In this paper, a fast adaptive exposure algorithm is proposed for the particularity of space environment and the irreversibility of spatial tasks. The algorithm is based on energy analysis, weighting statistics on the target according to the statistics of the image, and using the lookup table method to calculate the optimal exposure time. The double objective adjustment range is set, so that the automatic exposure algorithm converges better. The experimental results show that the algorithm can obtain the best exposure time quickly and stably. The speed of exposure is fast and the stability is high. It is very suitable for space scene detection. The correlation algorithm has been successfully applied to multiple on orbit models.
-
Key words:
- space camera /
- remote sensing /
- automatic exposure
-
Overview: The automatic exposure technology controls the imaging brightness by adjusting the exposure time of thecamera, so that the image is suitable for visual observation. In space field, space camera based on CMOS image sensorhas been widely used for its high integration, small volume and light weight. It plays a very important role in spacecraftkey action records, earth remote sensing and near earth observation. Because of the particularity of space environment, the irreversibility of space tasks and the lag of manual intervention, the camera can be intelligent and autonomous, which is very important. The camera can quickly and automatically adjust the imaging parameters and get the best imaging results for the first time. In view of the particularity of space environment and the irreversibility of space tasks, afast adaptive exposure algorithm is proposed, which can separate the target from the background and adjust the exposure time quickly and steadily. It is very suitable for deep space exploration and remote sensing. First of all, taking theremote sensing image as an example, the imaging link simulation is carried out. Special software is used to analyze theimaging link and get the target radiance. On the basis of the simulation analysis of the imaging link, according to theimaging characteristics of the lens and the sensitivity of the image sensor, the energy of the target to reach the imagesensor and the number of electrons converted from the lens are calculated to estimate the exposure time range, thussetting the initial exposure value. The target and background are separated by simple histogram distribution statistics.The different weights are applied to the target and the background. The luminance characteristics of the image are calculated by the weighted statistics. The exposure time is obtained by the look-up table method according to the result ofthe luminance characteristic. In the process of adjusting the exposure time, a dual target adjustment range is set up tomake the auto exposure algorithm converge well. The algorithm can be conveniently implemented on FPGA with lessresource occupancy and no complex operation. The algorithm is fast in computing speed and large in throughput, andcan be easily transplanted on various platforms. The experimental results show that the algorithm can quickly and stablyobtain the best exposure time, fast convergence speed, high stability, and less resource occupancy, which is very suitablefor space scene detection. The correlation algorithm has been successfully applied to multiple on orbit models, and alarge number of effective images have been obtained.
-
-
图 5 拍摄图像背景识别区域。(a)相机拍摄图像；(b) DN < 16暗背景区域；(c) 16≤DN < 32背景过渡区域
Figure 5. The background recognition area of the image. (a) The image taken by a camera; (b) DN < 16 dark background area; (c) 16 < DN < 32 background transition region
图 12 相机拍摄的特殊场景图像(窗外场景)
Figure 12. A special scene image taken by a camera (outside scene)
图 13 相机拍摄特殊场景直方图。(a)全图像直方图；(b)剔除背景后的直方图
Figure 13. The histogram of a special scene. (a) Full image histogram; (b) Remove background histogram
图 14 自动曝光在轨测试图像。(a)嫦娥三号降落相机在轨成像效果；(b)探月三期再入返回飞行试验器技术试验相机在轨成像效果；(c)珠海一号视频相机在轨自动曝光成像效果
Figure 14. An on-track image taken by an automatic exposure. (a) Imaging effect of Chang'e three landing camera; (b) Imaging effect of technical test camera in orbit; (c) Automatic exposure imaging effect of video camera on orbit
表 1 相机入瞳辐亮度计算条件
Table 1. Calculation conditions for the radiance of the camera pupil
项目 要求 轨道高度/km 1100 时间 6月23日 太阳高度角/(°) 20~70 大气模型 中纬度夏季 气溶胶模型 海面 能见度/km 23 地物反射率/% 5~70 谱段宽度/μm 0.5~0.9 
下载: 导出CSV
表 2 遥感成像辐亮度计算结果
Table 2. Calculation results of radiance of remote sensing imaging
太阳高度角/(°) 地面反射率(谱段范围：0.5 μm~0.9 μm) 0.05 0.10 0.20 0.30 0.40 0.50 0.55 0.60 0.65 0.70 20 13.11 18.36 29.02 39.89 50.98 62.30 68.05 73.86 79.73 85.66 30 17.08 25.54 42.71 60.22 78.09 96.33 105.59 114.95 124.41 133.96 40 20.83 32.23 55.36 78.96 103.04 127.61 140.09 152.70 165.45 178.32 50 24.88 38.82 67.11 95.96 125.41 155.47 170.73 186.15 201.73 217.48 60 29.47 45.47 77.93 111.05 144.85 179.34 196.86 214.56 232.45 250.53 70 31.98 49.49 85.03 121.28 158.28 196.04 215.21 234.59 254.17 273.96
下载: 导出CSV
表 3 相机成像所需的曝光时间
Table 3. Exposure time required for camera imaging
ms 太阳高度角/(°) 太阳高度角/(°) 0.05 0.10 0.20 0.30 0.40 0.50 0.55 0.60 0.65 0.70 20 7.90 5.64 3.57 2.59 2.03 1.67 1.53 1.41 1.30 1.21 30 6.07 4.05 2.43 1.72 1.33 1.08 0.98 0.90 0.83 0.77 40 4.97 3.21 1.87 1.31 1.00 0.81 0.74 0.68 0.62 0.58 50 4.16 2.67 1.55 1.08 0.83 0.66 0.61 0.56 0.51 0.48 60 3.51 2.28 1.33 0.94 0.72 0.58 0.52 0.48 0.45 0.41 70 3.24 2.09 1.22 0.85 0.65 0.52 0.48 0.45 0.40 0.38
下载: 导出CSV
-
[1] 岳涛, 张宏伟, 黄长宁, 等. "嫦娥二号"卫星CMOS相机技术及应用[J].航天返回与遥感, 2011, 32(2): 12–17. doi: 10.3969/j.issn.1009-8518.2011.02.002
Yue T, Zhang H W, Huang C N, et al. The application of Chang'E-2 CMOS camera technologies[J]. Spacecraft Recovery & Remote Sensing, 2011, 32(2): 12–17. doi: 10.3969/j.issn.1009-8518.2011.02.002
[2] 顾征, 杨孟飞, 薛博, 等.航天器动作状态的可视化遥测方法研究[J].载人航天, 2017, 23(2): 185‒190. doi: 10.3969/j.issn.1674-5825.2017.02.008
Gu Z, Yang M F, Xue B, et al. Research on visual telemetry method for spacecraft motion status[J]. Manned Spaceflight, 2017, 23(2): 185‒190. doi: 10.3969/j.issn.1674-5825.2017.02.008
[3] 杨作廷.基于月球探测全景相机自动曝光技术的研究[D].西安: 中国科学院西安光学精密机械研究所, 2012: 12–19.
Yang Z T. The auto exposure technology of the panoramic camera on lunar exploration[D]. Xi'an: Xi'an Institute of Optics and Precision Mechanics Chinese Academy of Science, 2012: 12–19.
[4] 张振华.一个大面阵CMOS图像传感器高速航天相机的设计与基于FPGA的实现[D].北京: 北京大学, 2014: 11–19.
[5] 陈绕青, 曹国, 毛志红.一种空间面阵CCD成像的曝光时间计算方法[J].计算机工程, 2012, 38(12): 1–4. doi: 10.3969/j.issn.1000-3428.2012.12.001
Chen R Q, Cao G, Mao Z H. Computation method of exposure time for space array CCD imaging[J]. Computer Engineering, 2012, 38(12): 1–4. doi: 10.3969/j.issn.1000-3428.2012.12.001
[6] 李宪圣, 任建岳, 任建伟, 等.空间相机在轨成像模式的建立[J].光学 精密工程, 2015, 23(7): 1852–1858. doi: 10.3788/OPE.20152307.1852
Li X S, Ren J Y, Ren J W, et al. Establishment of image model for on-board space camera[J]. Optics and Precision Engineering, 2015, 23(7): 1852–1858. doi: 10.3788/OPE.20152307.1852
[7] 李志勇, 杨校军.关于遥感卫星TDICCD相机动态范围设计的思考[J].航天返回与遥感, 2011, 32(1): 24–27. doi: 10.3969/j.issn.1009-8518.2011.01.005
Li Z Y, Yang X J. Consideration on designing of dynamic range for TDICCD camera on remote sensing satellites[J]. Spacecraft Recovery & Remote Sensing, 2011, 32(1): 24–27. doi: 10.3969/j.issn.1009-8518.2011.01.005
[8] 侯幸林, 罗海波, 周培培.基于局部信息熵最大的多曝光控制方法[J].红外与激光工程, 2017, 46(7): 263–269. doi: 10.3788/IRLA201746.0726001
Hou X L, Luo H B, Zhou P P. Multi-exposure control method based on maximum local information entropy[J]. Infrared and Laser Engineering, 2017, 46(7): 263–269. doi: 10.3788/IRLA201746.0726001
[9] 马泽龙, 高慧斌, 余毅, 等.采用图像直方图特征函数的高速相机自动曝光方法[J].光学 精密工程, 2017, 25(4): 1026–1035. doi: 10.3788/OPE.20172504.1026
Ma Z L, Gao H B, Yu Y, et al. Auto exposure control for high frame rate camera using image histogram feature function[J]. Optics and Precision Engineering, 2017, 25(4): 1026–1035. doi: 10.3788/OPE.20172504.1026
[10] 杨海涛, 常义林, 王静, 等.一种基于亮度直方图的自动曝光控制方法[J].光学学报, 2007, 27(5): 841–847. doi: 10.3321/j.issn:0253-2239.2007.05.016
Yang H T, Chang Y L, Wang J, et al. A new automatic exposure algorithm for video cameras using luminance histogram[J]. Acta Optica Sinica, 2007, 27(5): 841–847. doi: 10.3321/j.issn:0253-2239.2007.05.016
[11] 梁佳毅.高性能数码相机自动曝光算法研究与实现[D].上海: 复旦大学, 2008: 51–68.
[12] 游明琦, 刘勇.一种自动曝光实现方法: 200410102536.0[P]. 2008-01-09.
You M Q, Liu Y. Method for implementing automatic exposure: 200410102536.0[P]. 2008-01-09.
-
点击扫一扫
图(14)
表(3)
计量
- 文章访问数:
- PDF下载数:
- 施引文献: 0
