Positioning algorithm for laser spot center based on BP neural network and genetic algorithm
-
摘要:
针对振动环境中传统光斑中心定位算法存在的处理时间长、精度低等问题，本文提出一种基于遗传算法优化BP神经网络的光斑定位方法。使用BP神经网络对光斑位置进行预测，并通过遗传算法对神经网络进行优化。构建BP神经网络模型，将使用质心、形心、高斯拟合等方法求出的光斑中心位置以及形心法求出的光斑半径作为输入，对光斑真实中心位置进行预测。并使用遗传算法优化神经网络的权值和阈值，以增强预测效果。实验过程中，通过对光学系统外加干扰模拟振动环境，采集数据用于神经网络训练和算法验证。实验结果表明，优化前后的标定测试迭代次数分别为55和29，平均误差分别为0.81像素和0.45像素。由本文结果可知，在遗传算法的优化下，神经网络算法的迭代速度和预测精度均有所提高。
Abstract:Aming at the problems of long processing time and low accuracy of the traditional laser spot center positioning algorithm used in a vibrating environment. We proposed a laser spot center positioning method based on a genetic algorithm optimized BP neural network. A BP neural network was applied to predict the spot center position and a genetic algorithm was applied to optimize the neural network. Based on the BP neural network, the gray weighted centroid method, centroid method, Gaussian fitting method were used to obtain the spot center position, and the centroid method was used to obtain the radius of laser spot, on the above basis, we predicted the actual center position of the spot. Genetic algorithms were used to optimize the weights and thresholds of neural networks to improve prediction accuracy. An experimental platform is established to simulate the vibration environment by applying perturbations to the optical system and the data is collected to train neural network and verify the algorithm. The experimental results show that the number of calibration test iterations before and after optimization is 55 and 29, and the average errors are 0.81 pixels and 0.45 pixels, respectively. Under the optimization of the genetic algorithm, the iteration speed and prediction accuracy of the neural network algorithm is improved.
-
Key words:
- genetic algorithm /
- BP neural network /
- image processing /
- laser spot center
-
图 1 遗传算法优化BP神经网络流程图
Figure 1. Flow chart of genetic algorithm combined with BP neural network
图 10 进行控制后光斑沿x轴的变化情况
Figure 10. The vibration curves along x-axis for the spot center under controlling
表 1 训练数据
Table 1. Training data
Grayscale
centeringcentroid Gaussian
fittingradius Actual
Coordinate(252.30,
305.11)(255.43,
309.27)(254.17,
307,91)7 (253.30,
307,11)
下载: 导出CSV
表 2 两种网络性能对比
Table 2. Performance comparison of the two neural networks
Neural Network BP GA-BP Number of iterations 55 29 Mean error/pixel 0.76 0.42
下载: 导出CSV
表 3 两种网络预测方法的性能测试结果
Table 3. Performance test results by the two neural networks
Neural Network BP GA-BP Number of iterations 47 32 Mean error/pixel 1.21 0.73
下载: 导出CSV
-
[1] 徐抒岩, 张旭升, 范阔, 等. 大型光学系统波前检测中气流扰动的抑制[J]. 光学 精密工程,2020,28(1):80-89. doi: 10.3788/OPE.20202801.0080XU SH Y, ZHANG X SH, FAN K, et al. Suppression of airflow turbulence in wavefront measurement for large-aperture optical systems[J]. Optics and Precision Engineering, 2020, 28(1): 80-89. (in Chinese) doi: 10.3788/OPE.20202801.0080 [2] 周睿, 李新阳, 沈锋, 等. 基于两级高速倾斜镜闭环控制的光束稳定技术研究[J]. 光学学报,2016,36(12):1214002. doi: 10.3788/AOS201636.1214002ZHOU R, LI X Y, SHEN F, et al. Laser beam stabilizing system based on close loop control of two fast steering mirrors in series[J]. Acta Optica Sinica, 2016, 36(12): 1214002. (in Chinese) doi: 10.3788/AOS201636.1214002 [3] 周睿. 自适应光学实时信号处理及优化控制技术研究[D]. 成都: 中国科学院光电技术研究所, 2017.ZHOU R. Real time signal processing and optimal control technology on adaptive optics systems[D]. Chengdu: Institute of Optics and Electronics, Chinese Academy of Sciences, 2017. (in Chinese) [4] GANESAN A R, ARULMOZHIVARMAN P, MOHAN D, et al. Laser beam steering control system for free-space line of sight optical communication[J]. IETE Journal of Research, 2006, 52(6): 417-424. doi: 10.1080/03772063.2006.11416482 [5] ARNON S, KOPEIKA N S, KEDAR D, et al. Performance limitation of laser satellite communication due to vibrations and atmospheric turbulence: down‐link scenario[J]. International Journal of Satellite Communications and Networking, 2003, 21(6): 561-573. doi: 10.1002/sat.769 [6] 杨滨赫, 蔡引娣, 文志祥, 等. 长距离激光测量中光束漂移的自动补偿[J]. 光学 精密工程,2020,28(11):2393-2402. doi: 10.37188/OPE.20202811.2393YANG B H, CAI Y D, WEN ZH X, et al. Automatic compensation method for beam drift in long-distance laser measurement[J]. Optics and Precision Engineering, 2020, 28(11): 2393-2402. (in Chinese) doi: 10.37188/OPE.20202811.2393 [7] 孙晶, 黄普明, 幺周石. 大气湍流与平台微振动影响下的星地激光通信性能[J]. 激光与光电子学进展,2021,58(3):0301003.SUN J, HUANG P M, YAO ZH SH. Performance of satellite-to-ground laser communications under the influence of atmospheric turbulence and platform micro-vibration[J]. Laser &Optoelectronics Progress, 2021, 58(3): 0301003. (in Chinese) [8] 李海廷, 胡鑫, 曾双, 等. 大气湍流对半主动激光制导中光斑检测精度的影响[J]. 兵工学报,2021,42(2):297-307. doi: 10.3969/j.issn.1000-1093.2021.02.008LI H T, HU X, ZENG SH, et al. Influence of atmospheric turbulence on detection accuracy of laser spot[J]. Acta Armamentarii, 2021, 42(2): 297-307. (in Chinese) doi: 10.3969/j.issn.1000-1093.2021.02.008 [9] ZHU W, RUI X T. Adaptive control of a piezo-actuated steering mirror to restrain laser-beam jitter[J]. IEEE Transactions on Industrial Electronics, 2019, 66(10): 7873-7881. doi: 10.1109/TIE.2018.2885731 [10] 蒋佳雯, 康杰虎, 吴斌. 激光光斑中心高精度定位补偿算法研究[J]. 激光与光电子学进展,2021,58(14):1412002.JIANG J W, KANG J H, WU B. High precision positioning and compensation algorithm for laser spot center[J]. Laser &Optoelectronics Progress, 2021, 58(14): 1412002. (in Chinese) [11] 任守峰, 唐新明, 祝小勇, 等. 一种高分七号卫星激光足印光斑质心的高精度提取算法[J]. 光学学报,2021,41(10):1012001. doi: 10.3788/AOS202141.1012001REN SH F, TANG X M, ZHU X Y, et al. A high-precision extraction algorithm for centroid of laser footprint spot of GF-7 satellite[J]. Acta Optica Sinica, 2021, 41(10): 1012001. (in Chinese) doi: 10.3788/AOS202141.1012001 [12] 伍煜, 龙小祥, 杨坚, 等. 高分七号卫星足印相机激光光斑中心定位方法研究[J]. 光学学报,2021,41(17):1728001. doi: 10.3788/AOS202141.1728001WU Y, LONG X X, YANG J, et al. Laser spot center positioning method in gaofen-7 footprint camera[J]. Acta Optica Sinica, 2021, 41(17): 1728001. (in Chinese) doi: 10.3788/AOS202141.1728001 [13] 王丽丽, 胡中文, 季杭馨. 基于高斯拟合的激光光斑中心定位算法[J]. 应用光学,2012,33(5):985-990.WANG L L, HU ZH W, JI H X. Laser spot center location algorithm based on Gaussian fitting[J]. Journal of Applied Optics, 2012, 33(5): 985-990. (in Chinese) [14] 董书舟, 强佳, 舒嵘. 基于Hough变换的天基激光测距轨迹提取算法[J]. 半导体光电,2021,42(3):430-436.DONG SH ZH, QIANG J, SHU R. Space-based laser ranging trajectory extraction algorithm based on Hough transform[J]. Semiconductor Optoelectronics, 2021, 42(3): 430-436. (in Chinese) [15] 潘登, 李艳丽, 高东, 等. 基于椭圆拟合的多光斑/重叠光斑中心提取方法[J]. 光学学报,2020,40(14):1410001. doi: 10.3788/AOS202040.1410001PAN D, LI Y L, GAO D, et al. Center extraction method of multiple and overlapping faculae based on ellipse fitting[J]. Acta Optica Sinica, 2020, 40(14): 1410001. (in Chinese) doi: 10.3788/AOS202040.1410001 [16] 安哲. 大气湍流下的光斑定位方法研究[D]. 哈尔滨: 哈尔滨工业大学, 2019.AN ZH. Study on the method of spot location under atmospheric turbulence[D]. Harbin: Harbin Institute of Technology, 2019. (in Chinese) [17] 刘岳飞. 基于光流法的星间光通信信标光识别跟踪算法研究[D]. 哈尔滨: 哈尔滨工业大学, 2017.LIU Y F. Study on the reconition and tracking algorithm of the beacon for intersatellite laser communications based on optical flow method[D]. Harbin: Harbin Institute of Technology, 2017. (in Chinese) [18] 曹瑜彬. 基于深度学习的大气光场传输和光斑质心位置分析及预测[D]. 哈尔滨: 哈尔滨工业大学, 2020.CAO Y B. Deep learning based prediction and analysis for light fields propagating through atmosphere and optical image centroid position[D]. Harbin: Harbin Institute of Technology, 2020. (in Chinese) [19] 梁慧慧, 何秋生, 贾伟振, 等. 一种多特征融合的目标跟踪算法[J]. 液晶与显示,2020,35(6):583-594.LIANG H H, HE Q SH, JIA W ZH, et al. Multi-feature fusion target tracking algorithm[J]. Chinese Journal of Liquid Crystals and Displays, 2020, 35(6): 583-594. (in Chinese) [20] 兰旭婷, 郭中华, 李昌昊. 基于注意力与特征融合的光学遥感图像飞机目标检测[J]. 液晶与显示,2021,36(11):1506-1515. doi: 10.37188/CJLCD.2021-0088LAN X T, GUO ZH H, LI CH H. Attention and feature fusion for aircraft target detection in optical remote sensing images[J]. Chinese Journal of Liquid Crystals and Displays, 2021, 36(11): 1506-1515. (in Chinese) doi: 10.37188/CJLCD.2021-0088 [21] 张博, 龙慧, 刘刚. 基于特征约束与光流场模型的多通道视频目标跟踪算法[J]. 液晶与显示,2021,36(11):1554-1564. doi: 10.37188/CJLCD.2021-0113ZHANG B, LONG H, LIU G. Multi-channel video target tracking algorithm based on feature constraint and optical flow field model[J]. Chinese Journal of Liquid Crystals and Displays, 2021, 36(11): 1554-1564. (in Chinese) doi: 10.37188/CJLCD.2021-0113 [22] 迟书凯, 叶旋, 高翔, 等. 基于编码标记点的高精度运动估计[J]. 光学 精密工程,2021,29(7):1720-1730. doi: 10.37188/OPE.20212907.1720CHI SH K, YE X, GAO X, et al. Coded marker-based high-accuracy motion estimation[J]. Optics and Precision Engineering, 2021, 29(7): 1720-1730. (in Chinese) doi: 10.37188/OPE.20212907.1720 [23] 王向周, 陈冬清, 郑戍华, 等. 基于改进遗传优化算法的线阵相机标定方法[J]. 北京理工大学学报,2020,40(8):861-866. doi: 10.15918/j.tbit1001-0645.2019.239WANG X ZH, CHEN D Q, ZHENG SH H, et al. A linear array camera calibration method based on improved genetic optimization algorithm[J]. Transactions of Beijing Institute of Technology, 2020, 40(8): 861-866. (in Chinese) doi: 10.15918/j.tbit1001-0645.2019.239 -
下载:
计量
- 文章访问数: 816
- HTML全文浏览量: 443
- PDF下载量: 361
- 被引次数: 0
