多向前方交会与单隐层神经网络结合的近景前方交会法

doi:10.11947/j.AGCS.2020.20180600

测绘学报 ›› 2020, Vol. 49 ›› Issue (6): 736-745.doi: 10.11947/j.AGCS.2020.20180600

多向前方交会与单隐层神经网络结合的近景前方交会法

李佳田, 王聪聪, 阿晓荟, 晏玲, 朱志浩, 高鹏

昆明理工大学国土资源工程学院, 云南昆明 650093

收稿日期:2019-01-02 修回日期:2020-02-12 出版日期:2020-06-20 发布日期:2020-06-28
通讯作者: 王聪聪 E-mail:1083719493@qq.com
作者简介:李佳田(1975-),男,教授,博导,研究方向为数值最优化方法与机器场景理解。E-mail:ljtwcx@163.com
基金资助:
国家自然科学基金（41561082）

Method of close-range space intersection combining multi-image forward intersection with single hidden layer neural network

LI Jiatian, WANG Congcong, A Xiaohui, YAN Ling, ZHU Zhihao, GAO Peng

Faculty of Land Resource Engineering, Kunming University of Science and Technology, Kunming 650093, China

Received:2019-01-02 Revised:2020-02-12 Online:2020-06-20 Published:2020-06-28
Supported by:
The National Natural Science Foundation of China (No. 41561082)

摘要/Abstract

摘要： 针对三维坐标求解精度受非线性误差影响的问题，提出了一种多像前方交会与单隐层BP神经网络相结合的方法。基本过程为：①在样本点真实世界坐标已知的条件下，构建关于世界坐标的拉格朗日方程，对相机外参进行优化，以得到更高精度的三维坐标初值。②利用计算得到的三维坐标和真实三维坐标分别作为输入和输出参数对单隐层BP神经网络进行训练。③将三维坐标初值带入网络模型对其进行改正。试验结果表明：①相较于前方交会、稀疏光束法平差及典型神经网络方法，在试验装置环境视场内，本文方法解算精度较高，最大偏差为0.492 7 mm。②相较于典型神经网络方法，本文网络结构为3-6-3，结构简单且计算效率高。

关键词: 前方交会, 外方位参数, BP神经网络, 拉格朗日方程, 稀疏光束法平差

Abstract: Aiming at the problem that the three-dimensional coordinate solution precision is influenced by the non-linear error, a method of combining multi-image forward intersection with single hidden layer BP neural network is proposed in this paper. The steps are: ①In order to obtain the initial value of the three dimensional coordinates with higher accuracy, the external parameters of the camera are optimized by constructing the Lagrange equation about the world coordinates under the constraint of known real world coordinates of the sample point. ②The single hidden layer BP neural network is trained by using the calculated 3D coordinate and the real 3D coordinate as input and output parameters, respectively. ③The initial three-dimensional coordinate is corrected by brought into the model. Experiments show that: ①In the environmental field of view of the test device, the proposed method outperforms the space intersection, the sparse bundle adjustment and the other classic neural network methods, the maximum deviation is 0.492 7 mm. ②Compared with other classic neural network methods, the network structure of this paper is 3-6-3, the structure is simple and the calculation efficiency is high.

Key words: space intersection, exterior elements, BP neural network, Lagrange equation, sparse bundle adjustment

中图分类号:

P232

李佳田, 王聪聪, 阿晓荟, 晏玲, 朱志浩, 高鹏. 多向前方交会与单隐层神经网络结合的近景前方交会法[J]. 测绘学报, 2020, 49(6): 736-745.

LI Jiatian, WANG Congcong, A Xiaohui, YAN Ling, ZHU Zhihao, GAO Peng. Method of close-range space intersection combining multi-image forward intersection with single hidden layer neural network[J]. Acta Geodaetica et Cartographica Sinica, 2020, 49(6): 736-745.

参考文献 29

[1]	张剑清, 胡安文. 多基线摄影测量前方交会方法及精度分析[J]. 武汉大学学报(信息科学版), 2007, 32(10):847-851. ZHANG Jianqing, HU Anwen. Method and precision analysis of multi-baseline photogrammetry[J]. Geomatics and Information Science of Wuhan University, 2007, 32(10):847-851.
[2]	张祖勋, 杨春生, 张剑清, 等. 多基线-数字近景摄影测量[J]. 地理空间信息, 2007, 5(1):1-4. ZHANG Zuxun, YANG Chunsheng, ZHANG Jianqing, et al. Multi-baseline digital close-range photogrammetry[J]. Geospatial Information, 2007, 5(1):1-4.
[3]	袁修孝. 当代航空摄影测量加密的几种方法[J]. 武汉大学学报(信息科学版), 2007, 32(11):1001-1006. YUAN Xiuxiao. Modern methodologies for photogrammetric point determination[J]. Geomatics and Information Science of Wuhan University, 2007, 32(11):1001-1006.
[4]	单杰. 光束法平差简史与概要[J]. 武汉大学学报(信息科学版), 2018, 43(12):1797-1810. SHAN Jie. A brief history and essentials of bundle adjustment[J]. Geomatics and Information Science of Wuhan University, 2018, 43(12):1797-1810.
[5]	张剑清, 潘励, 王树根. 摄影测量学[M]. 2版. 武汉:武汉大学出版, 2009. ZHANG Jianqing, PAN Li, WANG Shugen. The principles of photogrammetry[M]. 2nd ed. Wuhan:Wuhan University Press, 2009.
[6]	隆昌宇, 邾继贵, 郭寅, 等. 基于非参数测量模型的摄影测量方法研究[J]. 光学学报, 2014, 34(12):1215004-1-1215004-9. LONG Changyu, ZHU Jigui, GUO Yin, et al. Study on close-range photogrammetry based on nonparameteric measurement model[J]. Acta Optica Sinica, 2014, 34(12):1215004-1-1215004-9.
[7]	李忠美, 边少锋, 翟勇. 多像空间前方交会的抗差总体最小二乘估计[J]. 测绘学报, 2017, 46(5):593-604. DOI:10.11947/j.AGCS.2017.20160081. LI Zhongmei, BIAN Shaofeng, QU Yong. Robust total least squares estimation of space intersection appropriate for multi-images[J]. Acta Geodaetica et Cartographica Sinica, 2017, 46(5):593-604. DOI:10.11947/j.AGCS.2017.20160081.
[8]	TRIGGS B, MCLAUCHLAN P F, HARTLEY R I, et al. Bundle adjustment-A modern synthesis[C]//Vision Algorithms:Theory and Practice. Berlin, Heidelberg:Springer, 2000:298-372.
[9]	王祥, 张永军, 黄山, 等. 旋转多基线摄影光束法平差法方程矩阵带宽优化[J]. 测绘学报, 2016, 45(2):170-177. DOI:10.11947/j.AGCS.2016.20150282. WANG Xiang, ZHANG Yongjun, HUANG Shan, et al. Bandwidth optimization of normal equation matrix in bundle block adjustment in multi-baseline rotational photography[J]. Acta Geodaetica et Cartographica Sinica, 2016, 45(2):170-177. DOI:10.11947/j.AGCS.2016.20150282.
[10]	徐振亮, 闫利, 马振玲, 等. 轴角描述的光束法平差新方法[J]. 武汉大学学报(信息科学版), 2015, 40(7):865-869, 876. XU Zhenliang, YAN Li, MA Zhenling, et al. A nevel bundle adjustment method based on the axis/angle expression[J]. Geomatics and Information Science of Wuhan University, 2015, 40(7):865-869, 876.
[11]	FRASER C S. Automatic camera calibration in close range photogrammetry[J]. Photogrammetric Engineering and Remote Sensing, 2013, 79(4):381-388.
[12]	许雄, 钟燕飞, 张良培, 等. 基于空间自相关BP神经网络的遥感影像亚像元定位[J]. 测绘学报, 2011, 40(3):307-311. XU Xiong, ZHONG Yanfei, ZHANG Liangpei, et al. A sub-pixel mapping algorithm based on BP neural network with spatial autocorrelation function for remote sensing imagery[J]. Acta Geodaetica et Cartographica Sinica, 2011, 40(3):307-311.
[13]	WU Bing, HAN Shaojun, XIAO Jin, et al. Error compensation based on BP neural network for airborne laser ranging[J]. Optics, 2016, 127(8):4083-4088.
[14]	DO Y. Application of neural networks for stereo-camera calibration[C]//International Joint Conference on Neural Networks. Washington, DC, USA:IEEE, 1999:2719-2722.
[15]	袁野, 欧宗瑛, 田中旭. 应用神经网络隐式视觉模型进行立体视觉的三维重建[J]. 计算机辅助设计与图形学学报, 2003, 15(3):293-296. YUAN Ye, OU Zongying, TIAN Zhongxu. 3D reconstruction of stereo vision using neural networks implicit vision model[J]. Journal of Computer-Aided Design & Computer Graphics, 2003, 15(3):293-296.
[16]	延和, 吴斌. 基于改进型神经网络的双目摄像机标定[J]. 西南科技大学学报, 2013, 28(4):66-70. YAN He, WU Bin. Improved neural network for binocular camera calibration[J]. Journal of Southwest University of Science and Technology, 2013, 28(4):66-70.
[17]	崔岸, 袁智, 王丹, 等. 基于改进Harris角点提取的摄像机神经网络标定技术[J]. 农业机械学报, 2009, 40(8):214-218. CUI An, YUAN Zhi, WANG Dan, et al. Neural network technique on camera calibration based on improved Harris corner extraction method[J]. Transactions of the Chinese Society for Agricultural Machinery, 2009, 40(8):214-218.
[18]	刘金颂, 原思聪, 江祥奎, 等. 基于Zernike矩和PSO算法的摄像机神经网络标定[J]. 光电子·激光, 2010, 21(9):1311-1314. LIU Jinsong, YUAN Sicong, JIANG Xiangkui, et al. Neural network on camera calibration based on Zernike moment and PSO algorithm[J]. Journal of Optoelectronics Laser, 2010, 21(9):1311-1314.
[19]	HU Z Y, WU F C. A note on the number of solutions of the noncoplanar P4P problem[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2002, 24(4):550-555.
[20]	胡占义, 雷成, 吴福朝. 关于P4P问题的一点讨论[J]. 自动化学报, 2001, 27(6):770-776. HU Zhanyi, LEI Cheng, WU Fuchao. A short note on P4P problem[J]. Acta Automatica Sinica, 2001, 27(6):770-776.
[21]	GONG Yuanzheng, MENG De, SEIBEL E J. Bound constrained bundle adjustment for reliable 3D reconstruction[J]. Optics Express, 2015, 23(8):10771-10785.
[22]	LI Yanyan, FAN Shiyue, SUN Yanbiao, et al. Bundle adjustment method using sparse BFGS solution[J]. Remote Sensing Letters, 2018, 9(8):789-798.
[23]	LOURAKIS M, ARGYROS A. The design and implementation of a generic sparse bundle adjustment software package based on the Levenberg-Marquardt algorithm[R]. Greece:Institute of Computer Science-Forth, 2004.
[24]	BASHEER I A, HAJMEER M. Artificial neural networks:fundamentals, computing, design, and application[J]. Journal of Microbiological Methods, 2000, 43(1):3-31.
[25]	DARKEN C, MOODY J. Towards faster stochastic gradient search[C]//Proceedings of the 4th International Conference on Neural Information Processing Systems. San Francisco, CA, United States:Morgan Kaufmann Publishers Inc., 1991:1009-1016.
[26]	ZHANG Zhengyou. A flexible new technique for camera calibration[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2000, 22(11):1330-1334.
[27]	AGARWAL S, SNAVELY N, SEITZ S M, et al. Bundle adjustment in the large[C]//Computer Vision-ECCV 2010. ECCV 2010. Lecture Notes in Computer Science, vol 6312. Heidelberg:Springer, 2010:29-42.
[28]	ZHOU Jun, XIA Tao, WANG Tingting, et al. A calibration method for binocular vision system based on CPSO-BP netural network[J]. Applied Mechanics and Materials, 2014, 490-491:1686-1691.
[29]	YAO Hengfeng, ZHANG Zhibin. Research of camera calibration based on genetic algorithm BP neural network[C]//Proceedings of IEEE International Conference on Information and Automation. Ningbo, China:IEEE, 2016:350-355.

多向前方交会与单隐层神经网络结合的近景前方交会法

Method of close-range space intersection combining multi-image forward intersection with single hidden layer neural network

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献 29

相关文章 4

编辑推荐

Metrics

本文评价

[1]	吕栋, 欧吉坤, 于胜文. 基于MEA-BP神经网络的卫星钟差预报[J]. 测绘学报, 2020, 49(8): 993-1003.
[2]	李忠美, 边少锋, 瞿勇. 多像空间前方交会的抗差总体最小二乘估计[J]. 测绘学报, 2017, 46(5): 593-604.
[3]	唐新明,张过,,祝小勇,潘红播,蒋永华,周平,王霞. 资源三号测绘卫星三线阵成像几何模型构建与精度初步验证[J]. 测绘学报, 2012, 41(2): 191-198.
[4]	刘汉丽周成虎朱阿兴李霖. 多子群遗传神经网络模型用于路口短时交通流量预测 [J]. 测绘学报, 2009, 38(4): 0-310.