Please wait a minute...
浙江大学学报(工学版)
J4  2010, Vol. 44 Issue (12): 2263-2268    DOI: 10.3785/j.issn.1008-973X.2010.12.006
自动化技术、计算机技术     
稳像中基于BP神经网络的颤振预测及改进
董文德, 徐之海, 李奇, 郑珍珍, 冯华君
浙江大学 现代光学仪器国家重点实验室, 浙江 杭州 310027
Vibration forecasting using BP neural network for
image stabilization and an improving method
DONG Wen-de, XU Zhi-hai, LI Qi, ZHENG Zhen-zhen, FENG Hua-jun
State Key Laboratory of Optical Instrumentation, Zhejiang University, Hangzhou 310027, China
 全文: PDF  HTML
摘要：

为了解决航拍过程中的图像抖动问题,研究了机载相机的颤振规律；提出在稳像过程中,利用BP (Back Propagation)神经网络的函数逼近功能对相机颤振规律进行模拟,预测相机颤振矢量的方法；针对单个BP神经网络稳定性较差且精度较低的问题,提出在预测网络上增加一个误差校正网络以提高预测精度的方法.该方法使用误差校正网络对预测网络输出的结果进行二次预测、补偿,提高了网络系统的稳定性和计算精度.仿真实验表明：在训练样本相同的情况下,预测网络和误差校正网络相结合的方法能够对相机颤振矢量进行高精度预测,且运算速度较快,满足了机载相机实时稳像的需求.
Abstract:

The vibration characteristic of airborne camera was studied to solve the image vibration in aerial photography. A method based on the ability of function approximation of BP neural network to simulate the vibration characteristic of airborne camera and predict the vibration displacement vectors during image stabilization was proposed. The inherent defects of BP neural network are its poor stability and low precision which cannot be accepted in the application of image stabilization. To overcome these problems, a new method combining two networks named prediction network and error correction network was proposed. The later network performs further prediction and compensation on the outputs of the former one, and thus optimizes the property of the network system. Experimental results show that with the same training samples, the combined network system is more stable and the outputs are of higher precision than that of a single network, and the computing is also fast, all of which meet the demands of realtime image stabilization in aerial photography.
出版日期: 2010-12-01
:  TP 389.1  
基金资助:

国家“973”重点基础研究发展规划资助项目（2009CB724006）；浙江省重大科技专项资金资助项目（2008C16018）;中国航天科技集团公司航天科技创新基金资助项目.
通讯作者: 冯华君,男,教授,博导.     E-mail: fenghj@zju.edu.cn
作者简介: 董文德（1984—）,男,山西大同人,博士生,从事空间稳像、图像复原技术研究.E-mail: kkll321@zju.edu.cn
服务　　
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章　　

引用本文:

董文德, 徐之海, 李奇, 郑珍珍, 冯华君. 稳像中基于BP神经网络的颤振预测及改进[J]. J4, 2010, 44(12): 2263-2268.

DONG Wen-de, XU Zhi-hai, LI Qi, ZHENG Zhen-zhen, FENG Hua-jun. Vibration forecasting using BP neural network for
image stabilization and an improving method. J4, 2010, 44(12): 2263-2268.

链接本文:

http://www.zjujournals.com/eng/CN/10.3785/j.issn.1008-973X.2010.12.006    或    http://www.zjujournals.com/eng/CN/Y2010/V44/I12/2263

［1］ 徐丽娜.神经网络控制［M］.哈尔滨：哈尔滨工业大学出版社,1999:6-18.
［2］ CONNOR J T, MARTIN R D, ATLAS L E, et al. Recurrent neural networks and robust time series prediction［J］. IEEE Transaction on Neural Networks,1994,5(2): 240-254.
［3］ 马炼,王银堂,张闻胜.神经网络技术在水文系列中长期预报中的应用［J］.水利水电技术,2002,33(2): 5-7.
MA Lian, WANG Yintang, ZHANG Wensheng. Long term hydrological forecasting with neural networks［J］. Water Resources and Hydropower Engineering, 2002,33(2): 5-7.
［4］ 胡金滨,唐旭清.人工神经网络的BP算法及其应用［J］.信息技术,2004,28(4): 1-4.
HU Jinbin, TANG Xuqing. BP algorithm and its application in artificial neural network［J］. Information Technology, 2004,28(4): 1-4.
［5］ 王维,张英堂.BP神经网络进行时间序列预测的不足及改进［J］.计算机工程与设计,2007,28(21):5292-5294.
WANG Wei, ZHANG Yingtang. Analysis and improving way of BP ANN in prediction time series data［J］. Computer Engineering and Design, 2007,28(21): 5292-5294.
［6］ 赵雪红,张来斌,樊建春.基于组合遗传神经网络的磨损趋势预测［J］.润滑与密封,2005（5）: 40-42.
ZHAO Xuehong, ZHANG Laibin, FAN Jianchun. Prediction of wear trend using combined BP neural network based on genetic algorithm［J］. Lubrication Engineering, 2005（5）: 40-42.
No related articles found!