华南理工大学学报(自然科学版) ›› 2012, Vol. 40 ›› Issue (3): 100-105.

• 电子、通信与自动控制 • 上一篇    下一篇

基于多面化分解模型的目标信息获取优化技术

闫奕名 张晔 高凤娇   

  1. 哈尔滨工业大学 图像与信息技术研究所,黑龙江 哈尔滨 150001
  • 收稿日期:2011-08-14 修回日期:2011-12-08 出版日期:2012-03-25 发布日期:2012-02-01
  • 通信作者: 闫奕名(1982-) ,男,博士生,主要从事遥感立体目标识别研究. E-mail:dragon_no.2@hotmail.com
  • 作者简介:闫奕名(1982-) ,男,博士生,主要从事遥感立体目标识别研究.
  • 基金资助:

    国家“863”计划项目( 2010AA0748)

Optimization of Target Information Acquisition Technology Based on Multi-Side Decomposition Model

Yan Yi-ming  Zhang Ye  Gao Feng-jiao   

  1.  Institute of Image and Information Technology,Harbin Institute of Technology,Harbin 150001,Heilongjiang,China
  • Received:2011-08-14 Revised:2011-12-08 Online:2012-03-25 Published:2012-02-01
  • Contact: 闫奕名(1982-) ,男,博士生,主要从事遥感立体目标识别研究. E-mail:dragon_no.2@hotmail.com
  • About author:闫奕名(1982-) ,男,博士生,主要从事遥感立体目标识别研究.
  • Supported by:

    国家“863”计划项目( 2010AA0748)

摘要: 为了能够通过观测角度设置的优化选择,实现目标信息获取技术的优化,以最少的观测点数获取最大化的信息量,提出了一种基于多面化分解的目标信息量描述模型,用于分析图像传感器在不同观测角度下的目标信息获取量. 该模型以各面信息获取量组合向量的形式来描述多面化分解的目标在某观测角度下的信息获取量,采用改进的蜂群算法来搜索最佳的观测角度,实现最大化的信息获取. 针对无人机多时相观测时的目标信息获取优化问题,采用文中方法进行观测角规划仿真实验. 结果表明: 经文中方法优化后,信息获取量平均提高了58.5%; 文中方法可对有一定观测角约束条件的情况进行高效规划,实现最大化的信息获取.

关键词: 信息分析, 目标多面化分解, 蜂群算法, 无人机

Abstract:

In order to optimize the target information acquisition technology by prudently selecting observing angles and to collect maximum information with the fewest observing points,a multi-side decomposition-based information content description model is proposed,which is used to measure the target information collected by optical image sensors at different observing angles. The main idea of this model is that,to estimate the information collected at different observing angles,the target is decomposed into multiple sides,and its information is described by the combined vector of the information collected from all sides. Moreover,the modified bee colony algorithm is utilized to search the optimal observing angle and to further obtain the maximum information content. The proposed method is then applied to the planning simulation of observing angles in the target information acquisition of an unmanned aerial vehicle with multi-temporal and relative observation. Experimental results show that the proposed optimization method improves the average information acquisition by 58.5%,and that it helps to perform efficient planning with maximum information acquisition when observing angles are limited.

Key words: information analysis, multi-side target decomposition, bee colony algorithm, unmanned aerial vehicles

中图分类号: