计算机与现代化 ›› 2010, Vol. 1 ›› Issue (5): 63-65.doi: 10.3969/j.issn.1006-2475.2010.05.019

• 图像处理 • 上一篇    下一篇

运动目标识别方法与仿真研究

朱参世,安 利   

  1. 空军工程大学工程学院,陕西 西安 710038
  • 收稿日期:2010-02-25 修回日期:1900-01-01 出版日期:2010-05-10 发布日期:2010-05-10

Moving Object Recognition Method and Simulation Study

ZHU Can-shi, AN Li   

  1. Engineering College, Air Force Engineering University, Xi'an 710038, China
  • Received:2010-02-25 Revised:1900-01-01 Online:2010-05-10 Published:2010-05-10

摘要: 运动目标识别与跟踪在军事和工业领域具有广泛的应用前景。典型的跟踪方法有相关匹配算法、光流算法、基于运动轮廓算法等。典型的识别方法有:模板匹配目标识别法、基于相似性度量识别法、基于遗传算法目标识别算法等。本文提出了一种基于神经网络的识别与跟踪方法,并研究了运动目标识别与跟踪仿真模型和实现方法。

关键词: 目标识别, 目标跟踪, 仿真

Abstract: The moving target recognition and tracking in the military and industry has a broad application prospects. The typical tracking methods are correlation matching algorithm, optical flow algorithm, motion-based contour algorithm, and so on. The typical identification methods include: template matching target recognition method, similarity measure based identification method, genetic algorithm based target recognition algorithm, and so on. This paper presents a neural network based identification and tracking method, and studies the identification and tracking of moving target simulation model and the realization method.

Key words: target identification, target tracking, simulation

中图分类号: