多Agent系统在遥感图像目标识别系统中的应用研究  

Research on Application of based on Multi-agent to Remote Sensing Image Object Recognition

作　　者：章雅娟[1] 

机构地区：[1]苏州大学

出　　处：《电脑开发与应用》2007年第11期20-23,共4页Computer Development & Applications

摘　　要：从遥感图像分类方法的基本原理入手,介绍了多A gent系统因其智能性和并行计算能力在遥感图像目标识别领域的应用,并将模糊数学、神经网络以及统计学理论与其有机地结合起来,发挥综合系统的整体优势去解决问题。这对今后数据融合技术在图像处理中的研究和应用有很大的意义。Starting with basic principles of remote sensing image classification principle, Multi-agent system is introduced into remote sensing image object recognition for its intelligence and parallel computing, and combining fuzzy mathematics, neural network, statistical learning theory with Multi-agent system, exerted the whole advantage of integrated system to solve the question. The design of architecture plays an important role in the research and applications of data fusion technology in image disposal in the future.

关 键 词：遥感图像 目标识别 多AGENT系统 分类器 纹理检测 数据融合 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]