CAJ | 학술논문

随着遥感技术的发展，将会有更多不同时间、空间分辨率的遥感图像数据应用于各个领域的工程中，遥感图像的模式识别便成为当前遥感图像应用研究领域中一个很重要的研究方向。人工神经网络（Artificial Network Networks，ANN）是一种模仿动物神经网络行为特征，进行分布式并行信息处理的算法数学模型。这种网络依靠系统的复杂程度，通过调整内部大量节点之间相互连接的关系，从而达到处理信息的目的。基于神经网络的遥感图像识别是遥感图像处理领域这几年的研究重点。本文首先介绍了人工神经网络和模式识别算法的基本理论知识，然后利用人工神经网络在Matlab工具箱环境中进行高空间分辨率的遥感图像识别。
수착요감기술적발전，장회유경다불동시간、공간분변솔적요감도상수거응용우각개영역적공정중，요감도상적모식식별편성위당전요감도상응용연구영역중일개흔중요적연구방향。인공신경망락（Artificial Network Networks，ANN）시일충모방동물신경망락행위특정，진행분포식병행신식처리적산법수학모형。저충망락의고계통적복잡정도，통과조정내부대량절점지간상호련접적관계，종이체도처리신식적목적。기우신경망락적요감도상식별시요감도상처리영역저궤년적연구중점。본문수선개소료인공신경망락화모식식별산법적기본이론지식，연후이용인공신경망락재Matlab공구상배경중진행고공간분변솔적요감도상식별。
With the development of the remote sensing technology, more and more remote sensing image data which are different in the time and space frequency will be applied to various fields of engineering. The pattern recognition of remote sensing imagery is becoming a very important research direction in the remote sensing image processing areas. Artificial Neural Networks （ANN） is a neural network to imitate animal behavior,which is the distributed parallel processing algorithms mathematical model. This network relies on the complexity of the system, by adjusting the internal connection between a large number of nodes,in order to achieve the purpose of dealing with information. Remote sensing based on neural network image recognition is the focus of the remote sensing image processing research in these years. First of all,this thesis introduces the basic theory of artificial neural networks and the pattern recognition,and uses the artificial neural network toolbox in the Matlab environment for the high spatial resolution remote sensing image recognition.