摘要
目标识别是红外成象GIF中的关键技术之一,利用神经网络可以完成目标识别的任务。在获得目标红外灰度图象傅立叶描述子特征的基础上,对BP神经网络、径向基函数神经网络和学习矢量量化神经网络在目标类型识别中的应用进行了研究。通过网络的设计及算法的仿真结果,比较了这几种神经网络的在目标识别方面的优缺点。
Target recognition is one of the key techniques in infrared imaging GIF technology, and target recognition can be achieved by using neural network. On the basis of introducing how to acquire the Fourier descriptors according to target infrared images, this text studies the applications of BP neural network, RBF neural network and LVQ neural network in target recognition. Target recognizing ability of the three sorts of neural networks is compared through network simulation.
出处
《探测与控制学报》
CSCD
北大核心
2005年第1期9-12,53,共5页
Journal of Detection & Control
关键词
傅立叶描述子
神经网络
目标识别
fourier descriptors
neural network
Target recognition
