| 基于邻域统计特性的概率神经网络及其在自动目标识别中的应用 |
| |
| 引用本文： | 刘天溶,沈定刚.基于邻域统计特性的概率神经网络及其在自动目标识别中的应用[J].红外与毫米波学报,1995,14(1):52-58. |
| |
| 作者姓名： | 刘天溶 沈定刚 |
| |
| 作者单位： | 上海交通大学光纤技术研究所 |
| |
| 基金项目： | 国家攀登计划认知科学（神经网络）重大关键项目 |
| |
| 摘 要： | 在概率神经的一种改进模型－ＦＤＯ网络的基础上，提出在设计网络收敛域时进一步考虑每一像素点周围８邻域的影响，对网络的作用函数加以修，使改进后的网络具有稳定性好且收敛速度快的优点。通过实验对改进前后网络的识别性能加以比较，证明改进后的网络特别适用于噪声图像的识别。
|
| 关 键 词： | 概率神经网络 邻域统计特性 目标识别 |
PROBABILISTIC NEURAL NETWORK BASED ON THE NEIGHBOR STATISTIC CHARACTER AND ITS APPLICATION IN AUTOMATIC TARGET RECOGNITION |
| |
| Lin Tianrong,Shen Dinggang,Qi Feihu.PROBABILISTIC NEURAL NETWORK BASED ON THE NEIGHBOR STATISTIC CHARACTER AND ITS APPLICATION IN AUTOMATIC TARGET RECOGNITION[J].Journal of Infrared and Millimeter Waves,1995,14(1):52-58. |
| |
| Authors: | Lin Tianrong Shen Dinggang Qi Feihu |
| |
| Abstract: | Based on one of the improved probabilistic neural network models-FDO neural network, the paper presents an idea of taking the effect of the & neighbors of every pixel into consideration when designing the convergent domain of the network. The activation function of the network is modified, thus endowing the network with good stability and high running speed. The recognition capabilities of FDO network and its improved version are compared in the simulation experiment. The result proves that the improved network is especially suitable for the recognition of targets with Gaussian noise. |
| |
| Keywords: | artificial neural network probabilistic neural network convergent domain Gaussian noise |
| 本文献已被 CNKI 维普 等数据库收录！ |
| 点击此处可从《红外与毫米波学报》浏览原始摘要信息 |
|
点击此处可从《红外与毫米波学报》下载全文 |
