| 基于BP网络的血液气味识别模型的建立 |
| |
| 引用本文： | 龙成生,王辛,单军,吴德华,宋珍华.基于BP网络的血液气味识别模型的建立[J].信息技术,2012(6):33-35. |
| |
| 作者姓名： | 龙成生 王辛 单军 吴德华 宋珍华 |
| |
| 作者单位： | 公安部南京警犬研究所,南京,210012 |
| |
| 摘 要： | 以血液气味样品的气相色谱质谱分析结果为基础,建立了一个基于BP人工神经网络的血液气味识别模型,并利用Matlab计算平台对此模型进行了优化、训练和测试.此模型的网络结构为9×13×1,隐含层传递函数为tansig,输出层传递函数为logsig,训练函数为trainrp.优化后的模型对血液样品的正确识别率为100％.
|
| 关 键 词： | 血液气味 识别 BP网络 |
Identification of blood odors based on BP artificial neural network |
| |
| LONG Cheng-sheng
,
WANG Xin
,
SHAN Jun
,
WU De-hua
,
SONG Zhen-hua.Identification of blood odors based on BP artificial neural network[J].Information Technology,2012(6):33-35. |
| |
| Authors: | LONG Cheng-sheng WANG Xin SHAN Jun WU De-hua SONG Zhen-hua |
| |
| Institution: | (Nanjing Police Dog Research Institute of Public Security Ministry,Nanjing 210012,China) |
| |
| Abstract: | This paper proposed an identification model for blood odors based on Back Propagation Artificial Neural Network.The features were extracted from the data of blood samples which were analyzed by means of Gas Chromatography-Mass Spectrometer.Matlab was used to optimize the model,which then was trained and tested using human and animal blood odor samples.The architecture of the model was 9×13×1,the transfer functions of the hidden layer and the output layer were tansig and logsig,respectively,the train function was trainrp.The optimized model correctly identified the train and test samples with the accuracy of 100%. |
| |
| Keywords: | blood odor recognition BP ANN |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
