基于多信息融合的疫苗培养基检测研究  

Research on detection of vaccines medium based on multi-information fusion

作　　者：崔建鹏[1] 徐彦伟[1] 颉潭成[1] 南翔[1] 王俊澎 刘守川 

机构地区：[1]河南科技大学机电工程学院,河南洛阳471003 [2]洛阳普莱柯生物工程股份有限公司,河南洛阳471000

出　　处：《传感器与微系统》2014年第5期18-20,23,共4页Transducer and Microsystem Technologies

基　　金：国家自然科学基金资助项目(51305127);河南科技大学青年科学基金资助项目(2012QN024)

摘　　要：针对疫苗培养基检测准确率较低的问题,首次提出利用机器视觉、红外温度和光照度的多传感器信息融合技术检测疫苗培养基。通过BP神经网络将3种传感器采集到的信息在特征层进行融合,在模型的建立过程中,对各个信息的主成分因子数进行了优化。从实验结果看,模型对预测集检测的识别率为97%,实验结果表明:通过多信息融合方法检测疫苗培养基是可行的,检测结果的准确性和稳定性相对于单个模型有明显提高。Aiming at problem of low detection accuracy of vaccine medium, first propose using machine vision, multi-sensor information fusion technology of infrared temperature and illumination to detect vaccine medium. Through BP neural network,information collected by three kinds of sensors is fused in feature level, in process of model building, numbers of principal components factor of each message are optimized. From experimental results, detection recognition rate of model on prediction set is 97 %, test results show that detecting vaccine medium through multi-information fusion method is feasible, accuracy and stability of test results are improved significantly relative to individual model.

关 键 词：多信息融合 疫苗培养基 检测 机器视觉 红外温度 光照度 

分 类 号：TP274.5[自动化与计算机技术—检测技术与自动化装置]