摘要
激光荧光光谱数据具有规模大的特点，传统模式识别方法的识别效率低，实时性差，为了解决当前激光荧光光谱数据识别算地存在的不足，以获更高精度的激光荧光光谱数据，设计了一种大数据背景下的激光荧光光谱数据模式识别模型。首先对当前激光荧光光谱数据模式识别的研究现状进行分析，找到引起识别精度精的原因，然后根据大数据原理，引入了云计算平台，将激光荧光光谱数据模式识别划分多个子任务，并通过模式识别方法得到每一个子任务的识别结果，最后对多个子任务的识别结果进行融合，并在Matla2016平台下进行了激光荧光光谱数据模式识别仿真实验。结果表明，本文模型较好的解决了当前激光荧光光谱数据识别方法的弊端，改善了激光荧光光谱数据识别效果，而且激光荧光光谱数据的识别速度加快。
Laser fluorescence spectrum data is large scale, the traditional pattern recognition method has low effi- ciency and poor real -time performance. In order to solve these insufficiencies and get more acearate date, design a novel pattern recognition model in big data background. First, analyze the current research situation to find out the cause of recognition precision, in troduce the cloud computing platform and divide laser fluorescence spectrum data pattern recognition into several subtasks. Then, obtain each recognition results through pattern recognition method. Fi- nally, fuse the results and carry out simulations in Matla 2016 platform. The results show that this model well resolve the defects of current method and improve the recognition effect and speed.
作者
袁书萍
YUAN Shuping(Information Engineering School, Anhui Xinhua University, Hefei 230088, Chin)
出处
《激光杂志》
北大核心
2018年第5期124-127,共4页
Laser Journal
基金
安徽省教育厅项目(No.20158823)
校级自然科学重点研究项目(No.2017zr004)
关键词
大数据
激光荧光
光谱数据
模式识别
large data
laser fluorescence
spectral data
pattern recognition
