基于激光荧光光谱数据的疾病模式识别  

Recognition of disease patterns based on laser fluorescence spectrum data

作　　者：王英男[1] 

机构地区：[1]吉林农业大学工程技术学院,长春130118

出　　处：《激光杂志》2017年第11期117-120,共4页Laser Journal

基　　金：吉林省教育厅自然科学基金项目

摘　　要：为了解决当前疾病模式识别过程存在的精度低,速度慢等缺陷,设计了一种基于激光荧光光谱数据的疾病模式识别方法。首先收集激光荧光光谱数据,并对其进行消噪和降维处理,然后基于处理后的激光荧光光谱数据建立疾病模式识别的分类器,最后采用疾病模式识别的实验对本文方法的有效性进行测试,其疾病模式识别的精度高达95%以上,并与其它识别方法进行对比实验,本文方法的疾病模式识别结果具有十分明显的优势,实际应用价值更高。In order to solve the defects of low precision and slow speed in the current pattern recognition process,a disease pattern recognition method based on laser fluorescence spectrum data was designed. First collect laser fluorescence spectra data and carry on the denoising and dimensionality reduction. Then,establish disease pattern recognition classifier based on processed laser fluorescence spectrum data. Finally,the effectiveness of this disease pattern recognition were tested,the accuracy is up to 95%. and it has obvious advantages and higher application value compared with other methods.

关 键 词：疾病模式识别 激光荧光 光谱数据 数据去噪 

分 类 号：TN391[电子电信—物理电子学]