本文選題：疾病模式識(shí)別　切入點(diǎn)：激光熒光　出處：《激光雜志》2017年11期

【摘要】：為了解決當(dāng)前疾病模式識(shí)別過(guò)程存在的精度低,速度慢等缺陷,設(shè)計(jì)了一種基于激光熒光光譜數(shù)據(jù)的疾病模式識(shí)別方法。首先收集激光熒光光譜數(shù)據(jù),并對(duì)其進(jìn)行消噪和降維處理,然后基于處理后的激光熒光光譜數(shù)據(jù)建立疾病模式識(shí)別的分類器,最后采用疾病模式識(shí)別的實(shí)驗(yàn)對(duì)本文方法的有效性進(jìn)行測(cè)試,其疾病模式識(shí)別的精度高達(dá)95%以上,并與其它識(shí)別方法進(jìn)行對(duì)比實(shí)驗(yàn),本文方法的疾病模式識(shí)別結(jié)果具有十分明顯的優(yōu)勢(shì),實(shí)際應(yīng)用價(jià)值更高。
[Abstract]:In order to solve the problems of low precision and slow speed in the process of current disease pattern recognition, a method of disease pattern recognition based on laser fluorescence spectrum data was designed. Firstly, the laser fluorescence spectrum data were collected. Then the classification of disease pattern recognition is established based on the processed laser fluorescence spectrum data. Finally, the effectiveness of this method is tested by the experiment of disease pattern recognition. The accuracy of disease pattern recognition is up to 95%, and compared with other recognition methods, the result of disease pattern recognition in this paper has obvious advantages, and the practical application value is higher.
【作者單位】： 吉林農(nóng)業(yè)大學(xué)工程技術(shù)學(xué)院;
【基金】：吉林省教育廳自然科學(xué)基金項(xiàng)目
【分類號(hào)】：R312;TP391.4
，

本文編號(hào)：1675321