首页|高光谱成像技术鉴别长鞭红景天和四裂红景天

高光谱成像技术鉴别长鞭红景天和四裂红景天

扫码查看
目的 采用近红外高光谱成像技术(HSI)鉴别四裂红景天和长鞭红景天.方法 采集935~1720 nm长鞭红景天和四裂红景天的高光谱图像信息.采用主成分分析法(PCA)和正交-偏最小二乘判别分析法(OPLS-DA)进行分类判别.结果 OPLS-DA比PCA分析具有更好的分类性能,HSI和OPLS-DA分析结合可准确识别两种红景天.结论 所用方法可快速、无损地鉴别长鞭红景天和四裂红景天,为红景天的品种鉴定、质量控制和评估提供了参考.
Identification of Rhodiola fastigiata and Rhodiola quadrifida by Hyperspectral imaging technique
OBJECTIVE To identify the Rhodiola fastigiata and R.quadrifida by near-infrared hyperspectral imaging(HSI).METHODS HSI information of Rhodiola was collected in the wavelength range of 935 nm to 1720 nm.Principal component analysis(PCA)and Orthogonal partial least squares-discriminant analysis(OPLS-DA)were used for classification.RESULTS OPLS-DA had better classification performance than PCA,and the combination of HSI and OPLS-DA analysis could accurately identify the two Rhodiola species.CONCLUSION This method can quickly and nondestructively identify R.fastigiata and R.quadrifida,which provides reference for species identification,quality control and evaluation of Rhodiola medicinal plants.

Hyperspectral imaging techniqueRhodiola fastigiataRhodiola quadrifidaSuluo MabaoChemical pattern recognitionPCAOPLS-DA

丁丹、李涛、杨语

展开 >

四川大学华西药学院,四川成都 610041

高光谱成像技术 四裂红景天 长鞭红景天 苏罗玛保 化学模式识别 主成分分析 正交-偏最小二乘法判别分析

四川大学—达州市人民政府战略合作专项资金项目四川省中医药管理局全国第四次中药资源普查科技专项

2021CDDZ-222019PC003

2024

华西药学杂志
四川大学,四川省药学会

华西药学杂志

CSTPCD北大核心
影响因子:0.624
ISSN:1006-0103
年,卷(期):2024.39(4)
  • 3