CAJ | 학술논문

基于近地高光谱成像技术结合化学计量学方法，实现了黑豆品种的鉴别。实验以三种不同颜色豆芯的黑豆为研究对象，采用高光谱成像系统采集380～1030 nm波段范围的高光谱图像，提取高光谱图像中的样本感兴趣区域平均光谱信息作为样本的光谱进行分析，建立黑豆品种的判别分析模型。共采集180个黑豆样本的180条平均光谱曲线。剔除明显噪声部分之后以440～943 nm范围光谱为黑豆样本的光谱，采用多元散射校正（multiplicative scatter correction ，MSC）对光谱曲线进行预处理。分别以全部光谱数据、主成分分析（principal component analysis ，PCA）提取的光谱特征信息、小波分析（wavelet transform ，WT）提取的光谱特征信息建立了偏最小二乘判别分析法（partial least squares discriminant analysis ，PLS-DA ），簇类独立模式识别法（soft independent modeling of class analogy ，SIMCA），最邻近节点算法（K-nearest neighbor algo-rithm ，KNN），支持向量机（support vector machine ，SVM），极限学习机（extreme learning machine ，ELM）等判别分析模型。以全谱的判别分析模型中，ELM 模型效果最优；以PCA提取的光谱特征信息建立的模型中，ELM模型也取得了最优的效果；以WT 提取的光谱特征信息建立的模型中，ELM 模型结识别效果最好，建模集和预测集识别正确率达到100％。在所有的判别分析模型中，W T-EL M模型取得了最优的识别效果。实验结果表明以高光谱成像技术对黑豆品种进行无损鉴别是可行的，且WT 用于提取光谱特征信息以及ELM模型用于判别黑豆品种能取得较好的效果。
기우근지고광보성상기술결합화학계량학방법，실현료흑두품충적감별。실험이삼충불동안색두심적흑두위연구대상，채용고광보성상계통채집380～1030 nm파단범위적고광보도상，제취고광보도상중적양본감흥취구역평균광보신식작위양본적광보진행분석，건립흑두품충적판별분석모형。공채집180개흑두양본적180조평균광보곡선。척제명현조성부분지후이440～943 nm범위광보위흑두양본적광보，채용다원산사교정（multiplicative scatter correction ，MSC）대광보곡선진행예처리。분별이전부광보수거、주성분분석（principal component analysis ，PCA）제취적광보특정신식、소파분석（wavelet transform ，WT）제취적광보특정신식건립료편최소이승판별분석법（partial least squares discriminant analysis ，PLS-DA ），족류독립모식식별법（soft independent modeling of class analogy ，SIMCA），최린근절점산법（K-nearest neighbor algo-rithm ，KNN），지지향량궤（support vector machine ，SVM），겁한학습궤（extreme learning machine ，ELM）등판별분석모형。이전보적판별분석모형중，ELM 모형효과최우；이PCA제취적광보특정신식건립적모형중，ELM모형야취득료최우적효과；이WT 제취적광보특정신식건립적모형중，ELM 모형결식별효과최호，건모집화예측집식별정학솔체도100％。재소유적판별분석모형중，W T-EL M모형취득료최우적식별효과。실험결과표명이고광보성상기술대흑두품충진행무손감별시가행적，차WT 용우제취광보특정신식이급ELM모형용우판별흑두품충능취득교호적효과。
In the present study ,hyperspectral imaging combined with chemometrics was successfully proposed to identify differ-ent varieties of black bean .The varieties of black bean were defined based on the three different colors of the bean core .The hy-perspectral images in the spectral range of 380~1 030 nm of black bean were acquired using the developed hyperspectral imaging system ,and the reflectance spectra were extracted from the region of interest (ROI) in the images .The average spectrum of a ROI of the sample in the images was used to represent the spectrum of the sample and build classification models .In total ,180 spectra of 180 samples were extracted .The wavelengths from 440 to 943 nm were used for analysis after the removal of the spec-tral region with absolute noises ,and 440~943 nm spectra were preprocessed by multiplicative scatter correction (MSC) .Five classification methods ,including partial least squares discriminant analysis (PLS-DA) ,soft independent modeling of class analo-gy (SIMCA) ,K-nearest neighbor algorithm (KNN) ,support vector machine (SVM) and extreme learning machine (ELM) , were used to build discriminant models using the preprocessed full spectra ,the feature information extracted by principal compo-nent analysis (PCA) and the feature information extracted by wavelet transform (WT ) from the preprocessed spectra ,respec-tively .Among all the classification models using the preprocessed full spectra ,ELM models obtained the best performance ;among all the classification models using the feature information extracted from the preprocessed spectra by PCA ,ELM model also obtained the best classification accuracy ;and among all the classification models using the feature information extracted from the preprocessed spectra by WT ,ELM models obtained the best classification performance with 100% accuracy in both the cali-bration set and the prediction set .Among all classification models ,WT-ELM model obtained the best classification accuracy . The overall results indicated that it was feasible to identify black bean varieties nondestructively by using hyperspectral imaging , and WT could effectively extract feature information from spectra and ELM algorithm was effective to build high performance classification models .