1(Food and Pharmaceutical Engineering Institute, Guiyang University, Guiyang 550005, China); 2(Research Center of Nondestructive Testing for Agricultural Products, Guiyang University, Guiyang 550005, China)
摘要 利用可见/近红外光谱对苹果可溶性固形物含量进行检测,并建立了最优预测模型。通过400~1 000 nm高光谱成像系统采集了120个“富士”苹果图像,分析比较了二阶导数(second derivative,SD)、标准正态变换(standard normal variation,SNV)以及多元散射校正(multi-scatter calibration,MSC)3种光谱预处理方法对预测模型的检测效果;分别应用连续投影算法(successive proiection algorithm,SPA)和竞争性自适应重加权算法(competitive adaptive reweighted sampling,CARS)对光谱数据进行降维,进而建立基于特征光谱的误差反向传播(error back propagation,BP)网络和多元线性回归(multiple linear regression,MLR)预测模型。结果表明,二阶导数预处理后的BP网络模型优于原始光谱及其他预处理方法;通过提取特征波长建立的SPA-BP网络模型的预测性能最优,其预测集相关系数rp和均方根误差(root mean square error of prediction set,RMSEP)分别为0.87和0.52。这表明基于可见/近红外光谱检测苹果可溶性固形物含量是可行的。
Abstract: A model of predicting soluble solids content (SSC) of apple by visible-near infrared (Vis/NIR) spectroscopy was established and optimized. The hyperspectral images of 120 “Fuji” apples over 400-1 000 nm were obtained by hyperspectral imaging acquisition system. The effectiveness of the prediction model with pretreatment by second derivative, standard normal variation and multi-scatter calibration was compared and evaluated. Then the successive projection algorithm (SPA) and competitive adaptive reweighted sampling (CARS) methods were used to conduct data mining. Moreover, BP model and multiple linear regression(MLR) model were established based on characteristic spectra. The results showed that BP model with pretreatment by SD was superior to full spectra and other spectral pretreatments. And SPA-BP model based on characteristic spectra had an excellent prediction ability. The correlation coefficient rp and root mean square error of prediction (RMSEP) were 0.87 and 0.52, respectively. These results indicated that it's feasible to determine SSC of apples by Vis/NIR spectroscopy.
[1] 董金磊,郭文川.采后猕猴桃可溶性固形物含量的高光谱无损检测[J].食品科学,2015,36(16):101-106. [2] 尚静,孟庆龙,张艳,等.紫外/可见光谱技术无损检测李子可溶性固形物含量[J].食品工业科技,2020,41(3):228-231. [3] 汪金花,吴兵,徐国强,等.水泥胶砂中水泥水化的高光谱特征分析[J].硅酸盐通报,2019,38(11):3 646-3 653. [4] 王芹志,强锋,何建国,等.基于可见-近红外光谱预测灵武长枣脆度及模型优化[J].食品与发酵工业,2017,43(3):205-211. [5] 孙炳新,匡立学,徐方旭,等.苹果脆度的近红外无损检测[J].食品与发酵工业,2013,39(5):185-189. [6] JIANG J,CEN H,ZHANG C,et al.Nondestructive quality assessment of chili peppers using near-infrared hyperspectral imaging combined with multivariate analysis[J].Postharvest Biology and Technology,2018,146:147-154. [7] 孙旭东,刘燕德,李轶凡,等.鸭梨黑心病和可溶性固形物含量同时在线检测研究[J].农业机械学报,2016,47(1):232-238. [8] HOU J,ZHANG Y,SUN Y,et al.Prediction of firmness and pH for “Golden Delicious” apple based on elasticity index from modal analysis[J].Journal of Food Science,2018,83(1):1-9. [9] 郭志明,赵春江,黄文倩,等.苹果糖度高光谱图像可视化预测的光强度校正方法[J].农业机械学报,2015,46(7):227-232. [10] 杨峰,范亚民,李建龙,等.高光谱数据估测稻麦叶面积指数和叶绿素密度[J].农业工程学报,2010,26(2):237-243. [11] 叶建秋,黄丹平,田建平,等.高光谱图像技术检测大曲发酵过程中的水分含量[J].食品与发酵工业,2020,46(9):250-254. [12] GUO W C,ZHAO F,DONG J L.Nondestructive measurement of soluble solids content of kiwifruits using near-infrared hyperspectral imaging[J].Food Analytical Methods,2016,9:38-47. [13] LI B C,HOU B L,ZHANG D W,et al.Pears characteristics (soluble solids content and firmness prediction, varieties) testing methods based on visible-near infrared hyperspectral imaging[J].Optik,2016,127(5):2 624-2 630. [14] FAN S,HUANG W,GUO Z,et al.Prediction of soluble solids content and firmness of pears using hyperspectral reflectance imaging[J].Food Analytical Methods,2015,8:1 936-1 946. [15] YU X,LU H,WU D.Development of deep learning method for predicting firmness and soluble solid content of postharvest Korla fragrant pear using Vis/NIR hyperspectral reflectance imaging[J].Postharvest Biology and Technology,2018,141:39-49. [16] PU H,LIU D,WANG L,et al.Soluble solids content and pH prediction and maturity discrimination of lychee fruits using visible and near infrared hyperspectral imaging[J].Food Analytical Methods,2016,9(1):235-244. [17] 李瑞,傅隆生.基于高光谱图像的蓝莓糖度和硬度无损测量[J].农业工程学报,2017,33(S1):362-366. [18] LEIVA-VALENZUELA G A,LU R,AGUILERA J M.Assessment of internal quality of blueberries using hyperspectral transmittance and reflectance images with whole spectra or selected wavelengths[J].Innovative Food Science & Emerging Technologies,2014,24:2-13. [19] GALVÀO R K H,ARAUJO M C U,JOSÉ G E,et al.A method for calibration and validation subset partitioning[J].Talanta,2005,67(4):736-740. [20] 倪力军,张立国.基础化学计量学及其应用[M].上海:华东理工大学出版社,2011.