基于近红外(near-infrared,NIR)以及可见光(visible,VIS)高光谱技术快速评估窖泥总酸的分布。化学计量法结合计算机技术分析窖泥在近红外以及可见光波段下的高光谱数据,结合总酸实测值建立偏最小二乘回归、最小二乘支持向量机2种预测模型。根据模型的表现性能,最优模型为可见光区域下的SNV-SPA-SVM模型,训练集的决定系数R2cal为0.998 5,均方根误差为0.004 9 g/kg,测试集的决定系数R2pre为0.999 1,均方根误差为0.003 8 g/kg,并计算得到不同窖龄、不同层次窖泥总酸度的可视化分布图。结果表明,将高光谱技术应用于窖泥总酸的快速无损检测是可行的,此技术帮助白酒企业快速发现问题,及时调整工艺,防止窖泥酸化和老化现象的发生,同时为中国白酒行业传统技术的转型升级以及智能化在线实时监控窖泥质量提供了有力的技术支持。
The total acid distribution in pit mud was rapidly assessed using visible (VIS)/near-infrared (NIR) hyperspectral technology. The hyperspectral data of pit mud in the near-infrared and visible wavebands were analyzed by stoichiometry combined with computer technology. Two prediction models, the partial least squares regression (PLSR) and a least squares-support vector machine (LS-SVM) were constructed based on these measurements. The model performances indicated that the optimal model was the Standard Normal Variable Correction-Successive Projection Algorithm-SVM (SNV-SPA-SVM) in the visible region. The coefficient of determination R2cal of the calibration set was 0.998 5, root mean square error of calibration (RMSEC) was 0.004 9 g/kg, and the coefficient of the determination R2pre of the prediction set was 0.999 1. Furthermore, the root mean square error of prediction (RMSEP) was 0.003 8 g/kg and a visual distribution map of the total acid in pit mud was obtained. The results showed that it was feasible to employ hyperspectral technology for rapid and non-destructive detection of the total acid in pit mud, enabling baijiu enterprises to identify problems quickly, adjust processes in a timely manner, prevent pit mud acidification and aging, as well as provide strong technical support for the transformation and upgrading of traditional technology in the Chinese baijiu industry and intelligent online real-time monitoring of pit mud quality.
[1] ZHAO Yuping, ZHENG Xiangping,SONG Pu, et al. Characterization of volatiles in the six most well-known distilled spirits[J]. Journal of the American Society of Brewing Chemists, 2013, 71(3):161-169.
[2] 胡晓龙. 浓香型白酒窖泥中梭菌群落多样性与窖泥质量关联性研究[D].无锡:江南大学,2015.
[3] 赵东,郑佳,彭志云,等. 五粮浓香型发酵糟醅与窖泥的相关性研究[J]. 食品与发酵工业, 2019,45(6): 1-7.
[4] WANG Xueshan, DU Hai, XU Yan. Source tracking of prokaryotic communities in fermented grain of Chinese strong-flavor liquor[J]. International Journal of Food Microbiology, 2017, 244: 27-35.
[5] 章发盛,张学英. 预防酿酒窖泥老化的研究[J]. 酿酒, 2010, 37(6):45-46.
[6] HEGE E K,O'CONNELL D,JOHNDON W, et al. Hyperspectral imaging for astronomy and space surveillance in imaging Spectrometry IX[J]. International Society for Optics and Photonics,2004,5 159:380-391.
[7] EDELMAN G J,GASTON E,LEEOWEN G V T. Hyperspectral imaging for non-contact analysis of forensic traces[J]. Forensic Science International, 2012, 223(1-3): 28-39.
[8] MALKOFF D B, OLIVER W R. Hyperspectral imaging applied to forensic medicine[J]. Proceedings of Spie-the International Society for Optical Engineering, 2000, 3 920: 108-116.
[9] KUULA J, PÖLÖNENI H,PUUPPONEN H H, et al. Using VIS/NIR and IR spectral cameras for detecting and separating crime scene details[J]. Proceedings of Spie-the International Society for Optical Engineering, 2012, 8 359: 13.
[10] SCHULER R L,KISH P E,PLESE C A.Preliminary observations on the ability of hyperspectral imaging to provide detection and visualization of bloodstain patterns on black fabrics[J]. Journal of Forensic Sciences, 2015, 57(6): 1 562-1 569.
[11] FISCHER C,KAKOULLI I. Multispectral and hyperspectral imaging technologies in conservation: Current research and potential applications[J]. Studies in Conservation, 2006, 51(Sup 1): 3-16.
[12] LIANG Haida. Advances in multispectral and hyperspectral imaging for archaeology and art conservation[J]. Applied Physics A, 2012, 106(2): 309-323.
[13] AFROMOWITZ M A, CALLIS J B,HEIMBACH D M, et al. Multispectral imaging of burn wounds: A new clinical instrument for evaluating burn depth[J].IEEE transactions on bio-medical engineering, 1988, 35(10): 842-850.
[14] CARRASCO O,GOMEZ R B,CHAINANI A, et al. Hyperspectral imaging applied to medical diagnoses and food safety[J]. Proceedings of Spie-the International Society for Optical Engineering, 2003, 5 097: 215-221.
[15] ADAM E,MUTANGA O,RUGEGE D. Multispectral and hyperspectral remote sensing for identification and mapping of wetland vegetation: A review[J]. Wetlands Ecology & Management, 2010, 18(3): 281-296.
[16] GOVENDER M,CHETTY K,BULCOCK H. A review of hyperspectral remote sensing and its application in vegetation and water resource studies[J]. Water SA, 2009, 33(2):141-145.
[17] REN J,ZABALZA J,MARSHALL S, et al. Effective feature extraction and data reduction in remote sensing using hyperspectral imaging[J]. IEEE Signal Processing Magazine, 2014, 31(4): 149-154.
[18] ZHU Yaodi,ZOU Xiaobo,SHEN Tingting, et al. Determination of total acid content and moisture content during solid-state fermentation processes using hyperspectral imaging[J]. Journal of Food Engineering, 2016, 174: 75-84.
[19] MUNERA S,BESADA C,ALEIXOS N, et al. Non-destructive assessment of the internal quality of intact persimmon using colour and VIS/NIR hyperspectral imaging[J]. LWT - Food Science and Technology, 2017, 77: 241-248.
[20] LING Yan,XIONG Chuanwu,HAO Qu, et al. Non-destructive determination and visualisation of insoluble and soluble dietary fibre contents in fresh-cut celeries during storage periods using hyperspectral imaging technique[J]. Food Chemistry, 2017, 228: 249-256.
[21] FERREIRA D S,GALÃO O F,PALLONE J A L, et al. Comparison and application of near-infrared (NIR) and mid-infrared (MIR) spectroscopy for determination of quality parameters in soybean samples[J]. Food Control, 2014, 35(1): 227-232.
[22] HE Hongju, WU Di, SUN Dawen. Rapid and non-destructive determination of drip loss and pH distribution in farmed Atlantic salmon (Salmo salar) fillets using visible and near-infrared (Vis-NIR) hyperspectral imaging[J]. Food Chemistry, 2014, 156:394-401.
