利用高光谱技术检测大曲发酵品质时,获取的水分等高含量物质的高光谱数据可能掩盖对大曲质量评价至关重要的微量物质高光谱数据。为方便后续更微量物质的光谱曲线分解,需先排除水分等高含量物质的数据干扰,该文通过建立无监督的深度自编码模型,可实现大曲水分高光谱曲线分解。通过实验设计,采集与水混合后的成品曲粉光谱数据。首先编码部分,将混合大曲光谱曲线压缩为低维表示,即端元;解码部分,将光谱的低维表示解压重构为原始光谱曲线,结合比较不同目标函数,反向传递重构误差,更新解码权重;最终,通过端元解码出水与曲粉各自的光谱曲线,运用欧氏距离与皮尔逊相关系数方法从特征距离和相关系数两方面同时评价解混效果。实验显示,利用深度自编码解混模型,选择L-C目标函数得到的解混效果最好,解混曲线与纯曲粉曲线的欧式距离与皮尔逊相似度分别为0.342 7和0.996 7。研究表明,利用深度自编码网络能够对大曲高光谱数据进行解混,可为实现大曲高光谱微量物质检测提供理论支持和技术支撑。
When hyperspectral technology is used to detect the fermentation quality of Daqu, the obtained hyperspectral data of high-content substances, such as water, may conceal the hyperspectral data of trace substances, which are essential for the quality evaluation of Daqu. In order to eliminate the interference of water and other high-content substances and decompose the spectral curve of trace substances, the unsupervised deep autoencoder model was established to decompose the water hyperspectral curve of Daqu. Through the experimental design, the spectral data of the mixture of Daqu powder and water were collected. Firstly, in the coding part, the spectral curve of mixed Daqu was compressed into a low dimensional representation, that is, the end member; in the decoding part, the low dimensional representation of the spectrum was decompressed and reconstructed into an original spectral curve, and the decoding weight could be updated by comparing different objective functions and transferring the reconstruction error in reverse; finally, the spectral curves of water and Daqu powder were decoded through the end member, then the Euclidean distance and Pearson correlation coefficient method were used to evaluate the effect of the solution from two aspects of characteristic distance and correlation coefficient at the same time. The experiment showed that using deep autoencoder model and choosing the L-C objective function could get the best result, the Euclidean distance and Pearson similarity of the demixed curve and the pure powder curve were 0.342 7 and 0.996 7, respectively. The results showed that the deep autoencoder network can be used to demix the hyperspectral data of Daqu, which can provide theoretical and technical support for the detection of trace substances of Daqu hyperspectral.
[1] 程劲松, 李春扬.白酒质量控制技术的研究进展[J].食品安全质量检测学报, 2014, 5(7):2 248-2 262.
CHENG J S, LI C Y.Development of quality control technologies for Chinese liquor[J].Journal of Food Safety & Quality, 2014, 5(7):2 248-2 262.
[2] 隋明, 张崇军, 张凤英, 等.白酒质量控制技术的研究进展[J].食品工程, 2018(4):8-10.
SUI M, ZHANG C J, ZHANG F Y, et al.Research progress of liquor quality control technology[J].Food Engineering, 2018(4):8-10.
[3] 吴翠芳, 刘国英, 何宏魁.探索新时代白酒可持续发展的途径[J].酿酒, 2018, 45(3):10-12.
WU C F, LIU G Y, HE H K.Exploring the way of sustainable development of baijiu in the new age[J].Liquor Making, 2018, 45(3):10-12.
[4] 徐姿静, 徐占成, 唐清兰.中国传统固态大曲酒风味质量特色的研究进展[J].酿酒科技, 2017(1):84-89.
XU Z J, XU Z C, TANG Q L.Research progress in the flavor & quality characteristics of traditional chinese daqu baijiu by solid-state fermentation[J].Liquor-Making Science & Technology, 2017(1):84-89.
[5] 敖宗华, 陕小虎, 沈才洪, 等.国内主要大曲相关标准及研究进展[J].酿酒科技, 2010(2):104-108.
AO Z H, SHAN X H, SHEN C H, et al.Domestic related quality standards of daqu & its research progress[J].Liquor-Making Science & Technology, 2010(2):104-108.
[6] 沈怡方. 白酒生产技术全书[M].北京:中国轻工业出版社.1998.
SHEN Y F.Complete Book of Liquor Production Technology[M].Beijing:China Light Industry Press,1998.
[7] 袁迎辉, 林子瑜.高光谱遥感技术综述[J].中国水运(学术版), 2007(8):155-157.
YUAN Y H, LIN Z Y.Review of hyperspectral remote sensing technology[J].China Water Transport(Academic Version), 2007(8):155-157.
[8] 张达, 郑玉权.高光谱遥感的发展与应用[J].光学与光电技术, 2013, 11(3):67-73.
ZHANG D,ZHENG Y Q.Hyperspectral remote sensing and its development and application review[J].Optics & Optoelectronic Technology, 2013, 11(3):67-73.
[9] 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.
YE J Q, HUANG D P, TIAN J P, et al.Detection of water content in Daqu during fermentation using hyperspectral image technology[J].Food and Fermentation Industries, 2020, 46(9):250-254.
[10] 高泽东, 郝群, 刘宇, 等.高光谱成像与应用技术发展[J].计测技术, 2019, 39(4):24-34.
GAO Z D, HAO Q, LIU Y, et al.Hyperspectral imaging and application technology development[J].Metrology & Measurement Technology, 2019, 39(4):24-34.
[11] 张芯豪, 黄丹平, 田建平, 等.基于机器视觉的大曲质量检测系统研究[J].食品与机械, 2018, 34(4):80-84.
ZHANG X H, HUANG D P, TIAN J P.et al.Research on the daqu quality detection system based on machine vision[J].Food & Machinery, 2018, 34(4):80-84.
[12] 庞云璇. 谈高光谱遥感技术的应用、发展与展望[J].中国新通信, 2019, 21(11):73-74.
PANG Y X.Application, development and prospect of hyperspectral remote sensing technology[J].China New Telecommunications, 2019, 21(11):73-74.
[13] 高晓惠. 高光谱数据处理技术研究[D].西安:中国科学院研究生院(西安光学精密机械研究所), 2013.
GAO X H.Research on technology of hyperspectral data processing[D].Xi'an:Chinese Academy of Sciences(Xi'an Institute of Optics & Precision Mechanics), 2013.
[14] 储成泉. 高光谱图像解混技术的研究[D].重庆:重庆邮电大学, 2019.
CHU C Q.Research on hyperspectral image unmixing technology[D].Chongqing:Chongqing University of Posts and Telecommunications, 2019.
[15] BIOUCAS-DIAS J M, PLAZA A, CAMPS-VALLS G, et al.Hyperspectral remote sensing data analysis and future challenges[J].IEEE Geoscience and Remote Sensing Magazine, 2013, 1(2):6-36.
[16] BIOUCAS-DIAS J M, PLAZA A, DOBIGEON N, et al.Hyperspectral unmixing overview:Geometrical, statistical, and sparse regression-based approaches[J].IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2012, 5(2):354-379.
[17] LU X, DONG L, YUAN Y.Subspace clustering constrained sparse NMF for hyperspectral unmixing[J].IEEE Transactions on Geoscience and Remote Sensing, 2020, 58(5):3 007-3 019.
[18] ZHANG S, AGATHOS A, LI J.Robust minimum volume simplex analysis for hyperspectral unmixing[J].IEEE Transactions on Geoscience and Remote Sensing, 2017, 55(11):6 431-6 439.
[19] RAJABI R, GHASSEMIAN H.Spectral unmixing of hyperspectral imagery using multilayer NMF[J].IEEE Geoscience and Remote Sensing Letters, 2015, 12(1):38-42.
[20] ZHANG X, SUN Y, ZHANG J, et al.Hyperspectral unmixing via deep convolutional neural networks[J].IEEE Geoscience and Remote Sensing Letters, 2018, 15(11):1 755-1 759.
[21] PALSSON B, SIGURDSSON J, SVEINSSON J R, et al.Hyperspectral unmixing using a neural network autoencoder[J].IEEE Access, 2018, 1(6):25 646-25 656.
[22] YUAN J, ZHANG Y J, GAO F P.An overview on linear hyperspectral unmixing[J].Journal of Infrared and Millimeter Waves, 2018, 37(5):553-571.
[23] HEYLEN R, PARENTE M, GADER P.A review of nonlinear hyperspectral unmixing methods[J].IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2014, 7(6):1 844-1 868.
[24] WANG M, ZHAO M, CHEN J, et al.Nonlinear unmixing of hyperspectral data via deep autoencoder networks[J].IEEE Geoscience and Remote Sensing Letters, 2019, 16(9):1 467-1 471.
