基于CARS-MIV-SVR的库尔勒香梨可溶性固体含量预测方法

doi:10.3964/j.issn.1000-0593(2019)11-3547-06

Abstract
References
Related Citation (15)

Download: PDF (2909 KB)
Export: BibTeX | EndNote (RIS)

Abstract In order to classify and set different prices on the basis of soluble solid content (SSC) of korla pears and promote the development of post-harvest processing healthily in standardization and industrialization, a fast, precise and nondestructive method to detect soluble solid content of korla pears was determined by applying hyperspectral reflectance imaging technology. 157 korla pears freshly and with no surface damage were collected as samples. Hyperspectral images with a spectral range of 400~1000 nm of pears were acquired by hyperspectral imaging system. Then the region of interest (ROI) function of ENVI 5.3 software was used to conduct spectral data extraction from each hyperspectral image of pear. Totally, 157 pear samples were divided into calibration set (105) and prediction set (52) based on the Kennard-Stone（KS）sample set partitioning method. The research compared the influence of accuracy of modeling in terms of the spectrum pretreatment methods of original spectrum, standard normal variate (SNV), multiplicative scatter correction (MSC), first derivative (FD) and second derivative (SD). The SNV was applied for smoothing and denoising of the original hyperspectral data. A variable selection method combining competitive adaptive reweighted sampling and mean impact value (CARS-MIV) was utilized to extract the characteristic variables from full spectrum (FS). The modeled samples of competitive adaptive reweighted sampling (CARS) are generated by random selection of Monte Carlo sampling, and the regression coefficients of variables will change accordingly. The absolute value of regression coefficients cannot fully reflect the importance of variables, and affect the accuracy of the model. To lower the impact, the mean impact value (MIV) algorithm is applied to select the independent variables for secondary screening, and the variables with bigger correlation are selected for modeling and analysis. In this paper, the variables selected by CARS, successive projection algorithm (SPA) and Monte-Carlo uninformative variable elimination (MCUVE) were used for comparison. Finally, the spectral information selected from full wavelength and the spectral information selected from four characteristic wavelength selection method were taken as input vector to build support vector regression（SVR）model to predict soluble solid content of korla pears. The performances of the models were evaluated by the root of mean square of calibration (RMSEC), the root of mean square of prediction (RMSEP), the correlation coefficient of calibration (R_c) and the correlation coefficient of prediction (R_p). By means of comparison, the CARS-MIV-SVR models achieved the optimal performance with the R_c reaching 0.985 94 and R_p up to 0.946 31. The RMSEC and RMSEP are 0.185 85 and 0.403 33 respectively. These experimental results demonstrated that CSRS-MIV method can efficiently improve the stability and accuracy of wavelength selection, and optimize the precision of prediction model. The hyperspectral technique combined with CARS-MIV-SVR model can meet the needs of determination of soluble solid content and be used to classify and set different prices on the basis of SSC of korla pears.

Key words：Spectral analysis; Soluble solid content; Variable selection; Competitive adaptive reweighted sampling-mean impact value; Support vector regression

Received: 2018-10-09 Accepted: 2019-02-10

ZTFLH:

TP391

Corresponding Authors: LI Guang-hui E-mail: ghli@jiangnan.edu.cn

Cite this article:

URL:

https://www.gpxygpfx.com/EN/10.3964/j.issn.1000-0593(2019)11-3547-06 OR https://www.gpxygpfx.com/EN/Y2019/V39/I11/3547

[1] Tan W, Sun L, Yang F, et al. Optik, 2018, 154: 581.
[2] Gabriel A Leiva-Valenzuela, RenfuLu José Miguel Aguilera. Innovative Food Science and Emerging Technologies, 2014, 24: 2.
[3] JIE Deng-fei, LI Ze-hai, ZHAO Jun-wei, et al（介邓飞, 李泽海, 赵竣威, 等）. Chinese Journal of Luminescence（发光学报）, 2017, 38(5): 685.
[4] GU Wen-jun, TIAN You-wen, ZHANG Fang, et al（古文君, 田有文, 张芳, 等）. Journal of Shenyang Aricultural University（沈阳农业大学学报）, 2017, 48(5): 584.
[5] Rajkumar P, Wang N, EImasry G, et al. Journal of Food Engineering, 2012, 108: 194.
[6] Wang A, Fu X, Xie L. Food Analytical Methods, 2015, 8(6): 1403.
[7] Li J, Tian X, Huang W, et al. Food Analytical Methods, 2016, 9(11): 3087.
[8] HE Hong-ju, HU Li-ping, LI Wu, et al（何洪巨, 胡丽萍, 李武, 等）. Food and Nutrition in China（中国食物与营养）, 2016, 22(10): 53.
[9] Dong J, Guo W. Food Analytical Methods, 2015, 8(10): 2635.
[10] Hui L, Liang L, Zhang L, et al. Transactions of the Chinese Society of Agricultural Engineering, 2013, 29(11): 139.
[11] FENG Di, JI Jian-wei, ZHANG Li, et al（冯迪, 纪建伟, 张莉, 等）. Chinese Journal of Luminescence（发光学报）, 2017, 38(6): 799.
[12] Hu M H. Postharvest Biology & Technology, 2016, 115: 122.
[13] Liu D, Sun D W, Zeng X A. Food and Bioprocess Technology, 2014, 7(2): 307.
[14] WANG Hai-long, YANG Guo-guo, ZHANG Yu, et al（王海龙, 杨国国, 张瑜, 等）. Spectroscopy and Spectral Analysis（光谱学与光谱分析）, 2017, 37(7): 2115.
[15] CHEN Xin, LIU Fei（陈鑫, 刘飞）. Computers and Applied Chemistry（计算机与应用化学）, 2012, 29(7): 812.
[16] Fan S, Huang W, Guo Z, et al. Food Analytical Methods, 2015, 8(8): 1936.
[17] Hao Y, Sun X, Zhang H, et al. Sensor Letters, 2011, 9(3): 1229.
[18] SUN Jun, CONG Sun-li, MAO Han-ping, et al（孙俊, 丛孙丽, 毛罕平, 等）. Transactions of the Chinese Society of Agricultural Engineering（农业工程学报）, 2017, 33(5): 178.