1. College of Quality and Technical Supervision, Hebei University, Baoding 071002, China 2. College of Mechanical and Electrical Engineering, Agricultural University of Hebei, Baoding 071001, China
Abstract：Strawberry variety is a main factor that can influence strawberry fruit quality. The use of near-infrared reflectance spectroscopy was explored discriminate among samples of strawberry of different varieties. And the significance of difference among different varieties was analyzed by comparison of the chemical composition of the different varieties samples. The performance of models established using back propagation-artificial neural networks (BP-ANN), least squares-support vector machine and discriminant analysis were evaluated on spectra range of 4 545～9 090 cm-1. The optimal model was obtained by BP-ANN with a topology of 12-18-3, which correctly classified 96.68% of calibration set and 97.14% of prediction set. And the 94.95%, 97% and 98.29% classifications were given respectively for “Tianbao” (n=99), “Fengxiang” (n=100) and “Mingxing” (n=117). One-way analysis of variance was made for comparison of the mean values for soluble solids content (SSC), titratable acid (TA), pH value and SSC-TA ratio, and the statistically significant differences were found. Principal component analysis was performed on the four chemical compositions, and obvious clustering tendencies for different varieties were found. These results showed that NIR combined with BP-ANN can discriminate strawberry of different varieties effectively, and the difference in chemical compositions of different varieties strawberry might be a chemical validation for NIR results.
Key words：Strawberry;Near infrared spectroscopy;Back propagation-artificial neural networks;One-way analysis of variance
[1] LUO Ya, TANG Yong, FENG Shan, et al(罗 娅, 唐 勇, 冯 姗, 等). Food Science(食品科学), 2011, 32(7): 52. [2] CHEN Wei-ping(陈卫平). Acta Agriculturae Jiangxi(江西农业学报), 2010, 22(9): 46. [3] ZHAO Jie-wen, HU Huai-ping, ZOU Xiao-bo(赵杰文, 呼怀平, 邹小波). Transactions of the Chinese Society of Agricultural Engineering (农业工程学报), 2007, 23(4): 149. [4] Cen H Y, He Y, Huang M. European Food Research and Technology, 2007, 225(5-6): 699. [5] He Y, Li X L, Shao Y N. International Journal of Food Properties, 2007, 10(1): 9. [6] Li X L, He Y, Fang H. Journal of Food Engineering, 2007, 91: 357. [7] WANG Hui-rong, LI Wei-jun, LIU Yang-yang, et al(王徽蓉, 李卫军, 刘扬阳, 等). Spectroscopy and Spectral Analysis(光谱学与光谱分析)，2011, 31(3): 669. [8] Sánchez M T, José De la Haba M, Benítez-López M, et al. Journal of Food Engineering，2012, 110: 102. [9] LI Xian-feng, ZHU Wei-xing, JI Bin, et al(李先锋, 朱伟兴, 纪 滨, 等). Transactions of the Chinese Society for Agricultural Machinery(农业机械学报)，2010, 41(11): 168. [10] ZHAN Hui, LI Xiao-yu, ZHOU Zhu, et al(展 慧, 李小昱, 周 竹, 等). Transactions of the Chinese Society of Agricultural Engineering(农业工程学报)，2011, 27(2): 345. [11] Ito S, Kito I. Nosanbutsu Ryutsu Riyo Shiken Seisekisho，1999, 12. [12] Ito H. Acta Horticulturae，2002, 567: 751. [13] Ito H, Fukino-Ito N, Horie H. Acta Horticulturae, 2005, 687: 271. [14] Nishizawa T, Mori Y, Fukushima S, et al. Journal of the Japanese Society for Food Science and Technology, 2009, 56(4): 229. [15] Shao Y N, He Y. International Journal of Food Properties, 2008, 11: 102.
