Microsoft Word 邢素霞-二校-出2998,2999

Transcription

1 6 8 Vol. 6 No Journal of Food Safety and Quality Aug., 2015 邢素霞 *, 郭培源, 向灵孜, 梁超群 (, ) 摘要 : 目的 方法 30, 20, ;, 结果 , (RMSECV) ; , RMSECV , , ; , 结论,, 10%,,,,,, 关键词 : ; ; ; ; Nutrients components forecast on free-range and broiler chicken based on near infrared spectroscopy XING Su-Xia *, GUO Pei-Yuan, XIANG Ling-Zi, LIANG Chao-Qun (Beijing Technology and Business University, Beijing , China) ABSTRACT: Objective To build a fast predictive model about protein and fat content of chicken and broilers. Methods Thirty fresh samples from both chicken and broilers were collected separately and 2 quantitative analysis models were built for determination of protein and fat content by using 20 samples of each. The remaining 20 samples were analyzed by predictive analytics and error analytics. Results The correlation coefficients of protein model with 2 kinds of chicken were and when RMSECV were and in the chicken and broilers protein model, respectively. As for the fat model, the correlation coefficients of 2 kinds of chicken were and 0.952, and RMSECV were and 0.149, respectively. It turned out to be that the mean errors of predicted and actual outcomes were and 0.214, the standard deviations were and in protein s case, respectively. And mean errors of predicated and actual outcomes were and 0.149, the standard deviations were and in fat s case, respectively. Conclusion From the 基金项目 : ( )( ) Fund: Supported by the National Natural Science Foundation of China ( ), Beijing Natural Science Foundation of China ( ) * 通讯作者 :,,, xingsuxia@163.com *Corresponding author: XING Su-Xia, Associate Professor, Ph.D, Beijing Technology and Business University, Beijing , China. xingsuxia@163.com

2 8, : 2995 compared results, the difference between predicted and actual results was not significant, and the standard deviation was 10% or lower. In addition, the larger the correlation coefficient of the prediction model, the more accurate the prediction results. At the same time, it showed that fast predictive models based on NIR analysis technique and the least square method can provide efficient feasible and accurate approaches to the identification of chicken and broilers. Then, it is necessary to collect more samples to built the prediction model for improve the accuracy of forecasting results. KEY WORDS: chicken quality; near infrared spectroscopy; interval partial least squares; fat; protein 1 引言,,,, [1,2],,,,, [3-5],,,,,, [6,7] [8] 148,, (RMSECV) , [9],, , 0.8 [10] 200,,(PLS), [11], 100%,,,, 2 材料与方法 2.1 材料与试剂,, 30 30, 30 20, 10,, min 2.2 仪器与设备 Foodscan ( FoodScan-Foss ) 2.3 蛋白质 脂肪含量测定理化试验 GB GB/T , 2.4 光谱数据采集与预处理 1, 850~1050 nm, 2 cm -1, 2 mm,,, 6 cm Win ISI, MatLab 2012b,,, 25

3 2996 6,, 6 cm Foodscan, 2,, 1 Foodscan Fig. 1 Foodscan near infrared reflection (NIR) total grating transmission spectrum analyzer 2 Fig. 2 The chicken sample with the NIR spectrum analyzer (multiplicative scatter correlated rection, MSC),, 2.5 区间优化偏最小二乘 (ipls) 预测模型, () ( ) 2 ( ), [14] X Y, X Y (),,,,,,, (ipls) ipls 3 1),, RMSECV ; 2) n, n PLS, RMSECV ; 3) n RMSECV, (RMSECV ); 4), [15] n, , n, PLS, PLS, (RMSECV) PLS RMSECV,,, : n yˆ c yc c 1 ( ) RMSECV n : n, y c c ; yˆc c 3 结果与讨论 3.1 理化结果 1,,,, ;,, 2

4 8, : ipls Fig. 3 Flow chart of ipls model 3.2 最佳建模区间实验结果 (MSC) 4,,,, 2 3 RMSECV, RMSECV, RMSECV [15] 2, 16,, (best local model range, BLMR) 4, (optimal local model principal fraction, OLMPF) 8 18,, BLMR 6, OLMPF 4 RMSECV, ipls 表 1 鸡肉样本脂肪和蛋白质含量理化测定结果 Table 1 The results of chicken samples fat and PR (%) (%) Fig. 4 Spectral map of chicken sample after MSC

5 Table 2 表 2 土鸡 肉鸡的脂肪含量预测模型最佳局部模型结果表 The best local model result of the fat content forecast model of chicken and broiler RMSECV RMSECV Table 3 表 3 土鸡 肉鸡的蛋白质含量预测模型最佳局部模型结果表 The best local model result of PR content forecast model of chicken and broiler RMSECV RMSECV , 14,, 8; 16,, 8 RMSECV, ipls ipls ipls 预测结果与分析,,, 5 6, 5, (2 ), ipls , RMSECV Fig. 5 Comparison of PR forecast and measured value of free-ranged chicken. 6 Fig. 6 Comparison of PR forecast and measured value of broiler.

6 8, : ,,, r , RMSECV [8], RMSECV [8], 3.4 预测实验结果与讨论 ipls,, 4, 5, ipls, , ;,, , Fig. 7 7 Comparison of PR forecast and measured value of free-ranged chicken 8 Fig. 8 Comparison of PR forecast and measured value of broiler Table 4 表 4 土鸡 肉鸡蛋白质含量模型预测结果对比实际结果 Comparison of PR forecast and measured value of chicken and broiler Table 5 表 5 土鸡 肉鸡脂肪含量模型预测结果对比实际结果 Comparison of fat forecast and measured value of chicken and broiler

7 结论 PLS, , RMSECV ; , RMSECV , ; , , 参考文献 [1],. [J]., 2005, (8): Song ZJ, Zhao GX. Chicken quality and nutrition regulation [J]. Anim Nutr, 2005, (8): [2]. [D]. :, Yuan HG. Identification of chicken broth aroma component and the effects of age, gender and carcass part to flavor of chicken [D]. Nanjing: Nanjing Agricultural University, [3],,. [J]., 2006, (8): Jiang LL, Liu HG, Qi DS. Research progress of fatty acids in chicken [J]. Meat Res, 2006, (8): [4]. [J]., 2013, (1): 1 2. Guo L. Asian chicken production will account for 1/3 of global output [J]. China Anim Husbandry Vet Med, 2013, (1): 1 2. [5]. [J]., 2010, (19): 8 9. Tian ZF. Analysis of broiler industry in China [J]. Agric Technol Equip, 2010, (19): 8 9. [6] Prieto N, Roehe R, Lavin P, et al. Application of near infrared reflectance spectroscopy to predict meat products quality: a review [J]. Meat Sci, 2009, 83: [7] Prieto N, Roehe R, Lavin P, et al. Application of near infrared reflectance spectroscopy to predict meat products quality: areview [J]. Meat Sci, 2009, 83: [8], 旻,,. [J]., 2009, 31(2): Liu W, Wu HW, Sun DD, et al. Fast determination of fresh chicken using FT-NIRspectroscopy [J]. China Poultry, 2009, 31(2): [9]. X [D]. :, Shen J. Poultry meat quality detection by X ray and based on near infrared spectroscopy [D]. Nan Chang: Jiangxi Agricultural University, [10],,,. [C].. :,, 2012: Yan CJ, Li F, Wang L, et al. Study on the chicken quality index by near infrared spectroscopy [C]. Animal Nutrition Symposium of Animal Nutrition Society of China animal husbandry and Veterinary Society eleventh. Chang Sa: China animal husbandry and Veterinary Society. Animal nutrition branch, 2012: [11],. [J]., 2014, 32(6): Xiang LZ, Guo PY. Research on classifying chicken based on near infrared spectroscopy [J]. J Food Sci Technol, 2014, 32(6): [12]. NITS [D]. :, Wang L. Research on the content detection and variety identification of liquid milk based on NITS [D]. Yinchuan: Ningxia University, [13] Douglas F, Barbin, Gamal El M, et al. Non-destructive determination of chemical composition in intact and minced pork using near-infrared hyperspectral imaging [J]. Food Chem, 2013, (2 3): [14] Li YK, Jing JA consensus PLS method based on diverse wavelength variables models for analysis of near-infrared spectra [J]. Chemometr Intell Lab, 2014, 130: [15],.

8 8, : 3001 [J]., 2011, (1): 作者简介 Huang XW, Wang KL. Nondestructive detection of total acid content in vinegar by NIR combined with partial least squares method [J].Chin Cond, 2011, (1): 邢素霞, 副教授, 主要研究方向为光电信号检测 图像处理与模式识别 xingsuxia@163.com ( 责任编辑 : 李振飞 ) 食品安全监管 专题征稿,,,,,,,,,,,,,,,,,, ,, 投稿方式 : : jfoodsq@126.com 食品安全质量检测学报 编辑部