食品安全质量检测学报
食品安全質量檢測學報
식품안전질량검측학보
FOOD SAFETY AND QUALITY DETECTION TECHNOLOGY
2015年
8期
3014-3020
,共7页
马铃薯%抗性淀粉%可见/近红外反射光谱%逐步回归%多元线性回归
馬鈴藷%抗性澱粉%可見/近紅外反射光譜%逐步迴歸%多元線性迴歸
마령서%항성정분%가견/근홍외반사광보%축보회귀%다원선성회귀
potato%resistant starch%visual/near infrared reflective spectrum%step-wise discrimination%multiple linear regressions
目的：利用可见/近红外反射光谱技术无损检测新鲜马铃薯茎块中抗性淀粉的含量。方法使用光谱仪获取新鲜马铃薯在345~1100 nm波段范围内的漫反射光谱；分别使用Savitzky–Golay(S-G)平滑处理、多元散射校正(MSC)法和一阶导数法(1st-D)对反射光谱进行预处理；对(S-G)反射光谱、MSC处理光谱和1st-D光谱使用逐步回归法判别法选择最优波长组合,建立多元线性回归模型,使用全交叉验证法验证模型。结果结果表明,可见/近红外反射光谱经过一阶导数处理后,确定的8个最优波长(370、569、576、866、868、886、922和963 nm)组合建立模型的校正和验证结果最好：模型的校正结果为相关系数 R=0.996,标准差 SEC=0.521%；模型交叉验证相关系数 Rcv=0.982,验证标准差 SECV=0.791%。结论可见/近红外反射光谱技术可以较好地预测新鲜马铃薯茎块的抗性淀粉含量,本研究可为可见近红外光谱技术在马铃薯功能成分的快速检测提供一定的技术基础。
目的：利用可見/近紅外反射光譜技術無損檢測新鮮馬鈴藷莖塊中抗性澱粉的含量。方法使用光譜儀穫取新鮮馬鈴藷在345~1100 nm波段範圍內的漫反射光譜；分彆使用Savitzky–Golay(S-G)平滑處理、多元散射校正(MSC)法和一階導數法(1st-D)對反射光譜進行預處理；對(S-G)反射光譜、MSC處理光譜和1st-D光譜使用逐步迴歸法判彆法選擇最優波長組閤,建立多元線性迴歸模型,使用全交扠驗證法驗證模型。結果結果錶明,可見/近紅外反射光譜經過一階導數處理後,確定的8箇最優波長(370、569、576、866、868、886、922和963 nm)組閤建立模型的校正和驗證結果最好：模型的校正結果為相關繫數 R=0.996,標準差 SEC=0.521%；模型交扠驗證相關繫數 Rcv=0.982,驗證標準差 SECV=0.791%。結論可見/近紅外反射光譜技術可以較好地預測新鮮馬鈴藷莖塊的抗性澱粉含量,本研究可為可見近紅外光譜技術在馬鈴藷功能成分的快速檢測提供一定的技術基礎。
목적：이용가견/근홍외반사광보기술무손검측신선마령서경괴중항성정분적함량。방법사용광보의획취신선마령서재345~1100 nm파단범위내적만반사광보；분별사용Savitzky–Golay(S-G)평활처리、다원산사교정(MSC)법화일계도수법(1st-D)대반사광보진행예처리；대(S-G)반사광보、MSC처리광보화1st-D광보사용축보회귀법판별법선택최우파장조합,건립다원선성회귀모형,사용전교차험증법험증모형。결과결과표명,가견/근홍외반사광보경과일계도수처리후,학정적8개최우파장(370、569、576、866、868、886、922화963 nm)조합건립모형적교정화험증결과최호：모형적교정결과위상관계수 R=0.996,표준차 SEC=0.521%；모형교차험증상관계수 Rcv=0.982,험증표준차 SECV=0.791%。결론가견/근홍외반사광보기술가이교호지예측신선마령서경괴적항성정분함량,본연구가위가견근홍외광보기술재마령서공능성분적쾌속검측제공일정적기술기출。
Objective To determine whole potato tubers resistant starch (RS) content nondestructively by visual/near infrared reflectance (VIS/NIR) spectroscopy. Methods The VIS/NIR reflectance spectrum were collected in 345~1100 nm using spectrography. The spectrum was pretreated by using Savitzky–Golay smoothing (S-G), multiplicative scattering correction (MSC) and first derivative (1st-Der) methods, respectively. The optimal wavelength combinations from S-G, MSC and 1st-Der spectrum were selected by step-wise discrimination method and used to establish multi-linear regression (MLR) models to predict the potato tuber RS contents. Then, the full cross validation were used to examine the effect of model. Results The results showed that the revise and verification result of MLR model based on 8 optimal wavelengths (370, 569, 576, 866, 868, 886, 922 and 963 nm) was the best after 1st-Der reflectance. It showed that correlation coefficient and standard deviation of revise result was 0.996 and 0.521%, respectively, the correlation coefficient and the standard error of cross validation (SECV) of cross validation (Rcv) were 0.982 and 0.791%,respectively. Conclusion The VIS/NIR reflectance spectral technique was useful for nondestructive determination of potato RS content. And this study could provide an efficient means for the rapid and nondestructive determination of potato starch quality.
