Journal of the Korean Chemical Society 2005, Vol. 49, No. 6 Printed in the Republic of Korea 근적외선분광법을이용한새로운비파괴적계란신선도측정방법의개발에관한연구 ½ y*á Á Á Á z Á p Á Á½z ƒ rp lj w w (2005. 7. 5 ) Non-Destructive Measurement of the Egg Freshness by Near Infrared Spectrometry Sang Ho Kim*, Sang Jin Lee, Duk Su Lee, Won Bo Cho, Seong Hun Lee, Stuart Borden, Young A Woo, and Hyo Jin Kim Poultry Science Division, National Livestook Research Institute, Suwon 441-706, Korea Institute of Development, Spectrontech Co., Ltd. Department of Pharmacy, Dongduk Women s University, Seoul 136-714, Korea (Received July 5, 2005). x d w w w y (HU) d w ù,» w w š q d w Ÿ w. Ÿ d w 70 HU 100 HU rp d w, PLSR(Partial Least Square Regression) w w. (R)ƒ 0.99, 1.35 HU d ƒ w.» t sƒ w y d Ÿ q t d w, ª t d ƒ w w. : Ÿw, q, z š, t, y ABSTRACT. The objective of this study was to develop a non-destructive, rapid and accurate analytical method to measure the quality(freshness) of egg. Near-infrared(NIR) spectra of 48 egg samples ranging from 70 HU(Haugh Unit) from 100 HU, PLSR(Partial Least Squares Regression) was used to build calibration model using the multiplicative scatter correction spectra in the 1100~1750 nm range. The results were very promising with correlation coefficient of 0.99 and standard error of prediction(sep) of 1.35 HU. The overall results suggest that NIR spectroscopy can be used for rapid non-invasive determination of egg quality. Keywords: NIR Spectroscopy, Nondestructive Analysis, PLSR, Egg Quality, Haugh Unit w w s š m» ¾ t š. t» w ƒ, t œ ƒ wš. w t 531
532 ½ yá Á Á Á zá p Á Á½z ù ù d w y (Haugh Unit, HU) tx. 1,2. HU=100 log (H 1.7W 0.37 + 76) (1) HU=Haugh unit H= (mm) W=ù (g) HU HU ƒ š û ƒ û ùk ù. HU 70 w HU 70 w w q w. w HU q w ù ù d w» d w w w. q ƒ» d w. x ü w 1000 w mw m š ù d w 0.01%, HU w ù l 70 w HU š. HU w 25 3 70 w yƒ. ù þ m w m w ü þ w œ HU 30 w wš. q w d w v. Ÿ t C-H, N- H, O-H»ƒ 800~2500 nm q ƒ w w w Chemometricsƒ w» ¾ vj š y ù, Karl Norris w, z y š. 3 š, t,,, k y w š. w w 4 w,, yw, t, š š. Ÿ 5,6 w ƒ œ ù ¾ œ s š. Karen Ÿ w, g,, e ƒ w. 7 Scotter w Ÿ w. 8 Gerard Ÿ fv w. 9 x ù ù 10» t t y w» w Ÿ» wš. t sƒ w ù d w» w Ÿ w d w d w. w ƒ w, ù 95% š, ù ƒ w š j» w. d ƒ. t d w q rp d ww» w Fig. 1 l w. rp d w» w NIR (1100~1750 nm) InGaAs diode-array» HN 1100(Spectrontech, Korea) w. Ÿ 0.5 W halogen lamp w š, Ÿ d w w š d w w. d y xk(diffusereflectance). Fig. 2 y xk d d. x s š ó x Journal of the Korean Chemical Society
2005, Vol. 49, No. 6 Ÿ w q d w 533 Fig. 1. Schematic diagram of Near Infrared Spectrophotometer for the quality evaluation of eggs. Fig. 2. 3D-drawing of module for determination of quality characteristics of egg. s» s d w d w ó d yw w w. y rp yƒ w» w Ÿ w rp y w. d black- anodizing w y w, Ÿ y w. Ÿ SMA w š Ÿ HN1100 spectrometer w. d w» š xk Teflon PE(Polyethylene) ƒƒ w. l ful w, l w PLS(Partial Least Squares) ChemoHN 1100(Spectrontech, Korea) w. rp j» w z» w rp multiplicative scatter correction(msc) 1 w PLS w 11 w. ù. y (HU) d t p y (HU) d t x»(egg MULTI Tester: EMT-5200, Hikari Technology Ltd., 15-26 Abeno-Motomachi, Abeno-Ku, Osaka, 545-0034, Japan) w d š,, ù, ùy» d w, yw» w. š ü t y rp Fig. 3 d w rp. ü ùy ù, p ù. q 2062 l 2353 nm ùkù, overtone 1100 1750 nm 12,13,14 ƒ ùkù.» rp Teflon w d š, rp 0.2 AU 0.6AU ùkù. ùkù rp 1400 1500 nm ƒ ùkù. O-H band vj. š 1600 nm vjƒ ùkù. Fig. 3. NIR spectrum of egg using teflon reference.
534 ½ yá Á Á Á zá p Á Á½z Fig. 4. NIR spectrum of egg using PE reference. Fig. 5. Diagram for PRESS of HU in quality characteristics of egg using diffuse reflectance mode by PE reference at 1100 to 1750 nm. š Fig. 4» PE(Polyethylene) w, rp 0.4 AU l 1.0 AU ùkù. Ÿ»»ƒ w ùkù» reference» 10,000 l 15,000 count w. rp p» PE p w rp ùkû. Teflon» w d w PE» d w p q y ƒ. Teflon rp y ƒ PE p q p q» l ƒ y ùk ù. d» rp 1100 nm l 1750 nm ¾ 2 nm d w log 1/Txk Ÿ ùkü rp. rp vj yƒ j» yƒ. w Ÿ l w z w d w. 80 w w. x Ÿ» w w y ƒ ƒ PLS w w. PLS regression z w factor w» w cross-validation ww. Cross-validation data data validation data ù v w ƒ j PRESS(Prediction Residual Error Sum of Squares) y w w. Fig. 5 Factor PRESS y š. Fig. 5 ƒ 12 w ƒ yw.» 12 w PLS w 80 HU l 100 HU¾ Fig. 6 (R, correlation coefficient) 0.99 š, SEC(Standard Error of Calibration) 0.7 HU yw. SEC Ÿ w f(residual) l. Fig. 6. Calibration plot of HU in quality characteristics of egg by PLSR using NIR. Journal of the Korean Chemical Society
Ÿ w q d w 535 Fig. 7. Validation plots of HU in quality characteristics of egg by PLSR using NIR. Fig. 8. Average spectrum of NIR spectrum and loading value for quanti-tative model by PLSR using NIR. SEC = N š, K q ù factor. Fig. 6 w w w. Fig. 7 Validation set w w, x w y Ÿ w w. SEP (Standard Erro of Prediction) w. SEP = n i = 1 f 2 ------------------- N K 1 ( C actual C estimated ) 2 --------------------------------------------- n Cactual t ü t y š, C estimated Ÿ y. Fig. 8 PLSR loading NIR rp s³ r p. ƒ factor w loading 1350 nm 1600 nm rp ùkù w OH wš. rp 1400 nm w w w ³e ƒ w k y ƒ k w w. 15 1550 nm p w w š w NH stretching mode first overtone w. factor 5 loading wš. mw rp HU y rp p wš y w. Ÿ w d w ü t d w. rp rp y w, w» w PLS regression w, w d w SEP 1.35 HU. w» t w q x t x» w Ÿ w d w 80 œ w 5% ùk ü.» q x t x» w y w Ÿ q w x œ on-line at-line d wš z t. 2005, Vol. 49, No. 6
536 ½ yá Á Á Á zá p Á Á½z x 1. Arafa,GA.GS.,GR.GH. Harms,GR.GD. Miles, R. B. Christmas and J. H. Choi., Quality characteristics of eggs from different strains of hens as related to time of Oviposition, Poult, Sci., 1982, 61, 842-847. 2. Eisen, E. J. and B. B. Bohern., Some problems in the evaluation of egg albumen quality, Poult, Sci., 1963, 42, 74-83. 3. Williams, Phil and Norris, Karl: Near-Infrared Technology in the Agricultural and Food Industry, American Association of Cereal Chemists, Inc., Minnesota, U.S.A., 1987; p. 201. 4. Panford, J. A., Williams, P. C., and deman J. M., Analysis of Oilseeds for Protein, Oil, Fiber and Moisture by Near-Infrared Reflectance Spectroscopy, JAOCS. 1988, 65, 1627. 5. Jasper, W. J. and Kovacs, E. T., Using Neutral Networks and NIR Spectrophotometry to Identify Fibers, Textile Res. J. 1994, 64, 444. 6. Kim, Y. B., Studies on the Chemical Analysis in Raw Meat and Meat Product by Near-Infrared Spectroscopy, Seoul Nation University, Ph. D. thesis, 1996. 7. Bewig, K. M., and Unkesbay, Nan: Discrimination Analysis of Vegetable Oils by Near-Infrared Reflectance Spectroscopy, JAOCS. 1994, 71, 195. 8. Murray, I. and Cowa, I. A., Making Light Work: Advances in Near Infrared Spectroscopy, Weinheim, New York, U.S.A., 1991; p. 395. 9. Davies, A. M. C. and Williams, P., Near Infrared Spectroscopy: The Future Waves, NIR Publications, U.K., 1995; p. 411. 10. Gemperline, P. J., Webber L. D., Cox, F. O., Raw Materials Testing Using Soft Independent Modeling of Class Analogy Analysis of Near-Infrared Reflectance Spectra, Anal. Chem. 1989, 61, 138. 11. Beebe, Kenneth R. and Kowalski, Bruce R., An Introduction to Multivarate Calibration and Analysis, Anal. Chem. 1987, 59, 1007. 12. McShame, W. J., Applying NIR to Process Control, JAOCS 1989, 66, 641. 13. J. W. Hall, A. Pollard., Near-Infrared Spectrophotometry, Clin. Chem. 1992, 38, 1623-1631. 14. J. W. Hall, A. Pollard., Near-Infrared Spectroscopic Determination of Serum Total Proteins, Albumin, Globulins, and Urea, Clin. Biochem, 1993, 26, 483-490. 15. K. Murayama, Y. Ozaki., Two-dimensional near-ir correlation spectroscopy study of molten globule-like state of ovalbumin in acidic ph region: Simultaneous changes in hydration and secondary structure, Biospectroscopy, 2002, 67, 394-405. Journal of the Korean Chemical Society