JP2008089529A - Nondestructive measurement method of frozen ground fish meat component by near infrared analysis - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a method for simply and accurately measuring moisture and protein in frozen ground fish meat. <P>SOLUTION: In this near infrared analysis method of the ground fish meat, the ground fish meat to be measured is frozen, and a raw spectrum is subjected to WSC processing by using a fiber probe at a temperature below -5°C in the frozen state of the ground fish meat, and its secondary differential processing spectrum is subjected to PLS regression analysis. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a method for measuring components such as proteins contained in frozen surimi.

The quality of frozen surimi is measured in accordance with the quality inspection standards specified by the Fisheries Agency (Fisheries Agency notification, 6 water fishery No. 1065, April 1, 1994). Water content of frozen surimi is an essential test item, and protein is an optional measurement. It is an item. This is because the water and protein content is an important factor that directly affects the gel-forming ability, which is the most important quality characteristic of surimi. Quality inspection standards that are almost the same as the standards are established.
In the CODEX standard, the protein measurement method is the Kjeldahl method, and the total nitrogen content is obtained by the Kjeldahl method and multiplied by the "nitrogen-protein conversion factor (6.25 for fish)" The amount of crude protein is calculated.
The Kjeldahl method is a highly versatile method with high analytical accuracy, but the analysis test is complicated, and it takes time to measure, so it can be measured quickly and accurately at the manufacturing site where quickness and simplicity are required. New methods have been demanded.

  In recent years, near-infrared spectroscopy has attracted attention as a rapid component analysis method, and methods for measuring protein content by such analysis methods have been proposed (Patent Documents 1 to 3). Among these, the methods described in Patent Documents 1 and 2 are intended for cereals such as wheat flour having a relatively low water content, and for these objects, there is no interference in the absorption spectrum due to the presence of water. Although the peak could be observed in the low wavelength region (570-1120 nm) (Patent Document 2), it was not a method suitable for seafood with a high water content. In Patent Document 3, the flour dough containing a relatively large amount of water is observed in the high wavelength region (1100-2500 nm), but the fiber probe (low wavelength of 1100 nm or less for the attenuation of near infrared light). It was not applicable to the measurement of large block products that had to be used only for area observation.

  The present inventors analyzed water and protein in surimi components in order to investigate a rapid protein measurement method by near-infrared spectroscopic analysis by an interactance method using a fiber probe in a low wavelength region of 1100 nm or less. Tried (Non-Patent Document 1). The wavelength range when using a fiber probe is 400-1100 nm, but in this range, a peak indicating water is recognized around 970 nm, and from the influence, the protein itself to be observed around 910 nm The peak could not be confirmed, and the protein amount could only be estimated by the inverse correlation from the water content.

On the other hand, methods for measuring the fat content of frozen bonito and frozen jack mackerel by near infrared spectroscopy equipped with an optical fiber probe have been proposed (Non-Patent Documents 2 and 3). These methods can accurately measure the fat content of frozen fish, 0-5mm subcutaneously, and are suitable for sorting fish that contain a lot of fat near the epidermis, but are measured by the spectrum of water generated by thawing. It has been reported that errors occur (Non-Patent Document 2) and that the frozen sample is inferior to the raw sample in the evaluation of the calibration curve of the raw / frozen sample (Non-Patent Document 3). It did not suggest application.
Thus, there has been no known method for accurately measuring the moisture and protein content in the frozen surimi as it is.
Japanese Patent Application Laid-Open No. 560685 Japanese translation of PCT publication No. 2003-526079 JP-T-2001-527646 Udden M, Okazaki E, Fukushima H, Turza S, Yamashita Y, and Fukuda Y: Food Chem. 96, 491-495 (2006) Satoru Yamauchi, Toshio Sawada, Sumio Kawano, Journal of the Fisheries Society of Japan, 65 [4], 1999, Fisheries Society of Japan, 747-752 Koji Enomoto, Satoshi Hasegawa, Satoshi Ide, Sumio Kono, Journal of Japanese Society of Fisheries Science, 67 [4], 2001, Japan Society of Fisheries Science, 717-722

  An object of the present invention is to provide a method for accurately and easily measuring moisture and protein in frozen surimi.

When the surimi to be measured is frozen and the near-infrared analysis is performed while the surimi is in a frozen state, the inventors have observed a large peak that appears to indicate the presence of water near 1026 nm, and the protein near 910 nm. It was found that a peak indicating γ was clearly observed and the measurement accuracy was superior to that after thawing, and the present invention was achieved.
The protein peaks that could not be observed when unfrozen were clearly observed because the water peak that was around 970 nm moved to around 1026 nm when unfrozen. A peak is observed.

  According to the present invention, it is easy to operate because it measures moisture and protein in surimi in a frozen state, and can be measured with an optical fiber probe because it uses a wavelength of 1100 nm or less, and frozen surimi is in a large blocked state In addition, the content can be measured directly from the peak of the protein, so that highly accurate measurement is possible.

  As the frozen surimi used in the present invention, various products currently distributed all over the world are targeted. For example, in North America, Russia, and Hokkaido, walleye pollock, hockey, and pacific whiting, etc., in Southeast Asia and India, Itoyodai, Kintoki, Guchi, Ezo, etc., and in New Zealand and South America, southern bluefish, hoki, horse mackerel, etc. are mentioned.

As the near-infrared spectroscopic analysis apparatus used in the present invention, a commercially available apparatus can be used, but an apparatus equipped with a fiber probe is preferable for convenience of handling.
Although it is very difficult and practical to cut and measure a portion of frozen surimi, in the present invention, the fiber and probe are used to accurately determine the moisture and protein contents in a frozen surimi mass. It is possible to measure high.

  As statistical processing in the near-infrared analyzer, multiple regression analysis (MLR: 926nm (fish oil absorption wavelength) at the first wavelength, wavelength 2 to 4 is automatically selected by the computer in the range of 700-1100nm) And a method of MSC processing the raw spectrum, and a method of PLS regression analysis (partial least squares regression) of the second derivative spectrum, both of which can be applied in the present invention, PLS regression analysis is preferred because many peaks have multiple effects because they contain various components.

  The analysis of the present invention can be performed any number of times as long as the surimi is frozen. However, a sharp peak shift from 1024 nm (ice) to 970 nm (water) occurs between −5 ° C. and 0 ° C. In order to increase the accuracy, it is necessary to carry out at a temperature of 5 ° C. or lower, more preferably −10 ° C. or lower.

≪Creation of calibration curve≫
Various types of frozen surimi distributed in the market (Sumitou Frozen Surimi, 52 products) were used as samples. For near-infrared spectrum measurement, an interactance optical fiber probe (9 mm in diameter) attached to the near-infrared analyzer NIRSystems 6500 (NIRSystems Inc., Silver Spring, MD, USA) is contacted with the sample surface and scanned 32 times. Went. A white ceramic plate was used for reference spectrum measurement before sample measurement. Instrument operation and spectrum collection were performed using Vision software (Version 3.20, NIRSystems Inc., MD, USA). The obtained spectrum data (700-1100 nm) was analyzed by The Unscrambler software (Version 8.05, CAMO, NJ, USA). The near-infrared spectrum of the surimi measured (protein: 910 nm, moisture: 1026 nm) is shown in FIG.

  FIG. 2 is a plot of the second derivative value of the original spectrum shown in FIG. 1 and shows a water-derived peak (1026 nm) and a protein-derived peak (910 nm).

FIG. 3 is a graph showing the correlation between the protein content value obtained from these peak values using a calibration curve and the protein content value obtained by the Kjeldahl method, and R ² = 0.96. This indicates that the protein content could be measured with high accuracy.

≪Comparison of measurement accuracy≫
Near-infrared spectra of frozen and thawed surimi were measured as in Example 1. The water content and protein content as control analysis values were measured by the atmospheric pressure heat drying method and the Kjeldahl method, respectively. A calibration curve was created using the obtained near-infrared spectrum data and chemical analysis values, and the accuracy of each model was verified. The results are shown in Table 1.
As the numerical values in Table 1 indicate, the prediction accuracy of the protein and water was better in the frozen state than in the thawed state. This is considered to be because the peak wavelengths of protein and water are separated in the frozen state and do not affect each other.

In the near-infrared analysis, the near-infrared spectrum in the thawing process of surimi was measured in the same manner as in Example 1 in order to examine the change in absorbance at 970 nm (water) and 1024 nm (ice) with temperature change. Near-infrared spectra of walleye pollock frozen surimi (approx. 50 x 50 x 50 mm block) at various temperatures (from -20 ° C to 5 ° C: 45 points) while monitoring the temperature about 5 mm deep from the surface. Was measured. Thawing and measurement were performed in a polystyrene foam filled with a cold insulation material. FIG. 4 shows the results of plotting the values of 970 nm (water) and 1024 nm (ice) after second-order differentiation of the obtained spectrum data.
As is clear from FIG. 4, it can be seen that the absorbance changes abruptly between −5 ° C. and 0 ° C.
Therefore, in order to carry out the present invention, it is desirable to carry out at -5 ° C or lower, more preferably -10 ° C or lower.

  According to the present invention, real-time grading of frozen surimi is possible, and production of frozen surimi with more stable quality can be realized. Moreover, the quality control of the raw material before manufacture can be strictly performed, which is very useful for improving the quality of processed products.

Water and protein spectra of surimi A plot of the differential value of the original spectrum shown in FIG. Graph showing the correlation between the protein content value obtained from the peak value using the calibration curve and the protein content value obtained by the Kjeldahl method Graph showing changes in absorbance at 970nm (water) and 1024nm (ice) with temperature change

Claims (4)

  A near-infrared analysis method characterized by measuring moisture and protein in surimi produced from various seafood in a frozen state.   The near-infrared analysis method according to claim 1, wherein measurement is performed at a temperature of −5 ° C. or lower.   3. The near infrared analysis method according to claim 1, wherein a fiber probe is used.   The near-infrared analysis method according to any one of claims 1 to 3, wherein the raw spectrum is subjected to MSC processing, and the secondary differential processing spectrum is subjected to PLS regression analysis.