Method for rapidly predicting freezer storage time of frozen pork based on reflectance ratio of two near-infrared bands

Patent metadata

Field	Value
Publication number	US-11448592-B2
Application number	US-201716321042-A
Country	US
Kind code	B2
Filing date	Sep 20, 2017
Priority date	Jul 28, 2016
Publication date	Sep 20, 2022
Grant date	Sep 20, 2022

Abstract

The invention discloses a method for rapidly predicting freezer storage time of frozen pork based on reflectance ratio of two near-infrared bands. Firstly, the near-infrared spectra information of the frozen pork is obtained by a near-infrared spectrometer, and the near-infrared spectra are analyzed to obtain center values of bands centered on 1500 nm, 1350 nm and 1890 nm characteristic peaks. The ratio thereof is treated as a eigenvector, which is substituted into the characteristic exponential function based on the eigenvector and freezer storage time to calculate the freezer storage time of frozen pork. By near-infrared spectroscopy technology, the invention directly detects the freezer storage time of pork in a frozen state, significantly reduces the time required by the conventional method, and has the advantages of being fast and non-destructive.

First claim

We claim:

1. A method for rapidly predicting freezer storage time of frozen pork based on reflectance ratio of two near-infrared bands, characterized in that, it comprises the steps of:
(1) scanning frozen pork and calibrating to obtain near-infrared spectrum of the frozen pork;
(2) calculating a center value R m1350 of a band centered on a 1350 nm characteristic peak, a center value R m1980 of a band centered on a 1890 nm characteristic peak and a center value R m1500 of a band centered on a 1500 nm reference peak in the near-infrared spectrum of the frozen pork:
R m1350 =R 1348 ×0.1+ R 1349 ×0.2+ R 1350 ×0.4+ R 1351 ×0.2+ R 1352 ×0.1

R m1890 =R 1888 ×0.1+ R 1889 ×0.2+ R 1890 ×0.4+ R 1891 ×0.2+ R 1892 ×0.1

R m1500 =R 1498 ×0.1+ R 1499 ×0.2+ R 1500 ×0.4+ R 1551 ×0.2+ R 1552 ×0.1

wherein R 1348 , R 1349 , R 1350 , R 1351 , R 1352 , R 1888 , R 1889 , R 1890 , R 1891 , R 1892 , R 1498 , R 1499 , R 1500 , R 1551 , R 1552 are reflectance of the frozen pork at 1348 nm, 1349 nm, 1350 nm, 1351 nm, 1352 nm, 1888 nm, 1889 nm, 1890 nm, 1891 nm, 1892 nm, 1498 nm, 1499 nm, 1500 nm, 1501 nm, 1502 nm, respectively;
(3) calculating a standard center value S m1890 of the characteristic band centered on 1890 nm:
S m1890 =R m1890 /R m1500

(4) treating the ratio of R m1350 to S m1890 as an eigenvector I, wherein I=R m1350 /S m1890 ;
(5) substituting the eigenvector I into a characteristic exponential function t=1E+07e −97.34I which is based on the eigenvector I and freezer storage time t, so as to predict the freezer storage time of the frozen pork.

2. The method for rapidly predicting freezer storage time of frozen pork based on reflectance ratio of two near-infrared bands according to claim 1 , characterized in that, the characteristic exponential function based on the eigenvector I and freezer storage time is obtained by the following steps:
(a) picking frozen pork training samples from pork with different freezer storage time, scanning the frozen pork training samples and calibrating to obtain near-infrared spectra of the frozen pork;
(b) calculating a center value R′ m1350 of a band centered on a 1350 nm characteristic peak, a center value R′ m1980 of a band centered on a 1890 nm characteristic peak and a center value R′ m1500 of a band centered on a 1500 nm reference peak in the near-infrared spectrum of each of the frozen pork training samples:
R′ m1350 =R′ 1348 ×0.1+ R′ 1349 ×0.2+ R′ 1350 ×0.4+ R′ 1351 ×0.2+ R′ 1352 ×0.1

R′ m1890 =R′ 1888 ×0.1+ R′ 1889 ×0.2+ R′ 1890 ×0.4+ R′ 1891 ×0.2+ R′ 1892 ×0.1

R′ m1500 =R′ 1498 ×0.1+ R′ 1499 ×0.2+ R′ 1500 ×0.4+ R′ 1551 ×0.2+ R′ 1552 ×0.1

wherein R′ 1348 , R′ 1349 , R′ 1350 , R′ 1351 , R′ 1352 , R′ 1888 , R′ 1889 , R′ 1890 , R′ 1891 , R′ 1892 , R′ 1498 , R′ 1499 , R′ 1500 , R′ 1551 , R′ 1552 are reflectance of the frozen pork training samples at 1348 nm, 1349 nm, 1350 nm, 1351 nm, 1352 nm, 1888 nm, 1889 nm, 1890 nm, 1891 nm, 1892 nm, 1498 nm, 1499 nm, 1500 nm, 1501 nm, 1502 nm, respectively;
(c) calculating a standard center value S m1890 of the characteristic band centered on 1890 nm of the frozen pork training samples:
S′ m1890 =R′ m1890 /R′ m1500

(d) treating the ratio of R′ m1350 to S′ m1890 as an training eigenvector I′ of the frozen pork training samples, wherein I′=R′ m1350 /S′ m1890 ;
(e) establishing a model based on the training eigenvector the frozen pork training samples to obtain a characteristic exponential function t=1E+07e −97.34I —which is based on the training eigenvector I′ and freezer storage time t.

3. The method for rapidly predicting freezer storage time of frozen pork based on reflectance ratio of two near-infrared bands according to claim 2 , wherein the different freezer storage time of the pork are 1, 3, 6, 9, and 12 months, respectively.

4. The method for rapidly predicting freezer storage time of frozen pork based on reflectance ratio of two near-infrared bands according to claim 2 , wherein a regression coefficient of the characteristic exponential function based on the eigenvector I and the freezer storage time t is 0.9953.

5. The method for rapidly predicting freezer storage time of frozen pork based on reflectance ratio of two near-infrared bands according to claim 2 , wherein a prediction coefficient of determination of the characteristic exponential function based on the eigenvector I and the freezer storage time t is 0.9947, and a prediction root mean square error is 0.1256 month.

6. The method for rapidly predicting freezer storage time of frozen pork based on reflectance ratio of two near-infrared bands according to claim 2 , wherein said scanning the frozen pork training samples in step (a) is specifically scanning the frozen pork training samples in a frozen state;
and said scanning the frozen pork in step (1) is specifically scanning the frozen pork in a frozen state.

Assignees

• Univ South China Tech

Inventors

• Sun Dawen
• Cheng Weiwei
• Pu Hongbin
• Zhu Zhiwei
• Han Zhong

Classifications

• G01N21/359 Primary

  using near infrared light · CPC title

• G01N33/12

  Meat; Fish · CPC title

• G01N21/3563

  for analysing solids; Preparation of samples therefor · CPC title

• G01N21/59

  Transmissivity (G01N21/25 takes precedence) · CPC title