Identification of Rice from Different Origins by Two-dimensional Correlation Near Infrared Spectroscopy
CSTR:
Author:
Affiliation:

(1.Shanghai Institute of Technical Physics, Chinese Academy of Sciences, Shanghai 200083;2.Taicang Institute of Optoelectronics Technology, Suzhou 215411, Jiangsu;3.University of Chinese Academy of Sciences, Beijing 100049)

  • Article
  • | |
  • Metrics
  • |
  • Reference
  • |
  • Related
  • |
  • Cited by
  • | |
  • Comments
    Abstract:

    In order to realize the nondestructive discrimination of rice samples from different origins, a discrimination method based on synchronous two-dimensional correlation near infrared spectroscopy and Euclidean distance was proposed. After obtaining the standard two-dimensional correlation spectra of five types of rice samples respectively, the Euclidean distance between the test set samples and each standard correlation spectrum is calculated, and the discrimination is based on the minimum distance. The accuracy of this method can reach 95%. By defining the discrimination coefficient, the discrimination effect of various types of samples can also be judged more accurately. In order to compare the practical effects of two-dimensional correlation method and common modeling methods, partial least squares discriminant analysis and support vector machine classification are tried on the basis of the same data. The highest accuracy rates are 75% and 70% respectively, which verifies the effectiveness of two-dimensional correlation method. This study provides a possible method for nondestructive identification of rice samples, and can provide a reference for on-line detection.

    Reference
    Related
    Cited by
Get Citation
Share
Article Metrics
  • Abstract:
  • PDF:
  • HTML:
  • Cited by:
History
  • Received:September 23,2022
  • Online: November 22,2023
Article QR Code
Copyright :Journal of Chinese Institute of Food Science and Technology     京ICP备09084417号-4
Address :9/F, No. 8 North 3rd Street, Fucheng Road, Haidian District, Beijing, China      Postal code :100048
Telephone :010-65223596 65265376      E-mail :chinaspxb@vip.163.com
Supported by : Beijing E-Tiller Technology Development Co., Ltd.
Firefox, Chrome, IE10, IE11 are recommended. Other browsers are not recommended.