A method for identifying organic infant rice and common infant rice based on visible-near infrared spectroscopy technology

By combining visible-near-infrared spectroscopy with centrifugal extraction supernatant and an artificial intelligence model, the problem of distinguishing between organic and ordinary infant rice cereal has been solved, achieving efficient, accurate, and low-cost identification results.

CN122238245APending Publication Date: 2026-06-19NANJING TECH UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
NANJING TECH UNIV
Filing Date
2026-04-01
Publication Date
2026-06-19

AI Technical Summary

Technical Problem

Existing technologies make it difficult to quickly and easily distinguish between organic and regular infant rice cereal. Traditional testing methods are costly, time-consuming, and highly subjective, lacking convenient and reliable identification technology support.

Method used

Visible-near-infrared spectroscopy combined with centrifugal extraction of rice paste supernatant was used. A discrimination model was established by spectral acquisition and artificial intelligence models (PLS-DA, SVM-DA, XGBoost-DA) to identify the rice paste and supernatant by utilizing the difference in transmittance spectra.

Benefits of technology

It achieves efficient and accurate identification of organic and regular infant rice cereal, with low cost, speed, and no damage to the sample structure, and the identification accuracy rate reaches 83.33%~94.93%.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122238245A_ABST
    Figure CN122238245A_ABST
Patent Text Reader

Abstract

This invention discloses a method for identifying organic and conventional infant rice cereal based on visible-near-infrared spectroscopy, belonging to the field of spectral analysis technology. This method utilizes the differences in chemical composition and microstructure between organic and conventional infant rice cereal due to variations in planting and processing, resulting in different near-infrared light absorption characteristics in the rice paste and supernatant states, thus producing measurable differences in transmittance spectra. By collecting the transmittance spectra of the rice paste and supernatant of the two types of samples, spectral features are extracted, and a correlation model between near-infrared spectral information and sample category differences is constructed to achieve rapid prediction and classification of unknown samples. Results show that the SVM-DA model constructed based on rice paste spectral data has the highest prediction accuracy, reaching 90%; the XGB-DA model constructed based on supernatant spectral data has the highest prediction accuracy, reaching 97.83%. This invention has the advantages of simple operation and rapid detection, enabling efficient identification of organic infant rice cereal.
Need to check novelty before this filing date? Find Prior Art