Scholarlink

To ensure food safety and quality, the detection of impurities in solid foods is of paramount importance. The research investigates the application of machine learning techniques in conjunction with spectrometry for the identification and classification of impurities in solid food samples. The spectral data were specially collected for the experiment through a thorough examination of various solid food samples using near-infrared (NIR) spectrometry. Different types of machine learning models such as Random Forest classifications, Support-Vector, Ensemble Learning algorithms and Neural Networks were leveraged for classification. The study leverages the concept of NIR spectrometry as an utilization of machine learning algorithms to maintain food safety and quality assurance. The study tries to enable real-time contamination detection and classification during food processing. The implementation of experiment has achieved a validation accuracy of about 97.85% on an in-house dataset.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Integrated Approaches to Impurity Detection and Classification in Solid Foods Using Machine Learning and Spectrometry

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Axat Patel,
  • Hetul Patel,
  • Kush Shah,
  • Mansi Solanki,
  • Hardik Jayswal,
  • Atanu Benarjee,
  • Dipika Damodar,
  • Jitendra P. Chaudhari,
  • Poojasingh Chaudhari

摘要

To ensure food safety and quality, the detection of impurities in solid foods is of paramount importance. The research investigates the application of machine learning techniques in conjunction with spectrometry for the identification and classification of impurities in solid food samples. The spectral data were specially collected for the experiment through a thorough examination of various solid food samples using near-infrared (NIR) spectrometry. Different types of machine learning models such as Random Forest classifications, Support-Vector, Ensemble Learning algorithms and Neural Networks were leveraged for classification. The study leverages the concept of NIR spectrometry as an utilization of machine learning algorithms to maintain food safety and quality assurance. The study tries to enable real-time contamination detection and classification during food processing. The implementation of experiment has achieved a validation accuracy of about 97.85% on an in-house dataset.