Scholarlink

Milk is an essential dietary component, and is vulnerable to adulteration for profit, posing serious health risks. Traditional detection methods are slow and require advanced laboratories. This study addresses these issues by combining machine learning with Near-Infrared (NIR) Spectroscopy for fast, cost-effective, and accurate detection of milk impurities. Various additives like refined sugar, flour, coffee powder, and milk powder were analyzed in liquid and solid forms using pixel sensors. Spectral data underwent preprocessing, normalization, and feature extraction before classification with machine learning algorithms such as Random Forest, Support Vector Machine (SVM), K-Nearest Neighbors (KNN), Logistic Regression, and Multi-Layer Perceptron (MLP). A dataset of 20,000 spectral samples was split into training (70%) and test (30%) sets. The wavelength range used was 400–1000 nm for detecting milk impurities. Supervised machine learning achieved high accuracy: Random Forest led with 99.849%, followed by KNN at 99.3341%, and other classifiers showed promising results.

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

Utilizing Machine Learning Techniques and Spectrometry for Milk Impurity Detection

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Pranali M. Modi,
  • Maahi M. Shah,
  • Janvi D. Bhanushali,
  • Dhruvi A. Dhulia,
  • Madhuri N. Barochiya,
  • Hardikkumar Jayswal,
  • Ritesh Patel,
  • Nilesh Dubey,
  • Dipika Damodar,
  • Jaimin Undaviya

摘要

Milk is an essential dietary component, and is vulnerable to adulteration for profit, posing serious health risks. Traditional detection methods are slow and require advanced laboratories. This study addresses these issues by combining machine learning with Near-Infrared (NIR) Spectroscopy for fast, cost-effective, and accurate detection of milk impurities. Various additives like refined sugar, flour, coffee powder, and milk powder were analyzed in liquid and solid forms using pixel sensors. Spectral data underwent preprocessing, normalization, and feature extraction before classification with machine learning algorithms such as Random Forest, Support Vector Machine (SVM), K-Nearest Neighbors (KNN), Logistic Regression, and Multi-Layer Perceptron (MLP). A dataset of 20,000 spectral samples was split into training (70%) and test (30%) sets. The wavelength range used was 400–1000 nm for detecting milk impurities. Supervised machine learning achieved high accuracy: Random Forest led with 99.849%, followed by KNN at 99.3341%, and other classifiers showed promising results.