Scholarlink

<p>Aflatoxin B1 is a major contaminant in edible oils, posing a significant threat to food safety. In this study, an olfactory visualization system based on a colorimetric sensor array (CSA) was developed to rapidly detect aflatoxin B1 contamination in peanut oil. Nine different chemical dyes were selected to extract volatile organic compounds (VOCs) from peanut oil samples, and image data were then processed using a machine vision algorithm. Several classification models, including linear discriminant analysis (LDA), random forest (RF), and support vector machine (SVM), were evaluated, with SVM demonstrating the best performance. To further enhance the prediction accuracy, swarm intelligence optimization algorithms—specifically the Sparrow Search Algorithm (SSA), Non-dominated Sorting Genetic Algorithm II (NSGA-II), and Whale Optimization Algorithm (WOA)—were employed to optimize the selection of color feature variables. Among these, the WOA-SVM model achieved the highest accuracy, reaching 95.83% on the test set. These findings suggest that the integration of swarm intelligence optimization with the colorimetric sensor array (CSA) significantly enhances the detection accuracy of aflatoxin B1 in peanut oil.</p>

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

Combining Colorimetric Sensor Technology with Swarm Intelligence Feature Optimization Algorithm to Realize High Precision Identification of Aflatoxin-B1 Content in Peanut Oil

  • 期刊论文

  • 发表日期：

    2025年6月28日

查看全文
  • Congli Mei,
  • Jingwen Zhu,
  • Wencheng Zhu,
  • Huazhi Wang,
  • Hui Jiang

摘要

Aflatoxin B1 is a major contaminant in edible oils, posing a significant threat to food safety. In this study, an olfactory visualization system based on a colorimetric sensor array (CSA) was developed to rapidly detect aflatoxin B1 contamination in peanut oil. Nine different chemical dyes were selected to extract volatile organic compounds (VOCs) from peanut oil samples, and image data were then processed using a machine vision algorithm. Several classification models, including linear discriminant analysis (LDA), random forest (RF), and support vector machine (SVM), were evaluated, with SVM demonstrating the best performance. To further enhance the prediction accuracy, swarm intelligence optimization algorithms—specifically the Sparrow Search Algorithm (SSA), Non-dominated Sorting Genetic Algorithm II (NSGA-II), and Whale Optimization Algorithm (WOA)—were employed to optimize the selection of color feature variables. Among these, the WOA-SVM model achieved the highest accuracy, reaching 95.83% on the test set. These findings suggest that the integration of swarm intelligence optimization with the colorimetric sensor array (CSA) significantly enhances the detection accuracy of aflatoxin B1 in peanut oil.