Scholarlink

<p>This study develops a non-invasive, automated odor collection system to address the need for enhanced freshness assessment of grapes during storage and transportation. The system, designed to simulate a ventilated storage environment, minimizes human interference and maximizes the reliability of E-nose technology in evaluating grape freshness. It enables continuous monitoring of volatile organic compounds (VOCs), with preliminary E-nose data analysis indicating that sensor response patterns are associated with grape freshness and spoilage levels. Additionally, the application of Inflection Point Detection helps classify stages of grape deterioration. A preliminary assessment of several machine-learning models, enhanced by data augmentation techniques, was performed to evaluate their effectiveness in distinguishing between grape varieties. These results show promise for advancing agricultural storage monitoring through integrated data fusion and sophisticated algorithmic analysis. Further research is required to verify the robustness and precision of these methods. This work offers significant insights into improving quality control practices in the agricultural sector, potentially transforming handling processes with its innovative approach to storage environment simulation and data integrity assurance. </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

Exploring cross-variety fruit spoilage monitoring methods based on electronic nose: taking grapes as examples

  • 期刊论文

  • 发表日期：

    2025年5月20日

查看全文
  • Hongli Yang,
  • Haochen Huang,
  • Yuchi Sun,
  • Gang Yu,
  • Qi Lu,
  • Jiawei Yang,
  • Yingqing Xu

摘要

This study develops a non-invasive, automated odor collection system to address the need for enhanced freshness assessment of grapes during storage and transportation. The system, designed to simulate a ventilated storage environment, minimizes human interference and maximizes the reliability of E-nose technology in evaluating grape freshness. It enables continuous monitoring of volatile organic compounds (VOCs), with preliminary E-nose data analysis indicating that sensor response patterns are associated with grape freshness and spoilage levels. Additionally, the application of Inflection Point Detection helps classify stages of grape deterioration. A preliminary assessment of several machine-learning models, enhanced by data augmentation techniques, was performed to evaluate their effectiveness in distinguishing between grape varieties. These results show promise for advancing agricultural storage monitoring through integrated data fusion and sophisticated algorithmic analysis. Further research is required to verify the robustness and precision of these methods. This work offers significant insights into improving quality control practices in the agricultural sector, potentially transforming handling processes with its innovative approach to storage environment simulation and data integrity assurance.