Scholarlink

<p>Accurate quantification of edible-oil adulteration is constrained by an open-world problem: adulterants encountered at deployment often differ from those seen during training. Here we show that olive-oil volume fraction can be reliably recovered from excitation-emission matrix (EEM) fluorescence spectra even when the adulterant is entirely unseen during training. We propose an optimal-transport (OT) framework that aligns source- and target-domain spectral distributions in an unsupervised manner, requiring no target-domain concentration labels. Under a stringent leave-one-adulterant-out protocol across five adulterant systems, the OT-aligned support vector regression model achieves a mean <i>R</i><sup>2</sup> of 0.965 (s.d. = 0.004), substantially outperforming raw spectra (<i>R</i><sup>2</sup> = 0.831), PCA (<i>R</i><sup>2</sup> = 0.693) and PARAFAC (<i>R</i><sup>2</sup> = 0.659). These results indicate that olive-oil concentration is encoded as an intrinsic geometric coordinate within the spectral distribution, recoverable by optimal transport without adulterant-specific supervision.</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

Open-world quantification of olive-oil adulteration from fluorescence spectra via optimal-transport alignment

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年8月5日

查看全文
  • Ruopu Sun,
  • Qingke Li,
  • Tongwei Wang,
  • Zhanjun Li,
  • Hui Liu,
  • Juanjuan Gao,
  • Xiaodong Huang,
  • Yan Wang,
  • Xinmin Fan,
  • Jin Gao,
  • Christopher Q. Lan,
  • Chunyan Wang,
  • Lujun Zhang

摘要

Accurate quantification of edible-oil adulteration is constrained by an open-world problem: adulterants encountered at deployment often differ from those seen during training. Here we show that olive-oil volume fraction can be reliably recovered from excitation-emission matrix (EEM) fluorescence spectra even when the adulterant is entirely unseen during training. We propose an optimal-transport (OT) framework that aligns source- and target-domain spectral distributions in an unsupervised manner, requiring no target-domain concentration labels. Under a stringent leave-one-adulterant-out protocol across five adulterant systems, the OT-aligned support vector regression model achieves a mean R2 of 0.965 (s.d. = 0.004), substantially outperforming raw spectra (R2 = 0.831), PCA (R2 = 0.693) and PARAFAC (R2 = 0.659). These results indicate that olive-oil concentration is encoded as an intrinsic geometric coordinate within the spectral distribution, recoverable by optimal transport without adulterant-specific supervision.