Scholarlink

<p>The Malaysian jackfruit (<i>Artocarpus heterophyllus</i> cv. Tekam Yellow) industry is challenged by jackfruit bronzing, a physiological disorder associated with <i>Pantoea stewartii</i> subsp. <i>stewartii</i>, which induces internal discoloration without external symptoms. This study evaluates the feasibility of visible–near-infrared spectroscopy (Vis–NIRS) for the non-destructive prediction of soluble solid concentration (SSC) and ascorbic acid (AA) in jackfruit rind and flesh across 10, 12, and 14&#xa0;weeks after anthesis (WAA). A total of 27 fruits were analyzed, generating 2430 spectra (500–950&#xa0;nm) using an Ocean Optics HR4000 spectrometer. Partial least squares regression (PLSR) models were developed using different spectral preprocessing techniques, including Savitzky–Golay smoothing, standard normal variate (SNV), and multiplicative scatter correction (MSC). Model performance was evaluated using <i>R</i><sup>2</sup>, RMSE, bias, and residual predictive deviation (RPD). The best-performing models were obtained using SNV and MSC preprocessing, achieving coefficient of determination for validation (<i>R</i><sub>v</sub><sup>2</sup>) of 0.97 for SSC and 0.94 for AA, with corresponding root mean square error of validation values of 1.26% and 0.72&#xa0;mg/100&#xa0;g fresh weight, respectively. SNV- and MSC-based preprocessing significantly improved model robustness compared with raw spectra. Rind-based models showed higher stability than flesh-based models due to reduced tissue heterogeneity. These findings demonstrate that Vis–NIRS combined with chemometric modeling provides a rapid, accurate, and non-destructive approach for assessing internal quality and detecting bronzing-associated physiological changes in jackfruit.</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

Predicting the Soluble Solid Concentration and Ascorbic Acid of Intact Jackfruit var. Tekam Yellow Using Visible–Near-Infrared Spectroscopy: Comparison of Rind and Flesh Spectra

  • 期刊论文

  • 发表日期：

    2026年5月4日

查看全文
  • Kok Jeng Lim,
  • Phebe Ding,
  • Nazmi Mat Nawi,
  • Mashitah Jusoh,
  • Najidah Abdullah,
  • Siti Saripa Rabiah Mat Lazim

摘要

The Malaysian jackfruit (Artocarpus heterophyllus cv. Tekam Yellow) industry is challenged by jackfruit bronzing, a physiological disorder associated with Pantoea stewartii subsp. stewartii, which induces internal discoloration without external symptoms. This study evaluates the feasibility of visible–near-infrared spectroscopy (Vis–NIRS) for the non-destructive prediction of soluble solid concentration (SSC) and ascorbic acid (AA) in jackfruit rind and flesh across 10, 12, and 14 weeks after anthesis (WAA). A total of 27 fruits were analyzed, generating 2430 spectra (500–950 nm) using an Ocean Optics HR4000 spectrometer. Partial least squares regression (PLSR) models were developed using different spectral preprocessing techniques, including Savitzky–Golay smoothing, standard normal variate (SNV), and multiplicative scatter correction (MSC). Model performance was evaluated using R2, RMSE, bias, and residual predictive deviation (RPD). The best-performing models were obtained using SNV and MSC preprocessing, achieving coefficient of determination for validation (Rv2) of 0.97 for SSC and 0.94 for AA, with corresponding root mean square error of validation values of 1.26% and 0.72 mg/100 g fresh weight, respectively. SNV- and MSC-based preprocessing significantly improved model robustness compared with raw spectra. Rind-based models showed higher stability than flesh-based models due to reduced tissue heterogeneity. These findings demonstrate that Vis–NIRS combined with chemometric modeling provides a rapid, accurate, and non-destructive approach for assessing internal quality and detecting bronzing-associated physiological changes in jackfruit.