Scholarlink

<p>Fruit quality assessment has advanced considerably with the development of imaging technologies and computational data analysis. Conventional evaluation methods are often destructive, labor-intensive, and unsuitable for rapid industrial implementation. Hyperspectral imaging (HSI) has emerged as a&#xa0;powerful nondestructive technique that simultaneously acquires spatial and spectral information across continuous wavelength bands, enabling comprehensive characterization of fruit tissues. By integrating imaging and spectroscopy within a&#xa0;single spectral–spatial framework, HSI allows simultaneous evaluation of both internal and external quality attributes. HSI has been widely applied to predict and classify key parameters, including firmness, soluble solids content (SSC), total soluble solids (TSS), acidity, moisture content, bruising, contamination, chilling injury, and maturity stage, across diverse horticultural commodities. However, the high dimensionality and multicollinearity of hyperspectral datasets require advanced chemometric strategies for effective interpretation. The integration of spectral preprocessing methods, wavelength selection algorithms, and regression or classification models, including Partial Least Squares Regression (PLSR), Support Vector Machines (SVM), and artificial neural networks, enhances prediction accuracy while reducing data redundancy and computational burden. This review synthesizes the principles of HIS, mechanisms of light–tissue interaction, system configurations, and chemometric modeling techniques for efficient fruit quality assessment. Current applications, industrial relevance, and existing challenges are critically discussed. Although limitations related to cost, calibration transfer, and real-time processing remain, ongoing advancements in artificial intelligence, multispectral optimization, and automated grading technologies are expected to accelerate the practical adoption of integrated hyperspectral–chemometric systems in precision horticulture and postharvest quality management.</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

Integration of Hyperspectral Imaging and Chemometrics for Efficient Fruit Quality Assessment

  • 期刊论文

  • 发表日期：

    2026年3月31日

查看全文
  • Abdul Basit,
  • Ayesha Khan,
  • Yaqoob Sultan,
  • Hamza Ali,
  • Arifa Bano,
  • Abbas Khan

摘要

Fruit quality assessment has advanced considerably with the development of imaging technologies and computational data analysis. Conventional evaluation methods are often destructive, labor-intensive, and unsuitable for rapid industrial implementation. Hyperspectral imaging (HSI) has emerged as a powerful nondestructive technique that simultaneously acquires spatial and spectral information across continuous wavelength bands, enabling comprehensive characterization of fruit tissues. By integrating imaging and spectroscopy within a single spectral–spatial framework, HSI allows simultaneous evaluation of both internal and external quality attributes. HSI has been widely applied to predict and classify key parameters, including firmness, soluble solids content (SSC), total soluble solids (TSS), acidity, moisture content, bruising, contamination, chilling injury, and maturity stage, across diverse horticultural commodities. However, the high dimensionality and multicollinearity of hyperspectral datasets require advanced chemometric strategies for effective interpretation. The integration of spectral preprocessing methods, wavelength selection algorithms, and regression or classification models, including Partial Least Squares Regression (PLSR), Support Vector Machines (SVM), and artificial neural networks, enhances prediction accuracy while reducing data redundancy and computational burden. This review synthesizes the principles of HIS, mechanisms of light–tissue interaction, system configurations, and chemometric modeling techniques for efficient fruit quality assessment. Current applications, industrial relevance, and existing challenges are critically discussed. Although limitations related to cost, calibration transfer, and real-time processing remain, ongoing advancements in artificial intelligence, multispectral optimization, and automated grading technologies are expected to accelerate the practical adoption of integrated hyperspectral–chemometric systems in precision horticulture and postharvest quality management.