{"title":"利用先进的化学计量学进行稳健的光谱分析以评估水果质量","authors":"Xiaolei Zhang , Jie Yang","doi":"10.1016/j.tifs.2024.104612","DOIUrl":null,"url":null,"abstract":"<div><h3>Background</h3><p>The application of visible/near-infrared (Vis/NIR) spectroscopy for fruit quality evaluation has garnered significant attention over the past few decades. Various chemometric techniques have been developed to predict fruit quality from spectral data. However, the broad applicability of existing chemometric models is limited by unpredictable data variability caused by various biological factors, instrumental settings, and measurement conditions. Deep learning has emerged as a leading methodology, offering substantial improvements in the accuracy and robustness of fruit quality assessments.</p></div><div><h3>Scope and approach</h3><p>This review examines the challenges of model robustness in fruit spectral analysis, tracing the advancement from conventional chemometrics to deep learning approaches. Developments in chemometric methods to enhance model reliability are explored, encompassing dataset-level, variable-level, and model parameter-level strategies, while outlining their applicability and limitations. Recent advances in deep learning-based techniques, e.g., transfer learning, multi-task learning, multi-modal data fusion, and knowledge-guided model design, are further highlighted, providing prospective pathways for achieving superior model robustness.</p></div><div><h3>Key findings and conclusions</h3><p>Current chemometric methods have enhanced model accuracy and proven effective in fruit spectral analysis. While the results are improved for certain research objectives, many analyses remain dependent on specific dataset characteristics and manual feature engineering, such as preprocessing, which limits their generalizability. Deep learning techniques with advanced feature extraction capabilities have shown promise in reducing the need for manually engineered features and expanding model robustness. However, further investigation into the applicability and limitations of these models is crucial for their successful integration into chemometric analysis.</p></div>","PeriodicalId":441,"journal":{"name":"Trends in Food Science & Technology","volume":null,"pages":null},"PeriodicalIF":15.1000,"publicationDate":"2024-07-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Advanced chemometrics toward robust spectral analysis for fruit quality evaluation\",\"authors\":\"Xiaolei Zhang , Jie Yang\",\"doi\":\"10.1016/j.tifs.2024.104612\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><h3>Background</h3><p>The application of visible/near-infrared (Vis/NIR) spectroscopy for fruit quality evaluation has garnered significant attention over the past few decades. Various chemometric techniques have been developed to predict fruit quality from spectral data. However, the broad applicability of existing chemometric models is limited by unpredictable data variability caused by various biological factors, instrumental settings, and measurement conditions. Deep learning has emerged as a leading methodology, offering substantial improvements in the accuracy and robustness of fruit quality assessments.</p></div><div><h3>Scope and approach</h3><p>This review examines the challenges of model robustness in fruit spectral analysis, tracing the advancement from conventional chemometrics to deep learning approaches. Developments in chemometric methods to enhance model reliability are explored, encompassing dataset-level, variable-level, and model parameter-level strategies, while outlining their applicability and limitations. Recent advances in deep learning-based techniques, e.g., transfer learning, multi-task learning, multi-modal data fusion, and knowledge-guided model design, are further highlighted, providing prospective pathways for achieving superior model robustness.</p></div><div><h3>Key findings and conclusions</h3><p>Current chemometric methods have enhanced model accuracy and proven effective in fruit spectral analysis. While the results are improved for certain research objectives, many analyses remain dependent on specific dataset characteristics and manual feature engineering, such as preprocessing, which limits their generalizability. Deep learning techniques with advanced feature extraction capabilities have shown promise in reducing the need for manually engineered features and expanding model robustness. However, further investigation into the applicability and limitations of these models is crucial for their successful integration into chemometric analysis.</p></div>\",\"PeriodicalId\":441,\"journal\":{\"name\":\"Trends in Food Science & Technology\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":15.1000,\"publicationDate\":\"2024-07-02\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Trends in Food Science & Technology\",\"FirstCategoryId\":\"97\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S0924224424002887\",\"RegionNum\":1,\"RegionCategory\":\"农林科学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"FOOD SCIENCE & TECHNOLOGY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Trends in Food Science & Technology","FirstCategoryId":"97","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0924224424002887","RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"FOOD SCIENCE & TECHNOLOGY","Score":null,"Total":0}
Advanced chemometrics toward robust spectral analysis for fruit quality evaluation
Background
The application of visible/near-infrared (Vis/NIR) spectroscopy for fruit quality evaluation has garnered significant attention over the past few decades. Various chemometric techniques have been developed to predict fruit quality from spectral data. However, the broad applicability of existing chemometric models is limited by unpredictable data variability caused by various biological factors, instrumental settings, and measurement conditions. Deep learning has emerged as a leading methodology, offering substantial improvements in the accuracy and robustness of fruit quality assessments.
Scope and approach
This review examines the challenges of model robustness in fruit spectral analysis, tracing the advancement from conventional chemometrics to deep learning approaches. Developments in chemometric methods to enhance model reliability are explored, encompassing dataset-level, variable-level, and model parameter-level strategies, while outlining their applicability and limitations. Recent advances in deep learning-based techniques, e.g., transfer learning, multi-task learning, multi-modal data fusion, and knowledge-guided model design, are further highlighted, providing prospective pathways for achieving superior model robustness.
Key findings and conclusions
Current chemometric methods have enhanced model accuracy and proven effective in fruit spectral analysis. While the results are improved for certain research objectives, many analyses remain dependent on specific dataset characteristics and manual feature engineering, such as preprocessing, which limits their generalizability. Deep learning techniques with advanced feature extraction capabilities have shown promise in reducing the need for manually engineered features and expanding model robustness. However, further investigation into the applicability and limitations of these models is crucial for their successful integration into chemometric analysis.
期刊介绍:
Trends in Food Science & Technology is a prestigious international journal that specializes in peer-reviewed articles covering the latest advancements in technology, food science, and human nutrition. It serves as a bridge between specialized primary journals and general trade magazines, providing readable and scientifically rigorous reviews and commentaries on current research developments and their potential applications in the food industry.
Unlike traditional journals, Trends in Food Science & Technology does not publish original research papers. Instead, it focuses on critical and comprehensive reviews to offer valuable insights for professionals in the field. By bringing together cutting-edge research and industry applications, this journal plays a vital role in disseminating knowledge and facilitating advancements in the food science and technology sector.