This chapter presents a comprehensive review of hyperspectral imaging (HSI)–based approaches for non-destructive quantification of curcumin concentration in fresh turmeric rhizomes. Conventional laboratory techniques such as High-Performance Liquid Chromatography (HPLC), although accurate, are destructive, time-consuming, and unsuitable for large-scale or real-time quality assessment. Hyperspectral imaging integrates imaging and spectroscopy to capture both spatial and spectral information, enabling pixel-level chemical analysis. This chapter reviews curcumin chemistry, hyperspectral acquisition systems, spectral preprocessing techniques, chemometric and machine learning methods, and recent advances in deep learning–based spectral–spatial models. Insights from experimental studies using fresh turmeric rhizomes and HPLC ground truthing are incorporated to position deep learning frameworks, particularly three-dimensional convolutional neural networks (3D-CNNs), as a robust and scalable solution. Challenges, limitations, and future research directions, including cloud-based deployment, are also discussed.
Sarfaraz Pathan (Thu,) studied this question.