Hyperspectral Technology in Agricultural Soil Heavy Metal Detection: Current Applications and Future Directions

Soil heavy metal pollution has become a global environmental issue, posing serious threats to agricultural productivity, food safety, and human health. The emergence of hyperspectral technology provides new approaches for rapid and non-destructive detection of soil heavy metals. This paper reviews the application of hyperspectral technology in identifying heavy metal elements in agricultural soils. First, it introduces the fundamental principles of hyperspectral technology, including the relationship between soil spectral characteristics and heavy metal elements, hyperspectral data acquisition and processing, as well as spectral feature extraction and analysis methods. Then, the research progress in agricultural soil heavy metal identification using hyperspectral technology is detailed from three aspects: laboratory studies, field applications, and integrated applications with other technologies. Studies demonstrate that hyperspectral technology can achieve high-precision prediction of soil heavy metal content through interactions between soil components and heavy metals, utilizing methods such as Continuous Wavelet Transform (CWT) combined with Radial Basis Function (RBF) models. However, practical applications still face challenges including soil background interference, data complexity, and high operational costs. Finally, the paper discusses the advantages and limitations of hyperspectral technology in agricultural soil heavy metal identification, and prospects future development directions including technological improvements and innovations, expansion of application scopes, and establishment of standardization and normalization. Future research should focus on enhancing sensor resolution, optimizing algorithms, and establishing unified spectral databases to improve model generalizability, while promoting widespread application of hyperspectral technology in agricultural soil heavy metal monitoring through multidisciplinary collaboration.