Given the increasingly serious threat of heavy metal pollution in farmland to the safety of agricultural products and human health, it is of great significance to develop rapid, non-destructive detection technologies. Traditional chemical detection methods suffer from being destructive, time-consuming and costly. In contrast, spectrum and imaging technology has become the research hotspot due to its advantages of fast and non-destructive. Heavy metal pollution alters the physiological and biochemical characteristics and optical responses of crops. Spectrum and imaging technology can rapidly and non-destructively obtain the corresponding spectral and spatial information, providing a way for pollution screening and quantitative and qualitative analysis. This paper systematically reviews the application of spectral imaging techniques such as visible/near-infrared spectroscopy, hyperspectral imaging, fluorescence hyperspectral imaging, laser-induced breakdown spectroscopy, and Raman spectroscopy in this field, and identifies that these technologies existing limitations and challenges. Future research should focus on multi-technology integration, model universality improvement, and portable device development. Through algorithm optimization, cost control and the extension of the industrial chain, the in-depth application of spectrum and imaging technology should be promoted. This article aims to provide theoretical references and practical guidance for the research and application of rapid non-destructive detection technology for heavy metals in crops.