Computer Vision and XRF-IoT Sensor Systems for Detecting Heavy Metals in Export Crops: A Comprehensive Systematic Review

The increasing concern over heavy metal contamination in export crops has intensified research on the application of computer vision systems (CVS) and advanced sensing technologies within multi-level agricultural monitoring frameworks spanning soil contamination assessment, crop spectral diagnostics, and in situ elemental sensing. This study conducts a systematic literature review following Kitchenham’s methodology, from which 68 studies were finally included after screening and eligibility assessment. The review focuses on the use of hyperspectral imaging (HSI) and XRF-IoT sensors (X-ray fluorescence units enhanced with IoT connectivity) for detecting heavy metals in export crops, considering publications from the last seven years indexed in Web of Science Core Collection, Scopus, IEEE Xplore, EBSCOhost, and Springer Nature Link. The findings indicate that research is concentrated in highly digitalized countries, which limits its global applicability; moreover, a substantial proportion of studies is published in Q1 journals, although the methodologies are not always fully objective. Likewise, the most developed research lines are oriented toward image-based diagnostics and crop analysis. These results reveal a gap between technological advances in computer vision and their integration into agricultural decision-making aimed at improving the quality of export crops. It is recommended to foster research with greater geographical diversity, grounded in solid theoretical frameworks and an ethical perspective.