“…On the other hand, hyperspectral imaging can simultaneously obtain the target spectrum and image information, and is regarded as a technique with high-throughput plant phenotype potential (Pandey et al, 2017 ). Although there are many studies on nutrition monitoring using near-end hyperspectral imaging, most of them focus on quantitative monitoring and diagnosis of crop nitrogen (N), such as wheat (Mahajan et al, 2014 ; Jiang et al, 2021 ), rice (Men et al, 2021 ), maize (Furlanetto et al, 2021 ), cotton (Oliveira et al, 2020 ), rape (Liu et al, 2020a ), soybean (Chen et al, 2019 ), orange (Osco et al, 2019 , 2020a ), tea (Wang et al, 2020 ) and mango (Mahajan et al, 2021 ). At present, the quantitative monitoring research on crop K is also gradually carried out, but more studies often analyze the K together with other elements (Liu et al, 2020b ; Osco et al, 2020a , b ; Mahajan et al, 2021 ), and there are few studies only on the characteristics of single the K nutrient element.…”