With more than 20 years of development, the X-ray fluorescence core scanners (XRF-CS) have been widely applied in the elemental analysis of multi-type sediment cores, the paleoenvironment reconstruction, and the exploration of mineral reservoirs and their abundance, exhibiting great potential for application. However, there is a lack of studies on the influencing factors and correction of the elemental signals output by XRF-CS (especially in China), which restricts the proper use of XRF-CS and the accurate interpretation of their data. Compared with conventional XRF techniques, XRF-CS enjoy a high processing speed (only 1/10 of the time for conventional analysis), high continuity, non-destructive scanning, and a high resolution (up to 0.02 mm). However, XRF-CS only output semi-quantitative values of elemental signals and thus fail to accurately identify the element compositions. This study summarized the influencing factors of the values of the elemental signals output by XRF-CS in terms of instruments and cores, together with the degrees of the influences. On this basis, this study proposed achieving the balance between the intensity of elemental signals output by XRF-CS and cost by selecting appropriate scanning steps and exposure time on the premise of the optimal instrument setting. This study also suggested that the influences of water content and particle sizes on elemental signals should be eliminated as far as possible by drying in the air and smoothing the core surface during the scanning. To improve the accuracy of elemental signals output by XRF-CS, this study systematically introduced three types of international common calibration models and their application potential, namely the normalized median-scaled (NMS) model, the log-ratio calibration equation (LRCE) model, the improved multivariate log-ratio calibration (MLC) model, the normalized polynomial-scaled calibration (NPS) model, and polynomial-corrected multivariate log-ratio calibration (P-MLC) model. Finally, this study proposed further enhancing research on the comparative analysis of the influence exerted by the same factor among multiple types of XRF-CS; the optimization of calibration models and development of visual software packages; the equipment of multiple sensors for integrated scanning, and the extensive applications in the exploration and evaluation of geological and mineral resources.