In this study, it was aimed to perform the colorimetric determination of fluoride in human saliva using smartphone camera. The SPADNS method (The mixed acid-zirconyl-SPADNS reagent (4,5-Dihydroxy-3-(p-sulfophenylazo)-2,7-naphthalene disulfonic acid)) was adapted to digital image-based colorimetric detection system. Relationship between F− concentrations and RGB (Red, Green, Blue) values was used to obtain analytical linearity. The method was validated by ultraviolet visible (UV–vis) spectroscopy for fluoride in human saliva. In the optimal experimental conditions, detection limit, relative standard deviation and R2 values were determined as 0.06 mg/L, 0.31% and 0.995, respectively. Recovery results (99–118%) were calculated for saliva samples in different humans at two different concentrations. In addition, the amount of residual fluorine passing from toothpaste to human saliva in 5 min was calculated as 0.14 mg/L. The same results were obtained with the UV–vis device, proving that the developed method can be used easily in dentistry and medicine.