Metamer mismatching occurs in both the human visual system and digital imaging systems. Increasing the number of sensors in digital imaging systems is an efficient method to reduce the degree of metamer mismatching. This study investigates the number of sensors that are needed for the digital imaging systems to have a similar ability to the human sensor system in distinguishing colors from the perspective of metamer mismatching. Optimal spectral reflectance was generated, and a dataset with more than 47 million previously collected practical spectral reflectance functions was used to derive the metameric color pairs. Different sets of Gaussian-shaped functions were designed to model the spectral sensitivity functions of the digital imaging systems. Both the metamer mismatching volumes and color differences were used to characterize the degree of metamer mismatching for pairs of samples, which were metameric to the imaging systems but appeared different to the human sensor system. The results show that the practical metamer mismatching volumes are substantially smaller than the theoretical ones. The results also show that both the metamer mismatching volumes and color differences are significantly reduced by increasing the number of sensors from three to five for the digital imaging system but are only slightly reduced by further increasing the number of sensors from five to seven. This indicates that five sensors is an efficient and optimal solution for an imaging system with Gaussian-shaped sensors to have a similar ability in distinguishing colors compared to the human sensor system.