Microbial bacteria play an irreplaceable role in natural and human production and life; thus, determining their activities is an important issue. This study proposed a method to quantitatively determine the activity of microbial materials through extinction property calculation using infrared spectroscopy. Complex refractive indices of different active biomaterials were calculated based on their infrared spectra, and their extinction properties were determined using the discrete dipole approximation method. Using partial least squares (PLS), support vector regression (SVR), and extreme learning machine (ELM) regression, quantitative determination models of microbial materials based on infrared extinction properties were established to predict their activity. The results demonstrated that the model of least angle regression (LAR) combined with PLS exhibited better potential for the determination of biomaterial activity. The coefficient of determination (R2) values acquired by the optimal model for the three biomaterials were 0.9699, 0.9744, and 0.9621, respectively. These findings suggested that a rapid and accurate quantitative determination of microbial activity can be achieved based on extinction property.