The present study examined pedestrian movements in five pedestrian crossings in Adana, Mersin, and Isparta provinces in the Mediterranean Region of Turkey. The effects of factors that contribute significantly to pedestrian crossing time were considered. Linear (multiple linear regression [MLR]), nonlinear (artificial neural network [ANN], adaptive neuro-fuzzy inference system [ANFIS]), particle swarm optimization (PSO), and crossing time models reported in the literature were used to estimate crossing times and compare the estimations to the collected data. The nonlinear ANN and ANFIS models achieved better predictions than the linear MLR. The models from the literature achieved worse results compared to the other models due to the limited number of included parameters, The model coefficients were calibrated with PSO to improve regional specificity and the accuracy of the predictions improved. Calibrating the models according to the characteristics of the study region improves the accuracy of the findings.