Accurate estimation of population size is key to understanding the ecology of disease vectors, as well as the epidemiology of the pathogens they carry and to plan effective control activities. Population size can be estimated through mark-release-recapture (MRR) experiments that are based on the assumption that the ratio of recaptured individuals to the total captures approximates the ratio of marked individuals released to the total population. However, methods to obtain population size estimates usually consider pooled data and are often based on the total number of marked and unmarked captures. We here present a logistic regression model, based on the principle of the well-known Fisher-Ford method, specific for MRR experiments where the information available is the number of marked mosquitoes released, the number of marked and unmarked mosquitoes caught in each trap and on each day, and the geographic coordinates of the traps. The model estimates population size, taking into consideration the distance between release points and traps, the time between release and recapture, and the loss of marked mosquitoes to death or dispersal. The performance and accuracy of the logistic regression model has been assessed using simulated data from known population sizes. We then applied the model to data from MRR experiments with Aedes albopictus Skuse performed on the campus of "Sapienza" University in Rome (Italy).