Money counterfeiting is the illegal duplication of any currency for the use of deceiving any entity in exchange for a real-world value. Due to the advancements in computer vision in digital computing and the ill-effects of money counterfeiting, it had become one of the most prevalent issues in the fiscal system of any country that needs to be progressively solved. This paper investigated the use of AlexNet, VGG-16, and ResNet18 through transfer learning for the task of Philippine banknote counterfeit detection. The trained models of this paper achieved a testing accuracy of 81.25% for AlexNet, 95.96% for VGG-16, and 99.59% for ResNet-18. Despite achieving a lower testing accuracy, the trained ResNet-18 model of this study achieved a validation accuracy, specificity, precision, sensitivity, and F1-score of 100% on live testing through the developed web-based money counterfeit detection system..