This work implements a mobile application designed to address the issue of counterfeit Peruvian banknotes. The solution is based on the use of image processing libraries, particularly OpenCV, and the training of the YOLOv8 convolutional neural network (CNN) for the detection and segmentation of visible security features on authentic banknotes. The model was trained using a dataset of 1,000 images labeled with security features such as watermark, security thread, microprinting, live spark, and hidden number. Upon obtaining predictions, a veracity percentage is calculated as a response to the user. The results obtained with the application showed a 96.875% effectiveness rate in the classification of genuine and potentially counterfeit banknotes in real-world tests.