Implementation of an automatic feeding system for dogs and cats with artificial intelligence technical application of convolutional neural network (CNN).
In this project, an artificial intelligence (AI) system was implemented applying convolutional neural network (CNN) techniques for the detection and automatic feeding of dogs and cats in a foster home located at the Laureano Gómez educational institution, in San Agustín, department from Huila Colombia. Two main components were developed: the first, an image processing system and the second, an automatic learning model, the image processing system was in charge of collecting and properly structuring the data for its use in the algorithm that applies techniques of convolutional neural network, the machine learning model adapted to pattern recognition techniques to identify the animals in question. The implementation of the system allows a more efficient and real-time feeding for household animals passing through, which is expected to improve their quality of life. The metrics used to evaluate the performance of the system include precision and functionality, in the experimental tests of the prototype an indicator of success of the algorithm implemented in the identification of dogs and cats is shown above 90%. In conclusion, this project demonstrates the potential of artificial intelligence and its different applicable techniques, one of these being convolutional neural networks that provide a solution to everyday problems that seeks to improve the quality of life of animals within the home of passage in the Educational institution.
Copyright (c) 2022 Documentos de Trabajo ECBTI
