Machine Vision Model Based on Deep Learning to Prevent Theft in Real Estate

Ovalle C. · Proceedings of the LACCEI international Multi-conference for Engineering, Education and Technology (2023)

Resumen

Artificial intelligence has become an increasingly present concept today and terms such as Machine Learning or Deep Learning are increasingly used frequently in the development of technological projects and can have an impact on various social problems that, Unfortunately, they are lived day by day in many countries at the Latin American level, which is why the objective of this research was to develop an artificial vision model based on Deep Learning to prevent theft of real estate assets. Likewise, the research is of an applied type and of a quantitative approach, having a preexperimental cross-sectional design. Finally, after processing the data, it was obtained as results that the convolutional neural networks of Deep Learning are capable of distinguishing the curved lines of a face with very exact precision, only having a margin of error of less than 0.5% in the tests carried out. In addition, it is concluded that the proposed artificial vision model obtained a score in the range of 0.94 to 0.97 for both precision, accuracy, memory and F1, thus proving a positive measurement after applying confusion matrix metrics.

Instituciones: Universidad Tecnológica del Perú