Citas bibligráficas
Ramirez, A., (2019). Un algoritmo de detección y validación de rostros en entornos no controlados orientado al monitoreo biométrico de conductores vehiculares [Tesis, Universidad Nacional de Ingeniería]. http://hdl.handle.net/20.500.14076/19928
Ramirez, A., Un algoritmo de detección y validación de rostros en entornos no controlados orientado al monitoreo biométrico de conductores vehiculares [Tesis]. PE: Universidad Nacional de Ingeniería; 2019. http://hdl.handle.net/20.500.14076/19928
@mastersthesis{renati/710840,
title = "Un algoritmo de detección y validación de rostros en entornos no controlados orientado al monitoreo biométrico de conductores vehiculares",
author = "Ramirez Perez Gao, Alejandro Mohamark",
publisher = "Universidad Nacional de Ingeniería",
year = "2019"
}
This research work seeks to solve a biometrics problem in the areas of digital image processing and computer vision in order monitor public transport drivers through face identification. The aim of this work is to propose a face detection and identification algorithm that is robust to the various lighting conditions to which drivers may be subjected in wild environments. The fast local histogram specification algorithm was implemented to reduce the noise due lighting conditions and a posterior study was made to analyze how this algorithm affects the performance in the tasks of face detection and identification, which are based on descriptors generated by histograms of oriented gradients and a convolutional neural network with FaceNet architecture respectively. Finally, support vector machine was used to carry out the classification task in face detection and identification stages, achieving a 71.98% sensibility and 97.65% specificity in face detection and a 99.2% accuracy in face identification. Those metrics were obtained using 6’094 images to train the algorithm and 1’523 images for testing, corresponding to 30 different people.
Este ítem está sujeto a una licencia Creative Commons Licencia Creative Commons