Detección de grietas en la vía pública para prevenir las caídas
Cargando...
Archivos
Fecha
2021
Profesor/a Guía
Facultad/escuela
Idioma
es
Título de la revista
ISSN de la revista
Título del volumen
Editor
Universidad Andrés Bello
Nombre de Curso
Licencia CC
Licencia CC
Resumen
La siguiente tesis analiza los pilares esenciales para lograr desarrollar una aplicación digital que detecte y reconozca grietas en caminos mediante imágenes, específicamente en veredas. Para apoyar la relevancia de la aplicación, se consultó un estudio acerca de accidentes de la población adulto mayor en veredas. El estudio considera una encuesta dirigida a la población adulto mayor, con una muestra de 100 adultos mayores. El 75% de ellos respondió que se han caído al menos una vez en una vereda. Adicionalmente, el 50% de ellos indica que esos accidentes fueron causados por la mala calidad del camino (Yannes, 2007).Para lograr el objetivo principal de la tesis, el cual es desarrollar un algoritmo que reconozca grietas en las veredas a través de imágenes, es esencial estudiar una de las ramas de Inteligencia Artificial llamada “Visión Computacional”. Finalmente, es estudió y aplicó redes neuronales en el entrenamiento del modelo de Inteligencia Artificial para lograr un óptimo reconocimiento de grietas auténticas en las veredas. Resultados precisos en el modelo podrían disminuir accidentes de la población adulto mayor en las veredas al tomar acciones preventivas en las grietas detectadas.
The following thesis analyzes the essential pillars to accomplish the development of a digital application which detects and recognizes cracks in roads from images, specifically in sidewalks.To support the relevance of the application, a study regarding the elderly population accidents in sidewalks was reviewed. It considers a survey targeted to the elderly population, with a sample of 100 elder adults. The 75% of them answered that they have fell at least once in a sidewalk. Additionally, 50% of them indicated that the accident was caused by the bad quality of the road (Yannes, 2007).To achieve the main goal of the thesis, which is to develop an algorithm to recognize cracks in sidewalks through images, it is essential to study one of the Artificial Intelligence branches named as “Computer Vision”.Finally, neural networks are studied and applied to train the Arficial Intelligence model towards an optimal recognition of authentical cracks in sidewalks. Accurate results in the model would lead todiminish accidents of the elderly population in sidewalks by taking preventive action on the detected cracks.
The following thesis analyzes the essential pillars to accomplish the development of a digital application which detects and recognizes cracks in roads from images, specifically in sidewalks.To support the relevance of the application, a study regarding the elderly population accidents in sidewalks was reviewed. It considers a survey targeted to the elderly population, with a sample of 100 elder adults. The 75% of them answered that they have fell at least once in a sidewalk. Additionally, 50% of them indicated that the accident was caused by the bad quality of the road (Yannes, 2007).To achieve the main goal of the thesis, which is to develop an algorithm to recognize cracks in sidewalks through images, it is essential to study one of the Artificial Intelligence branches named as “Computer Vision”.Finally, neural networks are studied and applied to train the Arficial Intelligence model towards an optimal recognition of authentical cracks in sidewalks. Accurate results in the model would lead todiminish accidents of the elderly population in sidewalks by taking preventive action on the detected cracks.
Notas
Tesis (Ingeniero Civil Informático)
Palabras clave
Procesamiento de Imagen, Prevención de Accidentes, Innovaciones Tecnológicas