Detección de motocicletas y del uso del casco de seguridad utilizando un algoritmo que emplea técnicas de procesamiento de imágenes y modelos de inteligencia artificial

In Colombia motorcyclists are the first victims of road accident. From 2001 to 2014, there were about 28,000 deaths. About half of these deaths occur as a result of the lack of use of passive protection elements (safety helmet). To address this problem, we propose an algorithm that uses image processing techniques in conjunction with artificial intelligence models for the detection of motorcycles and the use of helmets by the drivers. The precision in the classification of motorcycles reaches 97.14%, whereas that measure in the classification of helmets is of 85,29%. The algorithm proposed here, together with technological devices for traffic enforcement (license plate readers), could become the product that will be the cornerstone for transit and transport authorities, to formulate a public policy that will substantially reduce the mortality rate on this population. Such a policy will bring enormous benefits to society as a whole.

• Autores: Varón Forero, Mario Andrés
• Fecha de publicación: 2017
• Institución: Universidad de los Andes
• Repositorio: Séneca: repositorio Uniandes
• Idioma: spa
• OAI Identifier: oai:repositorio.uniandes.edu.co:1992/61895
• Acceso en línea: http://hdl.handle.net/1992/61895
• Palabra clave: Cascos de seguridad; Inteligencia artificial; Procesamiento de imágenes; Seguridad vial
• Rights: openAccess
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/