Category (2). In this paper we propose a preprocessing method of infrared imaging (IR) system designed to estimate the infrared radiation at the scene of acquisition. Radiation in the environment is estimated by using the eye area of the individuals present in the scene, to the region of the face is calculated with different metrics which seeks to establish an estimate of the amount of IR radiation. The metrics used correspond to energy, contrast, homogeneity and entropy using the correlation matrix of gray being used in order to characterize and establish IR radiation ranges in which optimal conditions presented images in regard to their characteristics the results obtained show that the eye area used as pattern has features that could be used to estimate changes of IR radiation in the environment, so it can be used to calculate the metrics described above which will serve as input algorithms to maximize contrast and illumination compensation for the purpose of exercising adaptive corrections on image sequences. Also the contrast maximization algorithm and illumination compensation demonstrated adequate corrections on the infrared images compared to corrections made by the art without adaptability is, when IR radiation is estimated by a pattern, however noise components of the image initial produce some corrections made clearer image noise.
La fatiga al conducir es uno de las principales causas de accidentes de tránsito. Los sistemas de detección de fatiga basados en visión por computador tienen gran potencial dada su característica de no-invasividad. Los principales desafíos que se presentan son los movimientos rápidos de ojos y boca, cambios de pose y variacio-nes de iluminación. En este trabajo se entrena un Modelo de Forma Activa para la detección de características faciales extraídas del modelo paramétrico Candide-3. Se describe la metodología de caracterización a partir del modelo paramétrico. Se evalúan de forma cuantitativa la exactitud en: la detección de características y la estimación de los parámetros asociados a la fatiga; analizando su robustez a variaciones de pose y variaciones locales de las regiones de interés. El modelo utilizado y la metodología de caracterización mostraron eficiencia para detectar la fatiga en el 100% de los casos evaluados.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.