COMPUTER VISION TECHNIQUES FOR TRAFFIC SIGN DETECTION IN ADAS APPLICATIONS

Fuente: Zenodo
Guardado en:
Detalles Bibliográficos
Autor principal: Taras Volodymyrovych, Kravchenko
Formato: Recurso digital
Lenguaje:inglés
Publicado: Zenodo 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866901526800236544
author Taras Volodymyrovych, Kravchenko
author_facet Taras Volodymyrovych, Kravchenko
contents <div> <table> <tbody> <tr> <td style="padding: 0in 9.0pt 0in 9.0pt;"> <p><em><span>this paper presents a comprehensive traffic sign recognition system designed to enhance advanced driver assistance systems (ADAS) and autonomous vehicles. The system employs a three-step algorithm comprising color segmentation, shape recognition, and a neural network-based classification to detect and identify various traffic signs in real time. Leveraging the advantages of color-based segmentation for rapid processing and combining it with sophisticated shape detection methods, our approach ensures high accuracy and precision even under challenging conditions such as varying illumination and occlusions. The integration of neural networks allows for effective classification across a broad range of sign types, addressing limitations seen in traditional methods. Our system’s ability to operate with standard onboard cameras, combined with its resilience to lighting variations, marks a significant advancement in traffic sign recognition technology. Extensive testing demonstrates its efficacy in real-world scenarios, highlighting its potential to enhance road safety and support autonomous driving technologies.</span></em></p> </td> </tr> </tbody> </table> </div>
format Recurso digital
id zenodo_https___doi_org_10_5281_zenodo_15847218
institution Zenodo
language eng
publishDate 2025
publisher Zenodo
record_format zenodo
spellingShingle COMPUTER VISION TECHNIQUES FOR TRAFFIC SIGN DETECTION IN ADAS APPLICATIONS
Taras Volodymyrovych, Kravchenko
artificial intelligence; computer vision; machine learning; neural networks; pattern recognition
<div> <table> <tbody> <tr> <td style="padding: 0in 9.0pt 0in 9.0pt;"> <p><em><span>this paper presents a comprehensive traffic sign recognition system designed to enhance advanced driver assistance systems (ADAS) and autonomous vehicles. The system employs a three-step algorithm comprising color segmentation, shape recognition, and a neural network-based classification to detect and identify various traffic signs in real time. Leveraging the advantages of color-based segmentation for rapid processing and combining it with sophisticated shape detection methods, our approach ensures high accuracy and precision even under challenging conditions such as varying illumination and occlusions. The integration of neural networks allows for effective classification across a broad range of sign types, addressing limitations seen in traditional methods. Our system’s ability to operate with standard onboard cameras, combined with its resilience to lighting variations, marks a significant advancement in traffic sign recognition technology. Extensive testing demonstrates its efficacy in real-world scenarios, highlighting its potential to enhance road safety and support autonomous driving technologies.</span></em></p> </td> </tr> </tbody> </table> </div>
title COMPUTER VISION TECHNIQUES FOR TRAFFIC SIGN DETECTION IN ADAS APPLICATIONS
topic artificial intelligence; computer vision; machine learning; neural networks; pattern recognition
url https://doi.org/10.5281/zenodo.15847218