Text-based CAPTCHA Vulnerability Assessment using a Deep Learning-based Solver

Producción científica: Capítulo del libro/informe/acta de congresoContribución a la conferenciarevisión exhaustiva

1 Cita (Scopus)

Resumen

The focus of this work is to test the security offered by Text-based CAPTCHAs. We present different types of CAPTCHAs and a preprocessing and segmentation process to clean noise in CAPTCHA images and crop digits or characters in single images. We present a convolutional neural network architecture trained under several hyperparameters, comparing multiple models with different batch sizes, epochs, and optimizers. We confirmed that using Text-based CAPTCHAs is no longer a secure mechanism for protection because, with simple computer vision techniques and current machine learning algorithms, they can be broken. We achieved a 90.49% accuracy with our model trained with a mix of four datasets and up to 97.10% with one dataset, which is enough to consider these schemes insecure in practice.

Idioma originalInglés
Título de la publicación alojadaETCM 2021 - 5th Ecuador Technical Chapters Meeting
EditoresMonica Karel Huerta, Sebastian Quevedo, Carlos Monsalve
EditorialInstitute of Electrical and Electronics Engineers Inc.
ISBN (versión digital)9781665441414
DOI
EstadoPublicada - 12 oct. 2021
Evento5th IEEE Ecuador Technical Chapters Meeting, ETCM 2021 - Cuenca, Ecuador
Duración: 12 oct. 202115 oct. 2021

Serie de la publicación

NombreETCM 2021 - 5th Ecuador Technical Chapters Meeting

Conferencia

Conferencia5th IEEE Ecuador Technical Chapters Meeting, ETCM 2021
País/TerritorioEcuador
CiudadCuenca
Período12/10/2115/10/21

Huella

Profundice en los temas de investigación de 'Text-based CAPTCHA Vulnerability Assessment using a Deep Learning-based Solver'. En conjunto forman una huella única.

Citar esto