Pulse-coupled neural network based on an adaptive Gabor filter for pavement crack segmentation

This article proposes a Pulse-Coupled Neural Network based on an adaptive Gabor filter for the segmentation of cracks in the pavement in digital images. By estimating the noise in the image, the parameters of the filter that convolves the neurons of the model are adjusted. As a result iterations wer...

Descripción completa

Detalles Bibliográficos
Autores: Luna Álvarez, A., Mújica Vargas, D., Rubio, J. de J., Rosales Silva, A.
Tipo de recurso: artículo
Estado:Versión publicada
Fecha de publicación:2024
País:México
Institución:UNIVERSIDAD NACIONAL AUTÓNOMA DE MÉXICO
Repositorio:Journal of Applied Research and Technology
Idioma:inglés
OAI Identifier:oai:ojs2.localhost:article/1837
Acceso en línea:https://jart.icat.unam.mx/index.php/jart/article/view/1837
Access Level:acceso abierto
Palabra clave:Pulse-coupled neural network
adaptive Gabor filter
parallel processing
crack segmentation
Descripción
Sumario:This article proposes a Pulse-Coupled Neural Network based on an adaptive Gabor filter for the segmentation of cracks in the pavement in digital images. By estimating the noise in the image, the parameters of the filter that convolves the neurons of the model are adjusted. As a result iterations were reduced to 2%with ? 90% precision. The algorithm was parallelized on the GPU and the processing time was reduced to n/NM regardless of the M and N dimensions of theimage.