Xlii Jornadas De Automática : Libro De Actas 2021
DOI: 10.17979/spudc.9788497498043.686
|View full text |Cite
|
Sign up to set email alerts
|

Análisis de técnicas de aumento de datos y entrenamiento en YOLOv3 para detección de objetos en imágenes RGB y TIR del UMA-SAR Dataset

Abstract: El uso de imágenes de los espectros visible (RGB) e infrarrojo térmico (TIR) para la detección de objetos puede resultar crucial en aplicaciones donde las condiciones de visibilidad están limitadas, como la robótica para búsqueda y rescate en catástrofes. Para ello resulta beneficioso analizar cómo las técnicas de aprendizaje profundo basadas en redes neuronales convolucionales (CNN) pueden aplicarse a ambas modalidades. En este artículo se analizan diferentes configuraciones y parámetros para el entrenamiento… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1

Citation Types

0
1
0

Year Published

2023
2023
2023
2023

Publication Types

Select...
1

Relationship

0
1

Authors

Journals

citations
Cited by 1 publication
(1 citation statement)
references
References 12 publications
0
1
0
Order By: Relevance
“…In both cases, postprocessing was necessary using a Python script, via the open-source library CVLIB (Ponnusamy, nf), with the capacity to detect people in photographs using artificial intelligence algorithms. The open source YOLOv3 was pretrained with RGB images from the COCO Dataset [36] because it offers good performance and balanced speed and precision and has been tested in similar cases, e.g., traffic congestion [37]. Once people were detected, all the pedestrians were counted.…”
Section: Data Relating To Cultural Visitsmentioning
confidence: 99%
“…In both cases, postprocessing was necessary using a Python script, via the open-source library CVLIB (Ponnusamy, nf), with the capacity to detect people in photographs using artificial intelligence algorithms. The open source YOLOv3 was pretrained with RGB images from the COCO Dataset [36] because it offers good performance and balanced speed and precision and has been tested in similar cases, e.g., traffic congestion [37]. Once people were detected, all the pedestrians were counted.…”
Section: Data Relating To Cultural Visitsmentioning
confidence: 99%