Rodolfo Lotte

Company Owner @ LCIARS / Consultant in Remote Sensing, Machine Learning and Computer Vision / PhD in Remote Sensing / MSc. in Applied Computing

Detection of points for road extraction in airborne SAR images: an approach based in Self-Organizing Maps

Rodolfo G. Lotte, Cláudia M. Almeida, José Demísio S. da Silva

Download paper

Abstract

The great advantage in the use of radar images is the possibility of survering areas often covered by clouds, since the imaging by active sensors is independent of atmospheric conditions in the region of interest. Often, the mapping from these images is done manually, requiring considerable time and effort. This article discusses the use of Self-Organizing Maps - SOM as a method of identifying points in images. Each point represents an identified element belonging to a road. It will examine two different models to capture patterns in the image, presenting as a result both the identified points as well as the percentage of agreement in both models. The quality in the identification of these points is essential in the performance of certain roads extractors in digital images.

Contextualization

Cite this paper

@article{lotte2011-cbic,
  title={Detection of points for road extraction in airborne SAR images: an approach based in Self-Organizing Maps},
  author={Lotte, R.G., Almeida, C. M., and Silva, J. D. S.},
  journal={Congresso Brasileiro de Inteligência Computacional (CBIC)},
  number={10},  
  year={2011}
}