Boat extraction in harbors from high resolution satellite images using mathematical morphology and marked point processes

Abstract : Boat extraction in harbors from high resolution remotely sensed imagery is a very difficult task, due to the particular distribution of the boats. A marked point process model has been devised for solving this task. This model uses a series of parameters, some of which are determined by hand through trial and error techniques. In this paper, we use mathematical morphology to automatically determine the global orientation of the objects, which is a key parameter of the model.
Liste complète des métadonnées

Cited literature [10 references]  Display  Hide  Download

https://hal.inria.fr/hal-00867592
Contributor : Paula Craciun <>
Submitted on : Monday, September 30, 2013 - 11:35:36 AM
Last modification on : Saturday, January 27, 2018 - 1:31:38 AM
Document(s) archivé(s) le : Tuesday, December 31, 2013 - 4:25:28 AM

File

GRETSI_13_FINAL.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-00867592, version 1

Collections

Citation

Paula Craciun, Josiane Zerubia. Boat extraction in harbors from high resolution satellite images using mathematical morphology and marked point processes. GRETSI - Traitement du Signal et des Images, Sep 2013, Brest, France. ⟨hal-00867592⟩

Share

Metrics

Record views

274

Files downloads

184