Multi-layer Sparse Coding Based Ship Detection for Remote Sensing Images

With the development of remote sensing technology, it becomes possible for the detection and identification of targets from remote sensing images. In this paper, we propose a new method integrating the bottom-up and the top-down mechanisms for the ship detection in high resolution satellite images....

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:2015 IEEE International Conference on Information Reuse and Integration s. 122 - 125
Hlavní autori: Zimeng Li, Daiqin Yang, Zhenzhong Chen
Médium: Konferenčný príspevok..
Jazyk:English
Vydavateľské údaje: IEEE 01.08.2015
Predmet:
Deformable models
Deformable Part Model
Encoding
Feature extraction
Global Saliency Detection
Marine vehicles
Multi-layer Sparse Coding
Remote sensing
Satellites
Ship Detection
Support vector machines
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Popis
Shrnutí:With the development of remote sensing technology, it becomes possible for the detection and identification of targets from remote sensing images. In this paper, we propose a new method integrating the bottom-up and the top-down mechanisms for the ship detection in high resolution satellite images. We use the multi-layer sparse coding to extract the features of the RS images. Then, we get the ship candidate regions by calculating the global saliency map which may have ships in it. Deformable part model is used to extract the ship features and latent support vector machine is used for the ship identification. As demonstrated in our experiments, the proposed approach can effectively detect ship in remote sensing images.
DOI:10.1109/IRI.2015.28