A Ship Detector Design Based on Deep Convolutional Neural Networks for Satellite Images

Ship targetclassification from satellite images is a challenging task with itsrequirements of feature extracting, advanced pre-processing, a variety ofparameters obtained from satellites and other type of images, and analyzing ofimages. The dissimilarity of results, enhanced dataset requirement, int...

Full description

Saved in:
Bibliographic Details
Main Authors: Ferhat Ucar, Deniz Korkmaz
Format: Article
Language:English
Published: Sakarya University 2020-02-01
Series:Sakarya Üniversitesi Fen Bilimleri Enstitüsü Dergisi
Subjects:
Online Access:https://dergipark.org.tr/tr/download/article-file/967515
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Ship targetclassification from satellite images is a challenging task with itsrequirements of feature extracting, advanced pre-processing, a variety ofparameters obtained from satellites and other type of images, and analyzing ofimages. The dissimilarity of results, enhanced dataset requirement, intricacyof the problem domain, general use of Synthetic Aperture Radar (SAR) images andproblems on generalizability are some topics of the issues related to shiptarget detection. In this study, we propose a deep convolutional neural networkmodel for detecting the ships using the satellite images as inputs. Our model has acquired an adequate accuracyvalue by just using a pre-processed satellite image input. Visual and graphicalresults of features at various layers and deconvolutions are also demonstratedfor a better understanding of the basic process.
ISSN:2147-835X