NORMA eResearch @NCI Library

Synthetic Aperture Radar (SAR) Automatic Target Recognition (ATR) of Ships using Data Augmentation and Deep Learning

Rayate, Devashish Vijay (2021) Synthetic Aperture Radar (SAR) Automatic Target Recognition (ATR) of Ships using Data Augmentation and Deep Learning. Masters thesis, Dublin, National College of Ireland.

[thumbnail of Master of Science]
Preview
PDF (Master of Science)
Download (3MB) | Preview
[thumbnail of Configuration manual]
Preview
PDF (Configuration manual)
Download (2MB) | Preview

Abstract

Automatic Target Recognition (ATR) using Synthetic Aperture Radar (SAR) has become a popular issue in studies, and it’s crucial for water target monitoring. Because SAR imagery are difficult to interpret directly, different machine learning methodologies have been implemented in recent times to recognize maritime objects from SAR data. Since standard deep learning models could only go so far, their effectiveness is limited, and they require more time to train. They often face overfitting issue owing to the unavailability of enough satellite imagery. To solve the SAR-ATR challenge of detecting maritime objects, this paper proposes a deep learning model called Mask R-CNN which uses ResNet101 as a backbone. The model can successfully recognize and segment ships. Data augmentation procedures such as rotating, flip, contrast and brightness modification, fading and intensifying the image are used to overcome challenges with limited SAR images. On the grounds of mean average precision (mAP), the model was analyzed with and without data augmentation. According to the results, the model’s mAP improved from 48.9% to 71.52% when it was trained using data augmentation and hyperparameter optimization.

Item Type: Thesis (Masters)
Uncontrolled Keywords: SAR; deep learning; convolution neural network; mask R-CNN; ResNet101; data augmentation; hyperparameter optimization
Subjects: Q Science > QA Mathematics > Electronic computers. Computer science
T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science
Q Science > QA Mathematics > Computer software
T Technology > T Technology (General) > Information Technology > Computer software
Divisions: School of Computing > Master of Science in Data Analytics
Depositing User: Clara Chan
Date Deposited: 14 Dec 2021 12:40
Last Modified: 14 Dec 2021 12:40
URI: https://norma.ncirl.ie/id/eprint/5217

Actions (login required)

View Item View Item