Detecting Ships from Satellite Images using Deep Learning Technique

Mukherjee, Sneham (2022) Detecting Ships from Satellite Images using Deep Learning Technique. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (3MB) | Preview
Preview	PDF (Configuration manual) Download (1MB) | Preview

Abstract

One of the most powerful and successful study areas in the marine surveillance application is ship detection from satellite photos. Beyond that, the suggested research will aid in the development of a powerful navy defense system, the tracking of illicit sea route activities, the tracking of fishing ships, the detection of lost ships, and so on. Detecting ship instances from satellite images is a challenging job. This study will present a Faster R-CNN model architecture based on Feature Engineering. Different types of augmentation techniques like Gaussian filter, Edge detection, horizontal flipping, random cropping, scaling and changing the colour of the image are used along with the Affine transformation. Adam Optimizer is used to construct the model and Binary Crossentropy is employed as a loss function. The model is trained with 50 epochs to get the best accuracy. The findings revealed that data augmentation can aid in the improvement of the model’s performance. The accuracy achieved from the model are 98.16% and 97.33% on validation and test data respectively with a precision score of 0.9820 and 0.9736. The method described in this research, when compared to the prior method in the literature review, may considerably enhance the efficiency of ships detection in large aerial photographs while also boosting accuracy in terms of accuracy.

Item Type:	Thesis (Masters)
Uncontrolled Keywords:	Satellite; Defence System; Faster R-CNN; Augmentation; Gaussian filter; Edge detection; Affine Transformation
Subjects:	Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science T Technology > TC Hydraulic engineering. Ocean engineering V Naval Science > V Naval Science (General) Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning
Divisions:	School of Computing > Master of Science in Data Analytics
Depositing User:	Tamara Malone
Date Deposited:	23 Feb 2023 15:29
Last Modified:	02 Mar 2023 08:43
URI:	https://norma.ncirl.ie/id/eprint/6234

Actions (login required)

View Item