NORMA eResearch @NCI Library

An Encoder-Decoder Framework for Remote Sensing Image Captioning

Mohan, Namita (2021) An Encoder-Decoder Framework for Remote Sensing Image Captioning. Masters thesis, Dublin, National College of Ireland.

[thumbnail of Master of Science]
PDF (Master of Science)
Download (1MB) | Preview
[thumbnail of Configuration manual]
PDF (Configuration manual)
Download (2MB) | Preview


Remote sensing image captioning involves generating human-like text to describe the content of images representing the earth’s surface captured from a distance. The challenge with remote sensing image captioning is identifying multiple objects with large scale differences and the relationship between these objects present in a single image. This research proposes a novel encoder-decoder framework to identify objects and their relationship for generating human-like descriptions of remote sensing images. The encoder-decoder framework consists of Capsule Network and Bidirectional LSTM. The Capsule Network is used to extract the object features from the images and Bi-directional LSTM is used to generate text descriptions based on the object features. This research used the popular RSICD dataset with over ten thousand remote sensing images for the training of the proposed framework. The implemented frameworks improved the BLEU-1 by 7%, BLEU-2 by 16%, BLEU-3 by 20% and BLEU-4 by 17% as compared to the traditional CNN-RNN model. The results demonstrate the power of Capsule Network compared to CNN however, the performance of the framework can be improved by using deeper architecture Capsule Network and much larger datasets. Remote sensing image captioning would be useful for image retrieval in all applications of remote sensing images such as development planning and disaster monitoring.

Item Type: Thesis (Masters)
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: 09 Dec 2021 12:33
Last Modified: 09 Dec 2021 12:33

Actions (login required)

View Item View Item