Remote sensing satellite images dataset for objects detection
Published: 5 April 2023| Version 1 | DOI: 10.17632/s5v4zz7yj5.1
Contributors:
ADEKANMI ADEGUN, , , Description
The dataset comprises of 92 satellite images containing 61 training set, 21 validation set and 10 testing set, together with the annotation files. Sample images from testing dataset are presented in Fig. 3. The images in the dataset were annotated using the roboflow application. The dataset contains 5 categories of objects: residence, roads, shoreline, swimming pool and vegetation. The dataset can be used to train, validate and test variants of YOLO models [2] for object detection from remote sensing satellite images.
Files
Institutions
University of KwaZulu-Natal
Categories
Earth Sciences, Machine Learning, Applied Computing in Earth Sciences