Generalised classification of fluvial scenes with deep learning methods [dataset]

Miyamoto, Hitoshi (Shibaura Institute of Tech., Japan); Dietrich, James (University of Northern Iowa); Dugdale, Steve (University of Nottingham); Carbonneau, Patrice (Durham University); Woodget, Amy (Loughborough University); Fonstad, Mark (University of Oregon); Breckon, Toby (Durham University)

Actions

Download Analytics Citations

Export to: EndNote | Zotero | Mendeley

Collections

This file is not currently in any collections.

Generalised classification of fluvial scenes with deep learning methods [dataset] Open Access

This dataset supports the development of a deep learning classifier specialised in hyperspatial (<10cm) resolution imagery of fluvial scenes such that can be obtained from a drone survey. The dataset has images from 11 rivers. this data is split into training and validation images, all with class label information totalling 5 billion classified pixels.

Descriptions

Resource type: Dataset
Contributors: Creator: Carbonneau, Patrice ¹
Data collector: Dugdale, Steve ²
Data collector: Miyamoto, Hitoshi ³
Data collector: Woodget, Amy ⁴
Data collector: Fonstad, Mark ⁵
Data collector: Dietrich, James ⁶
Data collector: Breckon, Toby ¹

¹ Durham University
² University of Nottingham
³ Shibaura Institute of Tech., Japan
⁴ Loughborough University
⁵ University of Oregon
⁶ University of Northern Iowa
Funder
Research methods: Deep learning
Other description: This data is linked to code in the following GitHub repository: https://github.com/geojames/Self-Supervised-Classification

The dataset is zipped in a single large file. Once downloaded, the internal file structure should be preserved since it partitions training and prediction (validation) data. The unzipped file structure is compatible with the source code available on GitHub.
Keyword: Deep learning
Fluvial remote sensing
UAV imagery
Subject: Remote sensing
Location
Language
Cited in
Identifier: ark:/32150/r2f1881k904
doi:10.15128/r2f1881k904
Rights: Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC)
Publisher: Durham University
Date Created

File Details

Depositor: N. Syrotiuk
Date Uploaded: 2 May 2019, 08:05:26
Date Modified: 2 May 2019, 15:05:11
Audit Status: Audits have not yet been run on this file.
Characterization: File format: zip (ZIP Format)

Mime type: application/zip

File size: 4200051587

Last modified: 2019:05:02 09:08:21+01:00

Filename: Deep_Learning_Data.zip

Original checksum: cf6e2eef2800b23f106cf7f71dde600a

Activity of users you follow
User Activity	Date
User P. Carbonneau has updated Generalised classification of fluvial scenes with deep learning methods [dataset]	almost 5 years ago