Datasets:
Tasks:
Image Classification
Modalities:
Image
Formats:
imagefolder
Size:
1K - 10K
ArXiv:
DOI:
License:
Update README.md
Browse files
README.md
CHANGED
|
@@ -20,6 +20,12 @@ size_categories:
|
|
| 20 |
|
| 21 |
This dataset provides imagery used for training and evaluation of models in the Reef Guidance System. All imagery was collected by the Australian Institute of Marine Science using the ReefScan™ Transom Marine Monitoring System.
|
| 22 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 23 |
## Dataset Details
|
| 24 |
|
| 25 |
This dataset contains high resolution images of the seafloor.
|
|
@@ -42,6 +48,7 @@ We provide multiple subsets with different labelling schemes for this dataset. D
|
|
| 42 |
- **Paper:** [Arxiv](link here)
|
| 43 |
|
| 44 |
To read more about the Reef Restoration and Adaptation Program: [RRAP](https://gbrrestoration.org/program/translation-to-deployment/)
|
|
|
|
| 45 |
To read more about the ReefScan™ Transom Marine Monitoring System [ReefScan™](https://www.aims.gov.au/research/technology/reefscan/reefscan-transom)
|
| 46 |
|
| 47 |
### Structure
|
|
@@ -52,8 +59,6 @@ We provide multiple labelling subsets for each location. The different subsets
|
|
| 52 |
|
| 53 |
All images were collected on the Great Barrier Reef in Australia.
|
| 54 |
|
| 55 |
-
We acknowledge the continuing Sea Country management and custodianship of the Great Barrier Reef by Aboriginal and Torres Strait Islander Traditional Owners whose rich cultures, heritage values, enduring connections and shared efforts protect the Reef for future generations.
|
| 56 |
-
|
| 57 |
The table below specifies the locations, depth ranges and labelling subsets for each of the 5 sites included in this dataset.
|
| 58 |
|
| 59 |
| Site | Location | Depth Range | Deployment Sequences | Patches | Weak Labels |
|
|
@@ -79,11 +84,11 @@ The whole images have associated metadata including:
|
|
| 79 |
- Depth: altitude or distance to seafloor
|
| 80 |
- Camera properties: focal length, exposure time, f-stop, ISO speed
|
| 81 |
|
| 82 |
-
Deployment Sequences are sequential frames to match the real coral dispensing scenario, labelled at the image-level as `Deploy' or `No-Deploy'.
|
| 83 |
|
| 84 |
-
Weak Labels are whole images that have been selected by our annotator to best represent the class i.e. the majority of the image matches the label. These images are labelled into `Deploy
|
| 85 |
|
| 86 |
-
Patches are smaller image crops which have been assigned a label. The patches are also labelled into `Deploy
|
| 87 |
|
| 88 |
|
| 89 |
### Annotations
|
|
@@ -113,12 +118,8 @@ If this dataset or the associated repository contributes to your research, we as
|
|
| 113 |
|
| 114 |
## Dataset Card Authors
|
| 115 |
|
| 116 |
-
Dr Scarlett Raine, QUT Centre for Robotics
|
| 117 |
-
|
| 118 |
-
Dr Benjamin Moshirian, Australian Institute of Marine Science
|
| 119 |
-
|
| 120 |
-
Dr Tobias Fischer, QUT Centre for Robotics
|
| 121 |
|
| 122 |
-
|
| 123 |
|
| 124 |
-
Dr
|
|
|
|
| 20 |
|
| 21 |
This dataset provides imagery used for training and evaluation of models in the Reef Guidance System. All imagery was collected by the Australian Institute of Marine Science using the ReefScan™ Transom Marine Monitoring System.
|
| 22 |
|
| 23 |
+
## Acknowledgement of Country
|
| 24 |
+
|
| 25 |
+
We acknowledge the continuing Sea Country management and custodianship of the Great Barrier Reef by Aboriginal and Torres Strait Islander Traditional Owners whose rich cultures, heritage values, enduring connections and shared efforts protect the Reef for future generations.
|
| 26 |
+
The Queensland University of Technology (QUT) acknowledges the Turrbal and Yugara, as the First Nations owners of the lands where QUT now stands. We pay respects to their Elders, lores, customs and creation spirits. We recognise that these lands have always been places of teaching, research and learning. QUT acknowledges the important role Aboriginal and Torres Strait Islander people play within the QUT community.
|
| 27 |
+
|
| 28 |
+
|
| 29 |
## Dataset Details
|
| 30 |
|
| 31 |
This dataset contains high resolution images of the seafloor.
|
|
|
|
| 48 |
- **Paper:** [Arxiv](link here)
|
| 49 |
|
| 50 |
To read more about the Reef Restoration and Adaptation Program: [RRAP](https://gbrrestoration.org/program/translation-to-deployment/)
|
| 51 |
+
|
| 52 |
To read more about the ReefScan™ Transom Marine Monitoring System [ReefScan™](https://www.aims.gov.au/research/technology/reefscan/reefscan-transom)
|
| 53 |
|
| 54 |
### Structure
|
|
|
|
| 59 |
|
| 60 |
All images were collected on the Great Barrier Reef in Australia.
|
| 61 |
|
|
|
|
|
|
|
| 62 |
The table below specifies the locations, depth ranges and labelling subsets for each of the 5 sites included in this dataset.
|
| 63 |
|
| 64 |
| Site | Location | Depth Range | Deployment Sequences | Patches | Weak Labels |
|
|
|
|
| 84 |
- Depth: altitude or distance to seafloor
|
| 85 |
- Camera properties: focal length, exposure time, f-stop, ISO speed
|
| 86 |
|
| 87 |
+
**Deployment Sequences** are sequential frames to match the real coral dispensing scenario, labelled at the image-level as `Deploy' or `No-Deploy'.
|
| 88 |
|
| 89 |
+
**Weak Labels** are whole images that have been selected by our annotator to best represent the class i.e. the majority of the image matches the label. These images are labelled into `Deploy`, `No-Deploy` or `Coral`. The `Coral` class is also considered as a case in which we would not dispense a coral device. We include this extra class to facilitate a parallel task for mapping coral coverage.
|
| 90 |
|
| 91 |
+
**Patches** are smaller image crops which have been assigned a label. The patches are also labelled into `Deploy`, `No-Deploy` and `Coral`.
|
| 92 |
|
| 93 |
|
| 94 |
### Annotations
|
|
|
|
| 118 |
|
| 119 |
## Dataset Card Authors
|
| 120 |
|
| 121 |
+
**Dataset Contact:** [Dr Scarlett Raine](https://www.qut.edu.au/about/our-people/academic-profiles/sg.raine), QUT Centre for Robotics, email: [email protected]
|
|
|
|
|
|
|
|
|
|
|
|
|
| 122 |
|
| 123 |
+
[Dr Benjamin Moshirian](https://gbrrestoration.org/research-team/benjamin-moshirian/), Australian Institute of Marine Science
|
| 124 |
|
| 125 |
+
[Dr Tobias Fischer](https://www.qut.edu.au/about/our-people/academic-profiles/tobias.fischer), QUT Centre for Robotics
|