kuwaaq/coral-health-classification

---
license: cc
---
# Coral health classification dataset

This dataset is a combination of three coral image datasets:
* Bleached and unbleached dataset (BU) by Jamil et al. [1]
* Bleached, healthy and dead dataset (BHD) by Jamil et al. [1]
* Dead subset of the EILAT dataset [2]

The labels of each image and source dataset is included in the filename i.e. "label_dataset_ID.png".

## More info about the dataset:
* Number of images: 1599
* Number of classes: 3 (healthy, unhealthy, dead)
* Healthy
* Number of images: 661
* 124 from BU
* 537 from BHD
* Unhealthy (bleached)
* Number of images: 508
* 134 from BU
* 374 from BHD
* Dead
* Number of images: 430
* 150 from BHD
* 280 from EILAT

---
References

[1] S. Jamil, M. Rahman, and A. Haider, “Bag of Features (BOF) based deep learning framework for bleached corals detection,” Big Data and Cognitive Computing, vol. 5, no. 4, p. 53, Oct. 2021, doi: 10.3390/bdcc5040053.

[2] A. Shihavuddin, N. Gracias, R. García, A. C. R. Gleason, and B. Gintert, “Image-Based coral reef classification and thematic mapping”, Remote Sensing, vol. 5, no. 4, pp. 1809–1841, Apr. 2013, doi: 10.3390/rs5041809.