Chula RBC-12 Dataset

The Chula-RBC-12 dataset is a dataset of red blood cell (RBC) blood smear images used in the paper “Red Blood Cell Segmentation with Overlapping Cell Separation and Classification from an Imbalanced Dataset”. The dataset includes 12 types of red blood cells, a total of 706 blood smear images, containing over 20,000 red cells. It was collected in 2019 at the Oxidative Red Cell Disorder Research Unit of Chulalongkorn University, using a DS-Fi2-L3 Nikon microscope at 1000× magnification.

Red Blood Cell Types

• 0 Normal cell
• 1 Large red cell
• 2 Small red cell
• 3 Spherical red cell
• 4 Target-shaped cell
• 5 Mouth-shaped cell
• 6 Elliptical cell
• 7 Tear-shaped cell
• 8 Thorn cell
• 9 Fragment cell
• 10 Unclassified
• 11 Hypochromic cell
• 12 Elliptical cell

Dataset

• The "Dataset" folder contains 738 red blood cell blood smear images with resolution 640×480.
• The "Label" folder contains label data, with filenames corresponding to the respective images. Files may contain multiple lines, each line storing in the following order:
  • x coordinate
  • y coordinate
  • red blood cell type number

Citation

If you use this dataset, please cite the following paper:

@misc{naruenatthanaset2021red, title={Red Blood Cell Segmentation with Overlapping Cell Separation and Classification on Imbalanced Dataset}, author={Korranat Naruenatthanaset and Thanarat H. Chalidabhongse and Duangdao Palasuwan and Nantheera Anantrasirichai and Attakorn Palasuwan}, year={2021}, eprint={2012.01321}, archivePrefix={arXiv}, primaryClass={eess.IV} }