Back to datasets
Dataset assetOpen Source CommunityMedical Image AnalysisBlood Cell Detection

BCCD Dataset

The BCCD dataset is a small‑scale collection for blood cell detection, containing three label types: red blood cells (RBC), white blood cells (WBC), and platelets. Images are in JPEG format with a resolution of 640 × 480 pixels, and annotations are provided in VOC‑style XML files.

Source
github
Created
Jan 18, 2019
Updated
Apr 3, 2020
Signals
245 views
Availability
Linked source ready
Overview

Dataset description and usage context

BCCD Dataset Overview

Description

  • Type: Blood cell detection dataset
  • Scale: Small
  • Source: Reconstructed from the original data and annotations of cosmicad and akshaylamba into VOC format
  • License: MIT License

Content

  • Label Types:
    • RBC (Red Blood Cell)
    • WBC (White Blood Cell)
    • Platelets
  • Structure:
    ├── BCCD
    │   ├── Annotations
    │   │       └── BloodImage_00XYZ.xml (364 items)
    │   ├── ImageSets
    │   └── JPEGImages
    │       └── BloodImage_00XYZ.jpg (364 items)
    ├── dataset
    │   └── mxnet
    ├── scripts
    │   ├── split.py
    │   └── visualize.py
    ├── example.jpg
    ├── LICENSE
    └── README.md
    
  • Image Info:
    • Format: JPEG
    • Size: 640 × 480

Annotation Format

  • VOC‑style .xml files
  • Example:
    <annotation>
    	...
    	<object>
    		<name>WBC</name>
    		...
    		<bndbox>
    			<xmin>260</xmin>
    			<ymin>177</ymin>
    			<xmax>491</xmax>
    			<ymax>376</ymax>
    		</bndbox>
    	</object>
    </annotation>
    
Need downstream help?

Pair the dataset with AI analysis and content workflows.

Once the source passes your review, move straight into summarization, transformation, report drafting, or presentation generation with the JuheAI toolchain.

Explore AI studio