Pap Smear Cervical Cell Image Database

This repository contains Pap smear cervical cell images, segmentation results, and classification outputs developed for abnormal cervical cell nuclei identification using image processing and machine learning techniques.

Repository Contents

The repository contains:

1. Original Pap Smear Images

• RGB cervical cytology images.
• Image dimensions: 768 × 568 pixels.

2. K-Means Segmentation Results

Contains segmentation outputs generated using K-Means clustering.

3. Mean Shift Segmentation Results

Contains oversegmentation and clustering outputs generated using mean shift segmentation.

4. Classification Output Results

Contains:

• abnormal nuclei identification results,
• normal nuclei visualization results,
• cascade classification outputs.

Dataset Characteristics

• Imaging modality: Pap smear microscopy
• Application:
• Cervical cancer screening
• Abnormal nuclei identification
• Image format: RGB images
• Image dimensions:
• 768 × 568 pixels

Dataset Source

The dataset was provided by:

Byriel, J.
Neuro-Fuzzy Classification of Cells in Cervical Smear.
M.Sc. Thesis, Technical University of Denmark, Department of Automation, 1999.

The images provided in this dataset were generated in:

TELLO-MIJARES, Santiago.
Abnormal Cervical Cell Nuclei Identification Using Spatial-Context-Saliency Regions and Two-Level Cascade Classifiers.
Journal of Medical Imaging and Health Informatics, 2019, Vol. 9, No. 3, pp. 426–435.