A Benchmark Image Dataset for Scabies Detection_2.0

This dataset contains clinical skin images for automated scabies detection using deep learning and computer vision techniques. The dataset has been prepared to support research in dermatological image analysis and to accompany a Data in Brief publication.

The dataset is organized into two main categories:

Original Data – Raw collected skin images

Augmented Data – Artificially expanded images generated from the original dataset to improve model robustness

Each category includes two classes:

Scabies

Healthy

All images are resized to atleast 512*512 and stored in JPG format to ensure compatibility with standard deep learning frameworks.

The original images were collected under a strict privacy protocol. Only affected regions were captured. No facial features or personally identifiable information are included in the dataset. All data have been anonymized prior to release.

Data augmentation techniques such as rotation, flipping, scaling, and brightness variation were applied to increase dataset diversity and enhance generalization performance in machine learning models. Augmented images are stored separately from the original images to maintain transparency and reproducibility.

This dataset is intended for:

Deep learning research

Medical image classification

Skin disease detection benchmarking

Computer vision applications in healthcare

The dataset may be used for academic and research purposes in accordance with the specified license.

If this dataset is used in research, please cite the associated Data in Brief publication.