Standardize the SipakMED and Herlev datasets

[Meet2304]

Objective

Standardize the SipakMED and Herlev datasets to create a consistent and clean input pipeline for model training and fine-tuning.

Task Details

• Design a preprocessing pipeline to bring both datasets into a unified format (e.g., resolution, color space, labeling scheme).
• Implement basic image processing steps such as resizing, normalization, and noise reduction.
• Apply consistent data augmentation strategies (rotation, flipping, color jitter, etc.).
• Investigate and apply stain normalization or correction techniques to reduce variation across samples.
• Ensure compatibility with downstream models and training scripts.

Deliverables

• Standardized dataset (or a script to generate it)
• Documentation describing the preprocessing steps and rationale
• Visual samples of pre- and post-processing for both datasets