High Resolution Fundus Image Database for Monomodal Single-Channel Image Registration

A database of 42 high-resolution image pairs derived from the High Resolution Fundus Image Database. It consists of grayscale intensity images, binary masks of the retinal area, and expert-annotated vessel segmentations, related by elastic deformations. The dataset was created by Johan Öfverstedt of Uppsala University.

Use Cases

• Developing image registration algorithms based on expert-annotated vessel segmentations.
• Testing similarity measures for images with thin linear structures that lack sufficient pixel overlap.
• Benchmarking monomodal, single-channel registration techniques using grayscale intensity images.
• Evaluating registration performance on retinal fundus images with provided binary masks.

Strengths

• Contains 42 high-resolution image pairs, providing a focused test set.
• Includes expert-annotated labels for vessels, offering a ground truth for evaluation.
• Provides binary masks of the retinal area alongside grayscale images.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• Last update date is unknown; freshness unverified.

Provenance

Source

Uppsala University

Collection Method

Created from original images from the High Resolution Fundus Image Database, with elastic deformations applied.