Image Similarity Ranking

Project Overview

This project compares a set of digital images to a single reference image called original.jpg. It uses perceptual hashing to measure visual similarity and ranks all other images from the most similar to the least similar.

How to Read the Results

Each comparison is based on a distance score. A lower distance means that the image is more visually similar to the original. A higher distance means that the image is more visually different.

• 0 = nearly identical to the original
• 1–5 = very similar to the original
• 6–10 = somewhat similar to the original
• Above 10 = different from the original

Research Relevance

This project relates to digital humanities and digital art history by addressing how digital images circulate in multiple versions and how computational methods can help distinguish close variants from unrelated works. It demonstrates how perceptual hashing can support visual analysis, image version tracking, and broader questions of transformation and digital provenance.