CVApr 3, 2018

A Modified Image Comparison Algorithm Using Histogram Features

arXiv:1804.01142v11 citations
Originality Synthesis-oriented
AI Analysis

This work addresses image comparison for applications like retrieval or analysis, but it is incremental as it builds on existing histogram-based methods.

The paper tackled the problem of color image content comparison by proposing a modified method that uses color histograms and considers color locations, achieving 97% average precision on a collection of about 700 images while maintaining scale and rotation invariance.

This article discuss the problem of color image content comparison. Particularly, methods of image content comparison are analyzed, restrictions of color histogram are described and a modified method of images content comparison is proposed. This method uses the color histograms and considers color locations. Testing and analyzing of based and modified algorithms are performed. The modified method shows 97% average precision for a collection containing about 700 images without loss of the advantages of based method, i.e. scale and rotation invariant.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes