# General framework for testing Poisson-Voronoi assumption for real microstructures

**Authors:** Martina Vittorietti, Piet J. J. Kok, Jilt Sietsma, Wei Li, Geurt Jongbloed  
**arXiv:** [1902.10981](https://arxiv.org/abs/1902.10981)  
**Date:** 2019-02-28  
**Summary page:** https://www.summarizepaper.com/en/arxiv-id/1902.10981v1/

## Key points

- The paper addresses the problem of modeling microstructures in Materials Science, Mathematics, and Statistics.
- It focuses on the Poisson-Voronoi diagram as a basic model for steel microstructure.
- The objective is to develop methods to test whether a real steel microstructure can be accurately approximated by the Poisson-Voronoi model.
- A general framework for testing the assumption is proposed based on images of 2D sections of real metals.
- Two different approaches are included depending on the use or non-use of periodic boundary conditions.
- Three different model tests are proposed within these approaches.
- The first two tests analyze the coefficient of variation and cumulative distribution function of cells' area in microstructure images.
- These tests assess how well the Poisson-Voronoi model captures variability and distribution patterns observed in real steel microstructures.
- The third test utilizes tools from Topological Data Analysis, specifically persistence landscapes, to evaluate how well the model represents complex structural characteristics present in real steel microstructures.
- The paper contributes a comprehensive framework for testing the validity of using Poisson-Voronoi diagrams as models for steel microstructures.

## Summary

The paper titled "General framework for testing Poisson-Voronoi assumption for real microstructures" addresses the problem of modeling microstructures in Materials Science, Mathematics, and Statistics. Specifically, it focuses on the Poisson-Voronoi diagram as a basic model for steel microstructure. This diagram has attractive mathematical properties and has been used to approximate single phase steel microstructures. The objective of the paper is to develop methods that can be used to test whether a real steel microstructure can be accurately approximated by the Poisson-Voronoi model. To achieve this, the authors propose a general framework for testing the Poisson-Voronoi assumption based on images of 2D sections of real metals. The framework includes two different approaches depending on the use or non-use of periodic boundary conditions. Three different model tests are proposed within these approaches. The first two tests are based on analyzing the coefficient of variation and cumulative distribution function of the cells' area in the microstructure images. These tests aim to assess how well the Poisson-Voronoi model captures the variability and distribution patterns observed in real steel microstructures. In addition to these statistical measures, the third test utilizes tools from Topological Data Analysis, specifically persistence landscapes. This approach explores topological features in order to evaluate how well the Poisson-Voronoi model represents complex structural characteristics present in real steel microstructures. Overall, this paper contributes a comprehensive framework for testing the validity of using Poisson-Voronoi diagrams as models for steel microstructures. By employing various statistical and topological analysis techniques, researchers can gain insights into whether this simple yet mathematically appealing model accurately approximates real-world microstructural patterns found in steel materials.