Machine Learning for Automatic Identification of Pores in Microscopic Images of Weathered Sandstone

Sep 17, 2026

Uploaded by Theodora

Pores act as channels for water, air, and microorganisms, accelerating the weathering of rock and altering its physical and mechanical properties. Microscopic pore-structure analysis is therefore essential for stone cultural heritage. This paper presents a multi-scale pore characterization method using backscattered scanning electron microscope (BSE) images with feature engineering and machine learning. Pixel-level 16 features are constructed and evaluated across four models to separate pores from mineral matrix.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages7

Reading Time18 min

AccessFree to download

Introduction

Significance of Yungang Grottoes and sandstone weathering

Role of microscopic pore structure

Limitations of traditional analysis methods

Machine learning-based pore identification

Data

Machine Learning for Automatic Identification of Pores in Microscopic Images of Weathered Sandstone from Yungang Grottoes Yuan Cheng 1,2, Jiamei Xue 1,2, Xiuwei Guo 1,2,3, Yue Zhang 1,2, Jizhong Huang 1,2, Haomin Yu 4 1 Institute for the Conservation of Cultural Heritage, School of Cultural Heritage and Information Management, Shanghai University, Shanghai, China, 200444 – [xjmshu@shu.edu.cn](xjmshu@shu.edu.cn), [chengyuan@shu.edu.cn](chengyuan@shu.edu.cn), [guoxw@shu.edu.cn](guoxw@shu.edu.cn), [2019zhangy@shu.edu.cn](2019zhangy@shu.edu.cn), [hjizhong@shu.edu.cn](hjizhong@shu.edu.cn) 2 Key Laboratory of Silicate Cultural Relics Conservation (Shanghai University), Ministry of Education – [xjmshu@shu.edu.cn](xjmshu@shu.edu.cn), [chengyuan@shu.edu.cn](chengyuan@shu.edu.cn), [guoxw@shu.edu.cn](guoxw@shu.edu.cn), [2019zhangy@shu.edu.cn](2019zhangy@shu.edu.cn), [hjizhong@shu.edu.cn](hjizhong@shu.edu.cn) 3 School of Mechanics and Engineering Science, Shanghai University, Shanghai, China, 200444 – [guoxw@shu.edu.cn](guoxw@shu.edu.cn) 4 Department of Computer Science, Aalborg University, Aalborg 9220, Denmark – [haominyu@cs.aau.dk](haominyu@cs.aau.dk) Keywords: Stone Cultural Heritage, Pore Structure, Machine Learning, Backscattered Electron, Image Segmentation. Abstract Pores, as one of the channels for water, air, and microorganisms to enter the rock, accelerate the rock weathering process and change the physical and mechanical properties of the rock. Therefore, the study of the microscopic pore structure is of great significance for stone cultural heritage. This paper proposes a multi-scale pore structure characterization method based on backscattered scanning electron microscope (BSE) images, integrating feature engineering and machine learning techniques. First, for each pixel, 16 features are generated to construct a feature engineering. Then, these features are input into four commonly used machine learning models (Random Forest, Support Vector Machine, Multi-Layer Perceptron, Gradient Boosting Decision Tree) to distinguish pore and mineral matrix. To verify the effectiveness of the multi-channel feature input method, we compared the confusion matrix parameters (accuracy, precision, recall, and F1-score) of the four models before and after adding the feature engineering, and found that the indicators of all models increased by 1%-15% . In addition, the study also found that the Random Forest model performed the best. It can effectively segment the pores in different new images, thus could be directly applied to weathered sandstone images, providing data support for the scientific protection of stone cultural heritages. 1. Introduction As a famous Buddhist sculptural site, the Yungang Grottoes were inscribed on the UNESCO World Heritage List in 2001, reflecting their significant historical, artistic, and cultural value. Yungang Grottoes is mainly composed of sandstone. Due to natural weathering process and anthropogenic influences, invaluable historical information on the surfaces of many grottoes has been lost (Liu et al., 2011) . Therefore, it is of great significance to understand the weathering mechanisms of the grottoes, evaluate the degree of grotto weathering, and thus formulate effective conservation and restoration strategies accordingly. The microscopic pore structure of sandstone plays a crucial role in the weathering process of cultural heritage. These pores provide pathways for water, air, and microorganisms to infiltrate the rock interior, thereby influencing water and heat transfer within the rock matrix (Hao et al., 2022; Zhao et al., 2024) . The pore structure also determines the permeability and mechanical properties of sandstone. Furthermore, pore characteristics are established parameters for quantifying the degree of rock weathering (Zhang et al., 2024a) . Consequently, analysis of the microscopic pore structure in weathered sandstone is essential for elucidating weathering mecha

FAQs

Why is studying microscopic pore structure important for stone cultural heritage?

Because pores enable water, air, and microorganisms to penetrate the rock, they accelerate weathering and change physical and mechanical properties, which directly affects conservation decisions.

What imaging and modeling approach does the paper propose?

It uses backscattered scanning electron microscope (BSE) images and a multi-scale method combining feature engineering with machine learning models to distinguish pores from the mineral matrix.

How does adding feature engineering affect model performance?

Confusion-matrix metrics (accuracy, precision, recall, F1-score) increase by about 1%–15% for all four evaluated models after feature engineering is added, and the Random Forest model performs best.

Sep 17, 2026

Uploaded by Theodora