Machine Learning of Porosity Analysis of Sintered Ceramics

Tan, Evan Joon Juan (2026) Machine Learning of Porosity Analysis of Sintered Ceramics. Final Year Project (Bachelor), Tunku Abdul Rahman University of Management and Technology.

Text EVAN TAN JOON JUAN_Full Text.pdf Restricted to Registered users only Download (17MB)

Abstract

This thesis explores the application of artificial intelligence, specifically deep learning techniques, for the automated detection and analysis of pores in hydroxyapatite, a bio-ceramic material extensively used in biomedical engineering. Accurate characterization of porosity in hydroxyapatite is critical, as it significantly influences the material’s mechanical strength, bioactivity, and overall performance in biomedical implants. Traditional pore detection methods such as Mercury Intrusion Porosimetry (MIP), X-ray Computed Tomography (CT), and gas adsorption techniques are often resource-intensive, potentially destructive, or limited in accessibility. To address these limitations, this research proposes a deep learning-based image analysis approach utilizing convolutional neural networks (CNNs) and the YOLO (You Only Look Once) family of object detection algorithms. The methodology encompasses the preparation and annotation of image datasets, model training, evaluation on validation and unseen data, and subsequent use of the trained models to extract quantitative pore information. The outcomes validate the potential of AI-driven techniques as a non-destructive and scalable alternative for pore analysis in materials science. This work contributes to the advancement of intelligent systems in microstructural characterization and offers a foundation for further research in automated materials evaluation