Machine Learning and Convolutional Neural Networks – Systems Based on Identification of Silicon Carbide Defects

DOI

10.2991/978-94-6463-986-5_49How to use a DOI?

Keywords

Silicon carbide crystals; Crystal defect detection; Cconvolutional neural networks (CNN); Machine learning (ML); Pre-trained model architecture

Abstract

Silicon carbide, which is a material we all call third-generation semiconductors, has special potential in 5G base stations and new energy vehicle charging. To make a better quality silicon carbide crystal, the key is to find out what defects it has. But now the factory mainly relies on manual inspection of these defects, this method is not reliable, often in the case of inequality or leakage. Better yet, we combine the method of the model that has been trained with incremental training, which can make the model more accurate and stable. Experiments have found that this method is particularly good, especially when classifying crystal defects, and even those small structural problems can be found. This method should be of great use in the fields of semiconductor production and optoelectronics. If the factory can use this automatic detection system, it can greatly improve the detection speed, reduce the error caused by manual operation, and reduce the production cost. It will be particularly easy to use in production, both flexible and accurate.

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Lei Zhang
PY  - 2026
DA  - 2026/02/18
TI  - Machine Learning and Convolutional Neural Networks – Systems Based on Identification of Silicon Carbide Defects
BT  - Proceedings of the 2025 International Conference on Electronics, Electrical and Grid Technology (ICEEGT 2025)
PB  - Atlantis Press
SP  - 475
EP  - 482
SN  - 2352-5401
UR  - https://doi.org/10.2991/978-94-6463-986-5_49
DO  - 10.2991/978-94-6463-986-5_49
ID  - Zhang2026
ER  -