journal article·2026-07-16

Study on the Identification of Wavy Features of Si3N4 Wafer Microdamage With Improved K‐Means Clustering by Incorporating the Centroid Growth Mechanism

Dongling Yu, Xianqi Liao, Zhixin Huang, Wenjie Dong, Chunjing Xu, Kun Hu, Nanxing Wu

Publisher page

Abstract

To address the problems of irregular wavy edge contours, blurred edges, and notch misidentification in microscratch damage on Si3N4 wafers, an improved K‐means clustering segmentation method with centroid growth fusion is proposed. This method aims to achieve complete segmentation of the wavy edge features of microscratch damage on Si3N4 wafers. Through in‐depth analysis of the intermittent notch characteristics of the wavy edge features of Si3N4 wafers, the initial clustering center is set based on the Euclidean distance probability model and combined with the image hierarchical clustering framework to achieve effective separation of image content. An image feature connectivity assessment mechanism is also introduced to enhance the extraction accuracy of target features. The experimental results show that the precision of the proposed method reaches 97.76% and the MAE reaches 3.65 in the comparison experiment with the real label image. This method effectively overcomes the interference of defect segmentation caused by irregular wavy edge contour, blurred edge, and misidentification of notch, and significantly improves the complete extraction ability of wavy edge feature defects of Si3N4 wafers. It provides key technical support for its promotion and application in high‐end manufacturing fields such as semiconductors.
Source: crossref · Quality: baseline

Research context

OpenAlex topics: Industrial Vision Systems and Defect Detection · Surface Roughness and Optical Measurements · Silicon and Solar Cell Technologies

SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.

Affiliations

School of Mechanical and Electronic Engineering Jingdezhen Ceramic University Jingdezhen Jiangxi China

Laboratory of Ceramic Material Processing Technology Engineering Jingdezhen Jiangxi China

Aecc Aero Science And Technology Co., Ltd. Chengdu Sichuan China

Funding

National Natural Science Foundation of China

References (21)

Show source-provided references
1. Facet‐Dependent Electrical Conductivity Properties of a 4H‐SiC Wafer10.1039/D2TC01981G ↗
2. Improvement in Surface Properties of Si Wafer by Mechanical Surface Treatment10.1016/j.jpcs.2019.109272 ↗
3. Ultra‐Thin Wafer Technology and Applications: A Review10.1016/j.mssp.2019.104681 ↗
4. Semiconductor Wafer Bonding for Solar Cell Applications: A Review10.1002/aesr.202300073 ↗
5. Investigation on Skidding Characteristics and Dynamic Responses of Rolling Element Bearing under Thermal Elastohyrdodynamic Lubrication10.1016/j.ymssp.2026.114403 ↗
6. Wafer‐to‐Wafer Bonding Fabrication Process‐Induced Wafer Warpage10.1109/TSM.2023.3284007 ↗
7. A Generalised Uncertain Decision Tree for Defect Classification of Multiple Wafer Maps10.1080/00207543.2019.1637035 ↗
8. A Unified Defect Pattern Analysis of Wafer Maps Using Density‐Based Clustering10.1109/ACCESS.2021.3084221 ↗
9. Similarity Matching of Wafer Bin Maps for Manufacturing Intelligence to Empower Industry 3.5 for Semiconductor Manufacturing10.1016/j.cie.2020.106358 ↗
10. Defect Engineering of p‐Type Silicon Heterojunction Solar Cells Fabricated Using Commercial‐Grade Low‐Lifetime Silicon Wafers10.1002/pip.3230 ↗
11. Wafer Defect Inspection Optimization With Partial Coverage—A Numerical Approach10.1109/TASE.2020.3024651 ↗
12. Detecting Material Defects Using a Scanning Laser Beam10.2514/1.T5464 ↗
13. Microcrack Damage Characterization of Aerospace Ceramic Bearings Based on Global Feature Stitching and Illumination‐Balanced Enhancement10.1016/j.ymssp.2026.114388 ↗
14. A Novel Fuzzy Clustering Based Method for Image Segmentation in RGB‐D Images10.1016/j.engappai.2022.104709 ↗
15. A Method for Classifying Citrus Surface Defects Based on Machine Vision10.1007/s11694-021-00868-z ↗
16. Segmentation of Tomato Leaf Images Based on Adaptive Clustering Number of K‐Means Algorithm10.1016/j.compag.2019.104962 ↗
17. An Efficient Split‐Merge Re‐Start for the K‐Means Algorithm
18. Investigations of Color Image Segmentation Based on Connectivity Measure, Shape Priority and Normalized Fuzzy Graph Cut10.1016/j.asoc.2023.110239 ↗
19. An IoT Based Modified Graph Cut Segmentation with Optimized Adaptive Connectivity and Shape Priors
20. Unsupervised Object Discovery with Pseudo Label Generated Using K‐Means and Self‐Supervised Transformer10.1016/j.neucom.2023.126326 ↗
21. K‐Means Algorithm Based on Flower Pollination Algorithm and Calinski‐Harabasz Index