2026-10-11T01:31:10Z https://www.scidb.cn/oai

10.57760/sciencedb.j00186.00893 2026-04-02T09:23:16Z

2026-04-02 User-Guided Interactive Machine Learning for High-Throughput, Multi-Scale Helium Bubble Segmentation and Quantification doi:10.57760/sciencedb.j00186.00893 en This project introduces a novel User-Guided Interactive Machine Learning Framework for high-throughput, multi-scale helium bubble segmentation and quantification. Combining advanced image processing techniques with deep learning, the framework leverages Mask R-CNN for semantic segmentation and incorporates GCM for refined analysis. This tool provides a scalable and customizable approach for helium bubble analysis, empowering researchers with precise, user-friendly, and interactive workflows. Multi-Scale Segmentation; Quantification Module; High-Throughput Analysis Zhiwei Zheng PUBLIC https://creativecommons.org/licenses/by-nc-nd/4.0/ dataset Science Data Bank