arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2308.09693cs.CVcs.LGeess.IV
收藏

用于更快速、鲁棒的 EBSD 数据采集的轻量级 Transformer

A Lightweight Transformer for Faster and Robust EBSD Data Collection

Carnegie Mellon University(卡内基梅隆大学);Air Force Research Laboratory(美国空军研究实验室)

Harry Dong, Sean Donegan, Megna Shah, Yuejie Chi

AI总结:

针对 3D EBSD 连续切片采集易缺失切片的问题,提出轻量级 Transformer 加投影算法的两步恢复方法,仅用合成数据自监督训练即在真实数据上取得更优精度。

摘要:

三维电子背散射衍射(EBSD)显微成像是材料科学众多应用中的关键工具,然而在艰难的采集过程中,尤其是通过连续切片法采集时,其数据质量可能大幅波动。幸运的是,3D EBSD 数据本质上具有序列性,这为使用 Transformer 进行数据处理和恢复提供了机会;Transformer 是最先进的深度学习架构,已在众多领域取得突破。为了增强对错误的鲁棒性并加快这种 3D EBSD 数据采集,我们提出了一种两步方法,用于恢复 3D EBSD 体数据中的缺失切片:该方法使用高效的 Transformer 模型,并采用投影算法处理 Transformer 的输出。为克服在高维数据稀缺情况下进行深度学习所面临的计算和实践障碍,我们仅使用合成 3D EBSD 数据并以自监督方式训练该模型;与现有方法相比,该模型在真实 3D EBSD 数据上获得了更优的恢复精度。

英文摘要:

Three dimensional electron back-scattered diffraction (EBSD) microscopy is a critical tool in many applications in materials science, yet its data quality can fluctuate greatly during the arduous collection process, particularly via serial-sectioning. Fortunately, 3D EBSD data is inherently sequential, opening up the opportunity to use transformers, state-of-the-art deep learning architectures that have made breakthroughs in a plethora of domains, for data processing and recovery. To be more robust to errors and accelerate this 3D EBSD data collection, we introduce a two step method that recovers missing slices in an 3D EBSD volume, using an efficient transformer model and a projection algorithm to process the transformer's outputs. Overcoming the computational and practical hurdles of deep learning with scarce high dimensional data, we train this model using only synthetic 3D EBSD data with self-supervision and obtain superior recovery accuracy on real 3D EBSD data, compared to existing methods.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑