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 2310.08302physics.data-ancond-mat.mtrl-sci
收藏

深度学习用于EELS高光谱图像解混——使用自编码器

Deep Learning for EELS hyperspectral images unmixing -- using autoencoders

N. Brun, G. Lambert, L. Bocher

AI总结:

本研究评估了基于自编码器的深度学习方法在STEM-EELS高光谱图像光谱解混中的性能,并与传统方法对比,结果表明自编码器具有巨大潜力,为探索更复杂的神经网络奠定了基础。

摘要:

在扫描透射电子显微镜(STEM)中进行的空间分辨电子能量损失谱(EELS)能够获取高光谱图像(HSIs)。光谱解混(SU)是将HSI中的每个光谱分解为对应于样品中存在的化合物的代表性光谱(端元)及其局部比例(丰度)的组合的过程。SU是一项复杂的任务,不同领域的研究者已开发了多种利用HSI的方法。然而,这些方法均未能完全满足STEM-EELS的要求。遥感领域的最新进展聚焦于深度学习技术,特别是自编码器(AEs),有潜力满足这些要求。在本研究中,评估了使用AE进行SU的深度学习方法的性能,并将其结果与传统方法进行了比较。为了定量评估解混过程的结果,创建了合成HSI并使用特定指标进行衡量。随后,将这些方法应用于一系列实验数据。研究结果表明,AE作为STEM-EELS SU工具的潜力巨大,标志着探索更复杂神经网络的一个起点。

英文摘要:

Spatially resolved Electron Energy-Loss Spectroscopy (EELS) conducted in a Scanning Transmission Electron Microscope (STEM) enables the acquisition of hyperspectral images (HSIs). Spectral unmixing (SU) is the process of decomposing each spectrum of an HSI into a combination of representative spectra (endmembers) corresponding to compounds present in the sample along with their local proportions (abundances). SU is a complex task, and various methods have been developed in different communities using HSIs. However, none of these methods fully satisfy the STEM-EELS requirements. Recent advancements in remote sensing, which focus on Deep Learning techniques, have the potential to meet these requirements, particularly Autoencoders (AEs). In this study, the performance of Deep Learning methods using AE for SU is evaluated, and their results are compared with traditional methods. Synthetic HSIs have been created to quantitatively assess the outcomes of the unmixing process using specific metrics. The methods are subsequently applied to a series of experimental data. The findings demonstrate the promising potential of AE as a tool for STEM-EELS SU, marking a starting point for exploring more sophisticated Neural Networks.

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。

↑