摘要
工作在超高真空环境下的扫描隧道显微镜(STM)具备原子级分辨率,广泛应用于材料表面结构的精细成像。然而,STM图像易受机械振动、电子噪声、环境扰动等多种因素影响,导致图像质量下降,严重制约其科学研究价值。为提升STM图像的可用性和精度,文章提出一种基于多尺度特征提取与注意力机制的深度学习图像修复模型——MAED-CNN。该模型采用U-Net编码−解码结构,融合多尺度卷积模块与通道注意力机制,并引入人工修复图像作为监督参考,有效增强对图像局部细节与全局结构的重构能力。在多个真实STM图像数据集上进行测试,MAED-CNN在PSNR、SSIM、UQI等评价指标上均优于现有主流图像修复模型,表现出更高的图像还原精度与稳定性。研究为STM图像智能修复提供了新思路,对提升纳米尺度成像技术的应用水平具有重要意义。
The Scanning Tunneling Microscope(STM),operating under ultra-high vacuum conditions,enables atomic-scale resolution imaging of material surfaces.However,STM images are often affected by various sources of noise,which degrades image quality.This paper proposes a deep learning model for STM image restoration,named MAED-CNN-Multi-scale Attention Encoder-Decoder Convolutional Neural Network.It uses artificially repaired STM images as references.The model leverages manually restored STM images as references and combines multi-scale convolution,attention modules,and an encoder-decoder U-Net architecture to transform noisy input images into high-quality,denoised outputs.Compared with several general deep learning models,the proposed model demonstrates superior performance in metrics such as PSNR,SSIM,and UQI.It effectively restores STM images and holds significant promise for advancing STM image restoration techniques and promoting research in imaging technologies.
作者
詹凌涛
范浩龙
张腾
王婷婷
曹雄柏
李燕
周贞如
张全震
杨惠霞
王业亮
ZHAN Lingtao;FAN Haolong;ZHANG Teng;WANG Tingting;CAO Xiongbai;LI Yan;ZHOU Zhenru;ZHANG Quanzhen;YANG Huixia;WANG Yeilang(School of Integrated Circuits and Electronics,Beijing Institute of Technology,Beijing Institute of Technology Yangtze River Delta Research Institute(Jiaxing),Key Laboratory of Low-Dimensional Quantum Structures and Devices,Beijing 100081,China)
出处
《真空科学与技术学报》
北大核心
2025年第8期686-695,共10页
Chinese Journal of Vacuum Science and Technology
基金
国家自然科学基金项目(62271048)。
关键词
深度学习
图像修复
扫描隧道显微镜
Deep Learning
Image Restoration
Scanning Tunneling Microscope
