Deep learning for enhancement of low-resolution and noisy scanning probe microscopy images

• Samuel Gelman,
• Irit Rosenhek-Goldian,
• Nir Kampf,
• Marek Patočka,
• Maricarmen Rios,
• Marcos Penedo,
• Georg Fantner,
• Amir Beker,
• Sidney R. Cohen and
• Ido Azuri

Beilstein J. Nanotechnol. 2025, 16, 1129–1140, doi:10.3762/bjnano.16.83

• , RDN, and EDSR) [27][28][29][30][31]. The model architectures are built of residual convolutional neural network blocks while each of them integrates a unique algorithmic component (see “Methods”, “Computational Pipeline” section: “Traditional methods and deep learning super-resolution models”). For

The round-robin approach applied to nanoinformatics: consensus prediction of nanomaterials zeta potential

• Dimitra-Danai Varsou,
• Arkaprava Banerjee,
• Joyita Roy,
• Kunal Roy,
• Giannis Savvas,
• Haralambos Sarimveis,
• Ewelina Wyrzykowska,
• Mateusz Balicki,
• Tomasz Puzyn,
• Georgia Melagraki,
• Iseult Lynch and
• Antreas Afantitis

Beilstein J. Nanotechnol. 2024, 15, 1536–1553, doi:10.3762/bjnano.15.121

• convolutional neural network to predict the ZP of 119 NPs based on their nanostructure images. The abovementioned studies are indicative examples of models that have been used for the computational assessment of NPs ZP. As research progresses, such models are expected to become increasingly sophisticated and

The role of convolutional neural networks in scanning probe microscopy: a review

• Ido Azuri,
• Irit Rosenhek-Goldian,
• Neta Regev-Rudzki,
• Georg Fantner and
• Sidney R. Cohen

Beilstein J. Nanotechnol. 2021, 12, 878–901, doi:10.3762/bjnano.12.66

• algorithms applied are not part of the deep learning family of algorithms. The most widely used neural network type in image analysis is the convolutional neural network (CNN) [26][28][29][30][31]. CNN uses several manipulations to reduce the demand on computing resources and increase efficiency, as will be

• computation to circumvent difficulties in sufficient data acquisition is to train deep convolutional neural networks on simulated data. Deep STORM is a deep convolutional neural network that achieves high-resolution images from images recorded by a standard inverted microscope under bad imaging conditions

Reducing molecular simulation time for AFM images based on super-resolution methods

• Zhipeng Dou,
• Jianqiang Qian,
• Yingzi Li,
• Rui Lin,
• Jianhai Wang,
• Peng Cheng and
• Zeyu Xu

Beilstein J. Nanotechnol. 2021, 12, 775–785, doi:10.3762/bjnano.12.61

• can be used to speed up the generation of training data and vary simulation resolution for AFM machine learning. Keywords: atomic force microscopy; Bayesian compressed sensing; convolutional neural network; molecular dynamics simulation; super resolution; Introduction Atomic force microscopy methods

• addition to using the signal sparsity, learning-based super-resolution reconstruction methods are also focused issues and trends in research. The super-resolution convolutional neural network (SRCNN) is the first neural network model for super-resolution reconstruction tasks [35]. The SRCNN consists of a

• three-layer convolutional neural network which directly learns an end-to-end mapping between low- and high-resolution images, making the reconstructed image as close to the original image as possible. Generally, increasing the network depth could improve the reconstruction accuracy. With the development