Back to Search Start Over

A Deep Learning Approach for Molecular Classification Based on AFM Images

A Deep Learning Approach for Molecular Classification Based on AFM Images

Authors :
Universidad de Sevilla. Departamento de Física Aplicada I
Comunidad de Madrid Industrial Doctorate Programme 2017 under reference number IND2017/IND-7793
Spanish MINECO Nd AEI/FEDER, UE project MAT2017-83273-R
Spanish Ministry of Science and Innovation, through the “María de Maeztu” Programme for Units of Excellence in R&D CEX2018-000805-M
Carracedo-Cosme, Jaime
Romero-Muñiz, Carlos
Pérez Pérez, Rubén
Universidad de Sevilla. Departamento de Física Aplicada I
Comunidad de Madrid Industrial Doctorate Programme 2017 under reference number IND2017/IND-7793
Spanish MINECO Nd AEI/FEDER, UE project MAT2017-83273-R
Spanish Ministry of Science and Innovation, through the “María de Maeztu” Programme for Units of Excellence in R&D CEX2018-000805-M
Carracedo-Cosme, Jaime
Romero-Muñiz, Carlos
Pérez Pérez, Rubén
Publication Year :
2021

Abstract

In spite of the unprecedented resolution provided by non-contact atomic force microscopy (AFM) with CO-functionalized and advances in the interpretation of the observed contrast, the unambiguous identification of molecular systems solely based on AFM images, without any prior information, remains an open problem. This work presents a first step towards the automatic classification of AFM experimental images by a deep learning model trained essentially with a theoretically generated dataset. We analyze the limitations of two standard models for pattern recognition when applied to AFM image classification and develop a model with the optimal depth to provide accurate results and to retain the ability to generalize. We show that a variational autoencoder (VAE) provides a very efficient way to incorporate, from very few experimental images, characteristic features into the training set that assure a high accuracy in the classification of both theoretical and experimental images.

Details

Database :
OAIster
Notes :
English
Publication Type :
Electronic Resource
Accession number :
edsoai.on1458371621
Document Type :
Electronic Resource