Next-generation deep learning based on simulators and synthetic data

2021 ◽  
Author(s):  
Celso M. de Melo ◽  
Antonio Torralba ◽  
Leonidas Guibas ◽  
James DiCarlo ◽  
Rama Chellappa ◽  
...  
Keyword(s):  
Deep Learning ◽  
Synthetic Data ◽  
Next Generation

scholarly journals Real UAV-Bird Image Classification Using CNN with a Synthetic Dataset

2021 ◽  
Vol 11 (9) ◽  
pp. 3863
Author(s):  
Ali Emre Öztürk ◽  
Ergun Erçelebi
Keyword(s):  
Deep Learning ◽  
Image Classification ◽  
Synthetic Data ◽  
Real Data ◽  
Corner Detection ◽  
Batch Size ◽  
Test Accuracy ◽  
Classification Problems ◽  
Auc Value ◽  
Classification Test

A large amount of training image data is required for solving image classification problems using deep learning (DL) networks. In this study, we aimed to train DL networks with synthetic images generated by using a game engine and determine the effects of the networks on performance when solving real-image classification problems. The study presents the results of using corner detection and nearest three-point selection (CDNTS) layers to classify bird and rotary-wing unmanned aerial vehicle (RW-UAV) images, provides a comprehensive comparison of two different experimental setups, and emphasizes the significant improvements in the performance in deep learning-based networks due to the inclusion of a CDNTS layer. Experiment 1 corresponds to training the commonly used deep learning-based networks with synthetic data and an image classification test on real data. Experiment 2 corresponds to training the CDNTS layer and commonly used deep learning-based networks with synthetic data and an image classification test on real data. In experiment 1, the best area under the curve (AUC) value for the image classification test accuracy was measured as 72%. In experiment 2, using the CDNTS layer, the AUC value for the image classification test accuracy was measured as 88.9%. A total of 432 different combinations of trainings were investigated in the experimental setups. The experiments were trained with various DL networks using four different optimizers by considering all combinations of batch size, learning rate, and dropout hyperparameters. The test accuracy AUC values for networks in experiment 1 ranged from 55% to 74%, whereas the test accuracy AUC values in experiment 2 networks with a CDNTS layer ranged from 76% to 89.9%. It was observed that the CDNTS layer has considerable effects on the image classification accuracy performance of deep learning-based networks. AUC, F-score, and test accuracy measures were used to validate the success of the networks.

scholarly journals A Novel Deep Learning-Based Diagnosis Method Applied to Power Quality Disturbances

Energies ◽  
2021 ◽  
Vol 14 (10) ◽  
pp. 2839
Author(s):  
Artvin-Darien Gonzalez-Abreu ◽  
Miguel Delgado-Prieto ◽  
Roque-Alfredo Osornio-Rios ◽  
Juan-Jose Saucedo-Dorantes ◽  
Rene-de-Jesus Romero-Troncoso
Keyword(s):  
Deep Learning ◽  
Power Quality ◽  
Electrical Power ◽  
Synthetic Data ◽  
Industrial Sector ◽  
Industrial Facilities ◽  
Diagnosis Method ◽  
Neural Network Structure ◽  
Power Quality Disturbances ◽  
Disturbance Detection

Monitoring electrical power quality has become a priority in the industrial sector background: avoiding unwanted effects that affect the whole performance at industrial facilities is an aim. The lack of commercial equipment capable of detecting them is a proven fact. Studies and research related to these types of grid behaviors are still a subject for which contributions are required. Although research has been conducted for disturbance detection, most methodologies consider only a few standardized disturbance combinations. This paper proposes an innovative deep learning-based diagnosis method to be applied on power quality disturbances, and it is based on three stages. Firstly, a domain fusion approach is considered in a feature extraction stage to characterize the electrical power grid. Secondly, an adaptive pattern characterization is carried out by considering a stacked autoencoder. Finally, a neural network structure is applied to identify disturbances. The proposed approach relies on the training and validation of the diagnosis system with synthetic data: single, double and triple disturbances combinations and different noise levels, also validated with available experimental measurements provided by IEEE 1159.2 Working Group. The proposed method achieves nearly a 100% hit rate allowing a far more practical application due to its capability of pattern characterization.

scholarly journals A deep learning approach for synthetic MRI based on two routine sequences and training with synthetic data

2021 ◽  
Vol 210 ◽  
pp. 106371
Author(s):  
Elisa Moya-Sáez ◽  
Óscar Peña-Nogales ◽  
Rodrigo de Luis-García ◽  
Carlos Alberola-López
Keyword(s):  
Deep Learning ◽  
Synthetic Data ◽  
Learning Approach ◽  
And Training ◽  
Synthetic Mri

In-Cabin vehicle synthetic data to test Deep Learning based human pose estimation models

2021 ◽  
Author(s):  
Luis Gustavo Tomal Ribas ◽  
Marta Pereira Cocron ◽  
Joed Lopes Da Silva ◽  
Alessandro Zimmer ◽  
Thomas Brandmeier
Keyword(s):  
Deep Learning ◽  
Pose Estimation ◽  
Synthetic Data ◽  
Human Pose Estimation ◽  
Human Pose ◽  
Estimation Models

REAL-TIME 2.5D INVERSION OF LWD RESISTIVITY MEASUREMENTS USING DEEP LEARNING FOR GEOSTEERING APPLICATIONS ACROSS FAULTED FORMATIONS

2021 ◽  
Author(s):  
Kyubo Noh ◽  
◽  
Carlos Torres-Verdín ◽  
David Pardo ◽  
◽  
...  
Keyword(s):  
Deep Learning ◽  
Real Time ◽  
Fault Plane ◽  
Synthetic Data ◽  
Field Measurements ◽  
Geological Structure ◽  
Inversion Method ◽  
Adaptive Finite Element ◽  
Resistivity Measurements ◽  
Lwd Resistivity

We develop a Deep Learning (DL) inversion method for the interpretation of 2.5-dimensional (2.5D) borehole resistivity measurements that requires negligible online computational costs. The method is successfully verified with the inversion of triaxial LWD resistivity measurements acquired across faulted and anisotropic formations. Our DL inversion workflow employs four independent DL architectures. The first one identifies the type of geological structure among several predefined types. Subsequently, the second, third, and fourth architectures estimate the corresponding spatial resistivity distributions that are parameterized (1) without the crossings of bed boundaries or fault plane, (2) with the crossing of a bed boundary but without the crossing of a fault plane, and (3) with the crossing of the fault plane, respectively. Each DL architecture employs convolutional layers and is trained with synthetic data obtained from an accurate high-order, mesh-adaptive finite-element forward numerical simulator. Numerical results confirm the importance of using multi-component resistivity measurements -specifically cross-coupling resistivity components- for the successful reconstruction of 2.5D resistivity distributions adjacent to the well trajectory. The feasibility and effectiveness of the developed inversion workflow is assessed with two synthetic examples inspired by actual field measurements. Results confirm that the proposed DL method successfully reconstructs 2.5D resistivity distributions, location and dip angles of bed boundaries, and the location of the fault plane, and is therefore reliable for real-time well geosteering applications.

scholarly journals Deep-learning with synthetic data enables automated picking of cryo-EM particle images of biological macromolecules

2019 ◽  
Author(s):  
Ruijie Yao ◽  
Jiaqiang Qian ◽  
Qiang Huang
Keyword(s):  
Deep Learning ◽  
Single Particle ◽  
Feature Matching ◽  
Synthetic Data ◽  
Supplementary Information ◽  
Particle Analysis ◽  
Biological Macromolecules ◽  
Low Contrast ◽  
3D Structures ◽  
Particle Images

Abstract Motivation Single-particle cryo-electron microscopy (cryo-EM) has become a powerful technique for determining 3D structures of biological macromolecules at near-atomic resolution. However, this approach requires picking huge numbers of macromolecular particle images from thousands of low-contrast, high-noisy electron micrographs. Although machine-learning methods were developed to get rid of this bottleneck, it still lacks universal methods that could automatically picking the noisy cryo-EM particles of various macromolecules. Results Here, we present a deep-learning segmentation model that employs fully convolutional networks trained with synthetic data of known 3D structures, called PARSED (PARticle SEgmentation Detector). Without using any experimental information, PARSED could automatically segment the cryo-EM particles in a whole micrograph at a time, enabling faster particle picking than previous template/feature-matching and particle-classification methods. Applications to six large public cryo-EM datasets clearly validated its universal ability to pick macromolecular particles of various sizes. Thus, our deep-learning method could break the particle-picking bottleneck in the single-particle analysis, and thereby accelerates the high-resolution structure determination by cryo-EM. Availability and implementation The PARSED package and user manual for noncommercial use are available as Supplementary Material (in the compressed file: parsed_v1.zip). Supplementary information Supplementary data are available at Bioinformatics online.

Sign in / Sign up

Export Citation Format

Share Document