scholarly journals An Application of Deep Neural Networks for Segmentation of Microtomographic Images of Rock Samples

Computers ◽  
2019 ◽  
Vol 8 (4) ◽  
pp. 72 ◽  
Author(s):  
Igor Varfolomeev ◽  
Ivan Yakimchuk ◽  
Ilia Safonov
Keyword(s):  
Neural Networks ◽  
Deep Neural Networks ◽  
Ground Truth ◽  
Indicator Kriging ◽  
Generalization Capability ◽  
Automatic Mode ◽  
Fine Structures ◽  
Segmentation Of Images ◽  
2D And 3D ◽  
Single Set

Image segmentation is a crucial step of almost any Digital Rock workflow. In this paper, we propose an approach for generation of a labelled dataset and investigate an application of three popular convolutional neural networks (CNN) architectures for segmentation of 3D microtomographic images of samples of various rocks. Our dataset contains eight pairs of images of five specimens of sand and sandstones. For each sample, we obtain a single set of microtomographic shadow projections, but run reconstruction twice: one regular high-quality reconstruction, and one using just a quarter of all available shadow projections. Thoughtful manual Indicator Kriging (IK) segmentation of the full-quality image is used as the ground truth for segmentation of images with reduced quality. We assess the generalization capability of CNN by splitting our dataset into training and validation sets by five different manners. In addition, we compare neural networks results with segmentation by IK and thresholding. Segmentation outcomes by 2D and 3D U-nets are comparable to IK, but the deep neural networks operate in automatic mode, and there is big room for improvements in solutions based on CNN. The main difficulties are associated with the segmentation of fine structures that are relatively uncommon in our dataset.

Deep neural networks for improving physical accuracy of 2D and 3D multi-mineral segmentation of rock micro-CT images

2021 ◽  
Vol 104 ◽  
pp. 107185 ◽  
Author(s):  
Ying Da Wang ◽  
Mehdi Shabaninejad ◽  
Ryan T. Armstrong ◽  
Peyman Mostaghimi
Keyword(s):  
Neural Networks ◽  
Deep Neural Networks ◽  
Ct Images ◽  
Micro Ct ◽  
2D And 3D

scholarly journals Particle-Based Fluid Surface Rendering with Neural Networks

2021 ◽  
Author(s):  
Viktória Burkus ◽  
Attila Kárpáti ◽  
László Szécsi
Keyword(s):  
Neural Networks ◽  
Deep Neural Networks ◽  
Ground Truth ◽  
Fluid Simulation ◽  
Image Manipulation ◽  
Surface Rendering ◽  
Fluid Surface ◽  
Image Pairs ◽  
The Cost

Surface reconstruction for particle-based fluid simulation is a computational challenge on par with the simula- tion itself. In real-time applications, splatting-style rendering approaches based on forward rendering of particle impostors are prevalent, but they suffer from noticeable artifacts. In this paper, we present a technique that combines forward rendering simulated features with deep-learning image manipulation to improve the rendering quality of splatting-style approaches to be perceptually similar to ray tracing solutions, circumventing the cost, complexity, and limitations of exact fluid surface rendering by replacing it with the flat cost of a neural network pass. Our solution is based on the idea of training generative deep neural networks with image pairs consisting of cheap particle impostor renders and ground truth high quality ray-traced images.

scholarly journals Video Super-Resolution Based on Generative Adversarial Network and Edge Enhancement

Electronics ◽  
2021 ◽  
Vol 10 (4) ◽  
pp. 459
Author(s):  
Jialu Wang ◽  
Guowei Teng ◽  
Ping An
Keyword(s):  
Neural Networks ◽  
Deep Neural Networks ◽  
Visual Experience ◽  
Super Resolution ◽  
Ground Truth ◽  
Intermediate Result ◽  
Edge Enhancement ◽  
Generative Adversarial Network ◽  
Adversarial Network ◽  
Sr Reconstruction

With the help of deep neural networks, video super-resolution (VSR) has made a huge breakthrough. However, these deep learning-based methods are rarely used in specific situations. In addition, training sets may not be suitable because many methods only assume that under ideal circumstances, low-resolution (LR) datasets are downgraded from high-resolution (HR) datasets in a fixed manner. In this paper, we proposed a model based on Generative Adversarial Network (GAN) and edge enhancement to perform super-resolution (SR) reconstruction for LR and blur videos, such as closed-circuit television (CCTV). The adversarial loss allows discriminators to be trained to distinguish between SR frames and ground truth (GT) frames, which is helpful to produce realistic and highly detailed results. The edge enhancement function uses the Laplacian edge module to perform edge enhancement on the intermediate result, which helps further improve the final results. In addition, we add the perceptual loss to the loss function to obtain a higher visual experience. At the same time, we also tried training network on different datasets. A large number of experiments show that our method has advantages in the Vid4 dataset and other LR videos.

scholarly journals Adversarial Erasing method based on graph neural network

2021 ◽  
Vol 2083 (4) ◽  
pp. 042083
Author(s):  
Shuhan Liu
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Loss Function ◽  
Deep Neural Networks ◽  
Ground Truth ◽  
Semantic Segmentation ◽  
Data Sets ◽  
Recent Developments ◽  
Weakly Supervised ◽  
Network Weight

Abstract Semantic segmentation is a traditional task that requires a large number of pixel-level ground truth label data sets, which is time-consuming and expensive. Recent developments in weakly-supervised settings have shown that reasonable performance can be obtained using only image-level labels. Classification is often used as an agent task to train deep neural networks and extract attention maps from them. The classification task only needs less supervision information to obtain the most discriminative part of the object. For this purpose, we propose a new end-to-end counter-wipe network. Compared with the baseline network, we propose a method to apply the graph neural network to obtain the first CAM. It is proposed to train the joint loss function to avoid the network weight sharing and cause the network to fall into a saddle point. Our experiments on the Pascal VOC2012 dataset show that 64.9% segmentation performance is obtained, which is an improvement of 2.1% compared to our baseline.

scholarly journals Frontal Cortex Segmentation of Brain PET Imaging Using Deep Neural Networks

2021 ◽  
Vol 15 ◽  
Author(s):  
Qianyi Zhan ◽  
Yuanyuan Liu ◽  
Yuan Liu ◽  
Wei Hu
Keyword(s):  
Neural Networks ◽  
Frontal Cortex ◽  
Pet Imaging ◽  
Deep Neural Networks ◽  
Ground Truth ◽  
Generative Adversarial Network ◽  
Learning Framework ◽  
Adversarial Network ◽  
Brain Pet ◽  
Positron Emission

18F-FDG positron emission tomography (PET) imaging of brain glucose use and amyloid accumulation is a research criteria for Alzheimer's disease (AD) diagnosis. Several PET studies have shown widespread metabolic deficits in the frontal cortex for AD patients. Therefore, studying frontal cortex changes is of great importance for AD research. This paper aims to segment frontal cortex from brain PET imaging using deep neural networks. The learning framework called Frontal cortex Segmentation model of brain PET imaging (FSPET) is proposed to tackle this problem. It combines the anatomical prior to frontal cortex into the segmentation model, which is based on conditional generative adversarial network and convolutional auto-encoder. The FSPET method is evaluated on a dataset of 30 brain PET imaging with ground truth annotated by a radiologist. Results that outperform other baselines demonstrate the effectiveness of the FSPET framework.

scholarly journals DAISM-DNN: Highly accurate cell type proportion estimation with in silico data augmentation and deep neural networks

2020 ◽  
Author(s):  
Yating Lin ◽  
Haojun Li ◽  
Xu Xiao ◽  
Wenxian Yang ◽  
Rongshan Yu
Keyword(s):  
Neural Networks ◽  
In Silico ◽  
Deep Neural Networks ◽  
Data Augmentation ◽  
Immune Cell ◽  
Ground Truth ◽  
Training Data ◽  
Cancer Treatments ◽  
Proportion Estimation ◽  
Real World Datasets

Understanding the immune-cell abundances of cancer and other disease-related tissues has an important role in guiding cancer treatments. We propose data augmentation through in silico mixing with deep neural networks (DAISM-DNN), where highly accurate and unbiased immune-cell proportion estimation is achieved through DNN with dataset-specific training data created from partial samples from the same batch with ground truth cell proportions. We evaluated the performance of DAISM-DNN on three publicly available real-world datasets and results showed that DAISM-DNN is robust against platform-specific variations among different datasets and outperforms other existing methods by a significant margin on all the datasets evaluated.

scholarly journals Controversial stimuli: Pitting neural networks against each other as models of human cognition

2020 ◽  
Vol 117 (47) ◽  
pp. 29330-29337 ◽  
Author(s):  
Tal Golan ◽  
Prashant C. Raju ◽  
Nikolaus Kriegeskorte
Keyword(s):  
Neural Networks ◽  
Visual Recognition ◽  
Deep Neural Networks ◽  
Human Subjects ◽  
Human Perception ◽  
Ground Truth ◽  
Natural Images ◽  
Human Cognition ◽  
Distribution Models ◽  
Adversarial Examples

Distinct scientific theories can make similar predictions. To adjudicate between theories, we must design experiments for which the theories make distinct predictions. Here we consider the problem of comparing deep neural networks as models of human visual recognition. To efficiently compare models’ ability to predict human responses, we synthesize controversial stimuli: images for which different models produce distinct responses. We applied this approach to two visual recognition tasks, handwritten digits (MNIST) and objects in small natural images (CIFAR-10). For each task, we synthesized controversial stimuli to maximize the disagreement among models which employed different architectures and recognition algorithms. Human subjects viewed hundreds of these stimuli, as well as natural examples, and judged the probability of presence of each digit/object category in each image. We quantified how accurately each model predicted the human judgments. The best-performing models were a generative analysis-by-synthesis model (based on variational autoencoders) for MNIST and a hybrid discriminative–generative joint energy model for CIFAR-10. These deep neural networks (DNNs), which model the distribution of images, performed better than purely discriminative DNNs, which learn only to map images to labels. None of the candidate models fully explained the human responses. Controversial stimuli generalize the concept of adversarial examples, obviating the need to assume a ground-truth model. Unlike natural images, controversial stimuli are not constrained to the stimulus distribution models are trained on, thus providing severe out-of-distribution tests that reveal the models’ inductive biases. Controversial stimuli therefore provide powerful probes of discrepancies between models and human perception.

scholarly journals Towards a Reliable Evaluation of Local Interpretation Methods

2021 ◽  
Vol 11 (6) ◽  
pp. 2732
Author(s):  
Jun Li ◽  
Daoyu Lin ◽  
Yang Wang ◽  
Guangluan Xu ◽  
Chibiao Ding
Keyword(s):  
Neural Networks ◽  
Deep Neural Networks ◽  
Ground Truth ◽  
Evaluation Framework ◽  
Three Dimensions ◽  
Data Sets ◽  
Annotation Tool ◽  
Visual Interpretation ◽  
Category Information ◽  
Interpretation Method

The growing use of deep neural networks in critical applications is making interpretability urgently to be solved. Local interpretation methods are the most prevalent and accepted approach for understanding and interpreting deep neural networks. How to effectively evaluate the local interpretation methods is challenging. To address this question, a unified evaluation framework is proposed, which assesses local interpretation methods from three dimensions: accuracy, persuasibility and class discriminativeness. Specifically, in order to assess correctness, we designed an interactive user feature annotation tool to provide ground truth for local interpretation methods. To verify the usefulness of the interpretation method, we iteratively display part of the interpretation results, and then ask users whether they agree with the category information. At the same time, we designed and built a set of evaluation data sets with a rich hierarchical structure. Surprisingly, one finding is that the existing visual interpretation methods cannot satisfy all evaluation dimensions at the same time, and each has its own shortcomings.

Sign in / Sign up

Export Citation Format

Share Document