scholarly journals Defect segmentation for multi-illumination quality control systems

2021 ◽  
Vol 32 (6) ◽  
Author(s):  
David Honzátko ◽  
Engin Türetken ◽  
Siavash A. Bigdeli ◽  
L. Andrea Dunbar ◽  
Pascal Fua
Keyword(s):  
Control Systems ◽  
Data Augmentation ◽  
Surface Inspection ◽  
Production Lines ◽  
Approach Training ◽  
Learning Techniques ◽  
Deep Networks ◽  
Automated Process ◽  
Image Collection ◽  
Small Image

AbstractThanks to recent advancements in image processing and deep learning techniques, visual surface inspection in production lines has become an automated process as long as all the defects are visible in a single or a few images. However, it is often necessary to inspect parts under many different illumination conditions to capture all the defects. Training deep networks to perform this task requires large quantities of annotated data, which are rarely available and cumbersome to obtain. To alleviate this problem, we devised an original augmentation approach that, given a small image collection, generates rotated versions of the images while preserving illumination effects, something that random rotations cannot do. We introduce three real multi-illumination datasets, on which we demonstrate the effectiveness of our illumination preserving rotation approach. Training deep neural architectures with our approach delivers a performance increase of up to 51% in terms of AuPRC score over using standard rotations to perform data augmentation.

DETERMINATION OF THE JUICE EVAPORATION MATHEMATICAL MODELS EFFICIENCY FOR AUTOMATED PROCESS CONTROL SYSTEMS

2019 ◽  
Vol 9 (1) ◽  
Author(s):  
�. ������� ◽  
◽  
�. ʳ�� ◽  
�. ������� ◽  
�. ������� ◽  
...  
Keyword(s):  
Process Control ◽  
Mathematical Models ◽  
Control Systems ◽  
Process Control Systems ◽  
Automated Process

scholarly journals Exploring the octanol–water partition coefficient dataset using deep learning techniques and data augmentation

2021 ◽  
Vol 4 (1) ◽  
Author(s):  
Nadin Ulrich ◽  
Kai-Uwe Goss ◽  
Andrea Ebert
Keyword(s):  
Partition Coefficient ◽  
Data Augmentation ◽  
Environmental Chemistry ◽  
Chemical Properties ◽  
Predictive Performance ◽  
Chemical Structures ◽  
Tautomeric Forms ◽  
Learning Techniques ◽  
Water Partition Coefficient ◽  
Using Data

AbstractToday more and more data are freely available. Based on these big datasets deep neural networks (DNNs) rapidly gain relevance in computational chemistry. Here, we explore the potential of DNNs to predict chemical properties from chemical structures. We have selected the octanol-water partition coefficient (log P) as an example, which plays an essential role in environmental chemistry and toxicology but also in chemical analysis. The predictive performance of the developed DNN is good with an rmse of 0.47 log units in the test dataset and an rmse of 0.33 for an external dataset from the SAMPL6 challenge. To this end, we trained the DNN using data augmentation considering all potential tautomeric forms of the chemicals. We further demonstrate how DNN models can help in the curation of the log P dataset by identifying potential errors, and address limitations of the dataset itself.

Adaptive Kanban control systems for two-stage production lines

2010 ◽  
Vol 20 (1/2/3/4) ◽  
pp. 75 ◽  
Author(s):  
Katsuhiko Takahashi ◽  
Katsumi Morikawa ◽  
Daisuke Hirotani ◽  
Takeshi Yoshikawa
Keyword(s):  
Control Systems ◽  
Production Lines ◽  
Two Stage ◽  
Kanban Control

scholarly journals Deep Learning for Laryngopharyngeal Reflux Diagnosis

2021 ◽  
Vol 11 (11) ◽  
pp. 4753
Author(s):  
Gen Ye ◽  
Chen Du ◽  
Tong Lin ◽  
Yan Yan ◽  
Jack Jiang
Keyword(s):  
Deep Learning ◽  
Speech Processing ◽  
Data Augmentation ◽  
Laryngopharyngeal Reflux ◽  
Ph Monitoring ◽  
Binary Classification ◽  
Classification Problem ◽  
Learning Approaches ◽  
Learning Techniques ◽  
Auc Value

(1) Background: Deep learning has become ubiquitous due to its impressive performance in various domains, such as varied as computer vision, natural language and speech processing, and game-playing. In this work, we investigated the performance of recent deep learning approaches on the laryngopharyngeal reflux (LPR) diagnosis task. (2) Methods: Our dataset is composed of 114 subjects with 37 pH-positive cases and 77 control cases. In contrast to prior work based on either reflux finding score (RFS) or pH monitoring, we directly take laryngoscope images as inputs to neural networks, as laryngoscopy is the most common and simple diagnostic method. The diagnosis task is formulated as a binary classification problem. We first tested a powerful backbone network that incorporates residual modules, attention mechanism and data augmentation. Furthermore, recent methods in transfer learning and few-shot learning were investigated. (3) Results: On our dataset, the performance is the best test classification accuracy is 73.4%, while the best AUC value is 76.2%. (4) Conclusions: This study demonstrates that deep learning techniques can be applied to classify LPR images automatically. Although the number of pH-positive images used for training is limited, deep network can still be capable of learning discriminant features with the advantage of technique.

APPLICATION OF INTERNET ARCHITECTURE IN CENTRALIZED AUTOMATED PROCESS CONTROL SYSTEMS

2021 ◽  
pp. 3-6
Author(s):  
D. Vasilchenko ◽  
A. Budilovskaya
Keyword(s):  
Process Control ◽  
Control Systems ◽  
Centralized Control ◽  
Distributed Objects ◽  
Internet Architecture ◽  
Process Control Systems ◽  
Scada Systems ◽  
Geographically Distributed ◽  
Automated Process ◽  
Use Of Internet

This article discusses the use of Internet architecture in centralized automated process control systems for the purpose of monitoring and managing geographically distributed objects. The hardware components of the proposed architecture are described and the required functions are formulated. The methods of implementing these functions of centralized control systems based on this architecture are proposed: using internal algorithms of SCADA systems, or using microprocessor subsystems. The difficulties that are likely to be encountered when implementing all the required functions in the system being developed are described.

A Method of Environmental Sound Classification Based on Residual Networks and Data Augmentation

2021 ◽  
pp. 2150018
Author(s):  
Jinfang Zeng ◽  
Youming Li ◽  
Yu Zhang ◽  
Da Chen
Keyword(s):  
Neural Networks ◽  
Data Augmentation ◽  
Machine Learning Techniques ◽  
Deep Convolutional Neural Networks ◽  
Mel Frequency Cepstral Coefficients ◽  
Environmental Sound ◽  
Sound Classification ◽  
Learning Techniques ◽  
Proposed Model ◽  
Audio Data

Environmental sound classification (ESC) is a challenging problem due to the complexity of sounds. To date, a variety of signal processing and machine learning techniques have been applied to ESC task, including matrix factorization, dictionary learning, wavelet filterbanks and deep neural networks. It is observed that features extracted from deeper networks tend to achieve higher performance than those extracted from shallow networks. However, in ESC task, only the deep convolutional neural networks (CNNs) which contain several layers are used and the residual networks are ignored, which lead to degradation in the performance. Meanwhile, a possible explanation for the limited exploration of CNNs and the difficulty to improve on simpler models is the relative scarcity of labeled data for ESC. In this paper, a residual network called EnvResNet for the ESC task is proposed. In addition, we propose to use audio data augmentation to overcome the problem of data scarcity. The experiments will be performed on the ESC-50 database. Combined with data augmentation, the proposed model outperforms baseline implementations relying on mel-frequency cepstral coefficients and achieves results comparable to other state-of-the-art approaches in terms of classification accuracy.

Reducing the Impact of Source Information Uncertainty in Intelligent Automated Thermal Process Control Systems in Ensuring Reliable Operation of Power Equipment

Vestnik MEI ◽  
2021 ◽  
Vol 3 (3) ◽  
pp. 78-87
Author(s):  
Edik K. Arakelyan ◽  
◽  
Ivan A. Shcherbatov ◽  
Keyword(s):  
Control System ◽  
Process Control ◽  
Control Systems ◽  
Process Control System ◽  
Power Equipment ◽  
Source Information ◽  
Equipment Operation ◽  
Process Control Systems ◽  
Automated Process ◽  
The Impact

The uncertainty of the source information is used to solve key tasks in an intelligent automated thermal process control system affects the calculation of control actions, the implementation of equipment optimal operating modes and, as a result, leads to degraded reliability. As a rule, this type of information can be qualitative (the use of expert knowledge) or quantitative in nature. In this regard, it is extremely important to reduce the impact of uncertainty. The aim of the study is to identify the types and origins of uncertainty in the source information used by an intelligent automated process control system and to develop approaches to reduce its impact on the reliability of power equipment operation. The approaches used to ensure the specified indicators of reliability, efficiency and environmental friendliness in modern intelligent automated process control systems are based on predictive strategies, according to which the technical condition of equipment and specific degradation processes are predicted. This means that various types of uncertainty can have a significant negative impact. To reduce the influence of uncertainty of the initial information that affects the reliability of power equipment operation, the use of artificial neural networks is proposed. Their application opens the possibility to predict the occurrence of equipment defects and failures based on retrospective data for specified forecast time intervals. A method for reducing the impact of anomalies contained in the source information used in an intelligent process control system for energy facilities is demonstrated. Data omissions and outliers are investigated, the elimination of which reduces the impact of uncertainty and improves the quality of solving key problems in intelligent automated process control systems. Experimental studies were carried out that made it possible to identify the mathematical methods for removing omissions and anomalies in the source information in the best way. Methodological aspects of eliminating various types of uncertainty that exist in managing of power facilities by means of intelligent automated process control systems at the key stages of the power equipment life cycle are described.

Sign in / Sign up

Export Citation Format

Share Document