Automated Multichannel Signal Classification Systems for Ultrasonic Nondestrucitve Evaluation

2006 ◽  
Vol 321-323 ◽  
pp. 1266-1269 ◽  
Author(s):  
J. Kim ◽  
P. Ramuhalli ◽  
L. Udpa ◽  
S. Udpa
Keyword(s):  
Neural Network ◽  
Spatial Correlation ◽  
Principal Component ◽  
Automated Analysis ◽  
Classification Systems ◽  
Inspection Systems ◽  
Weld Inspection ◽  
Correlation Information ◽  
Trained Neural Network ◽  
Analysis System

A key requirement in most ultrasonic weld inspection systems is the ability for rapid automated analysis to identify the type of flaw. Incorporation of spatial correlation information from adjacent A-scans can improve performance of the analysis system. This paper describes two neural network based classification techniques that use correlation of adjacent A-scans. The first method relies on differences in individual A-scans to classify signals using a trained neural network, with a post-processing mechanism to incorporate spatial correlation information. The second technique transforms a group of spatially localized signals using a 2-dimensional transform, and principal component analysis is applied to the transform coefficients to generate a reduced dimensional feature vectors for classification. Results of applying the proposed techniques to data obtained from weld inspection are presented, and the performances of the two approaches are compared.

Automatic Face Emotion Recognition With the Aid of Probability-Based Bird Swarm-Trained Neural Network

2021 ◽  
Vol 12 (4) ◽  
pp. 1-24
Author(s):  
Bhagyashri Devi ◽  
M. Mary Synthuja Jain Preetha
Keyword(s):  
Neural Network ◽  
Dimension Reduction ◽  
Face Detection ◽  
Orthogonal Transformation ◽  
Principal Component ◽  
Discrete Wavelet ◽  
The Face ◽  
Bird Swarm Algorithm ◽  
Trained Neural Network ◽  
Face Emotion Recognition

This paper intends to develop a novel FER model, which consists of four stages: (1) face detection, (2) feature extraction, (3) dimension reduction, and (4) classification. In this context, the face detection is done using Viola Jones method (VJ). It is the first object recognition model to offer better recognition rates in real-time. Further, features extraction techniques like local binary pattern (LBP) and discrete wavelet transform (DWT) are used for extracting the features from face detected images. Moreover, the dimension reduction of features is done using principal component analysis (PCA), which is an arithmetical process that exploits an orthogonal transformation to exchange a group of annotations of probably interrelated constraints. The classification procedure is performed using neural network (NN), with the new training algorithm called bird swarm algorithm, which is modified based on probability and hence termed as probability-based BSA (P-BSA).

Ferroelectric Memristive Networks for Dimensionality Reduction: A Process for Effectively Classifying Cancer Datasets

2019 ◽  
Vol 201 (1) ◽  
pp. 126-141
Author(s):  
P. Michael Preetam Raj ◽  
V. Jeffry Louis ◽  
Sumit Kumar Chatterjee ◽  
Sayan Kanungo ◽  
Souvik Kundu
Keyword(s):  
Breast Cancer ◽  
Neural Network ◽  
Power Dissipation ◽  
Principal Component ◽  
Ferroelectric Materials ◽  
Reduction Technique ◽  
Classification Systems ◽  
High Classification Accuracy ◽  
Neural Network Algorithm

In this work, a copper-doped (5%) zinc oxide (Cu:ZnO) ferroelectric materials-based memristor model was realized and it was employed to develop principal component analysis (PCA), a data dimension reduction technique. The developed PCA was utilized to efficaciously classify breast cancer datasets, which are considered as complex and big volumes of data. It was found that the controllable memristance variations were analogous to the weight modulations in the implemented neural network-based learning systems. Sanger’s rule was utilized to achieve unsupervised online learning in order to generate the principal components. On one side, the developed memristor-based PCA network was found to be effective to isolate distinct breast cancer classes with a high classification accuracy of 97.77% and the error in the classification of malignant cases as benign of 0.529%, a significantly low value. On the other side, the power dissipation was found to be 0.27 µW, which suggests the proposed memristive network is suitable for low-power applications. Further, a comparison was established with other existing non-memristor and non-PCA-based data classification systems. Furthermore, the devised less complex equations to implement PCA on this memristive crossbar array could be employed to implement any neural network algorithm.

scholarly journals Application of a Neural Network for the Analysis of the “Soyuz” Launch Vehicle Boosters’ Separation.

2020 ◽  
pp. 42-56
Author(s):  
M.M. Matushin ◽  
D.A. Makhalov
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Back Propagation ◽  
Automated Analysis ◽  
General Information ◽  
Network Configuration ◽  
Neural Network Learning ◽  
Network Learning ◽  
The Neural Network ◽  
Trained Neural Network

The paper discusses application of artificial intelligence (neural networks) technologies for automated analysis of dynamic processes of the “Soyuz” launch vehicle’s onboard systems. Cyclogram of strap-on boosters separa-tion as applied to this task, and telemetry measurement used to monitor this process are described. The general information about the construction of the used types of neural networks and about their learning using a back-propagation method is presented; the neural network configuration for solving the mentioned task, telemetry presentation format suitable for sup-plying power for the neural network, and features of the neural network learning are proposed. The approbation of the trained neural network for the analysis of launches of the “Soyuz-FG” and “Soyuz-2.1a” launch vehi-cles using telemetry in real-time and delayed modes was carried out.

Quality Inspection of the Riveting Process by Neural Networks

2013 ◽  
Vol 284-287 ◽  
pp. 2276-2280
Author(s):  
Huang Chi Chen ◽  
Yu Ju Chen ◽  
Hsing Ling Wang ◽  
Chuo Yean Chang ◽  
Pin Hsuin Weng ◽  
...  
Keyword(s):  
Neural Network ◽  
Inspection System ◽  
Quality Inspection ◽  
Quantum Neural Network ◽  
Riveting Process ◽  
Inspection Systems ◽  
Institute Of Technology ◽  
Trained Neural Network ◽  
The Impact ◽  
Chinese Air Force

This paper presents an automatic quality inspection system for the riveting process by using neural network (NN) techniques. Two types of neural models were used in studies. One is the conventional neural network and the other one is the quantum neural network which is expected to deal with the signals with fuzziness and uncertainty. The well-trained neural network could make an immediate diagnosis of the riveting quality based on the impact signals sensed. Thus, such NN inspection system can not only monitor the real time riveting process, but also give the assistance on the riveting quality verification. In order to demonstrate the superiority of neural network inspection system developed, the experimental data provided Chinese Air Force Institute of Technology was studied and simulated. The method of riveting quality index (RQI) was also performed as a comparison. From the simulation results shown, both of the proposed neural network inspection systems have the better verification accuracy than RQI method.

Digital Processing of STEM Images

1981 ◽  
Vol 39 ◽  
pp. 236-237
Author(s):  
A. V. Crewe ◽  
M. Ohtsuki
Keyword(s):  
High Resolution ◽  
Low Dose ◽  
Assembly Language ◽  
Processing System ◽  
Principal Component ◽  
Digital Processing ◽  
Image Analysis System ◽  
Black And White ◽  
Analysis System ◽  
Dose Imaging

We have assembled an image processing system for use with our high resolution STEM for the particular purpose of working with low dose images of biological specimens. The system is quite flexible, however, and can be used for a wide variety of images.The original images are stored on magnetic tape at the microscope using the digitized signals from the detectors. For low dose imaging, these are “first scan” exposures using an automatic montage system. One Nova minicomputer and one tape drive are dedicated to this task.The principal component of the image analysis system is a Lexidata 3400 frame store memory. This memory is arranged in a 640 x 512 x 16 bit configuration. Images are displayed simultaneously on two high resolution monitors, one color and one black and white. Interaction with the memory is obtained using a Nova 4 (32K) computer and a trackball and switch unit provided by Lexidata.The language used is BASIC and uses a variety of assembly language Calls, some provided by Lexidata, but the majority written by students (D. Kopf and N. Townes).

scholarly journals Analysis of the Nosema Cells Identification for Microscopic Images

Sensors ◽  
2021 ◽  
Vol 21 (9) ◽  
pp. 3068
Author(s):  
Soumaya Dghim ◽  
Carlos M. Travieso-González ◽  
Radim Burget
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Image Processing ◽  
Deep Learning ◽  
The Other ◽  
Support Vector ◽  
Learning Approaches ◽  
Microscopic Images ◽  
Trained Neural Network ◽  
Nosema Disease

The use of image processing tools, machine learning, and deep learning approaches has become very useful and robust in recent years. This paper introduces the detection of the Nosema disease, which is considered to be one of the most economically significant diseases today. This work shows a solution for recognizing and identifying Nosema cells between the other existing objects in the microscopic image. Two main strategies are examined. The first strategy uses image processing tools to extract the most valuable information and features from the dataset of microscopic images. Then, machine learning methods are applied, such as a neural network (ANN) and support vector machine (SVM) for detecting and classifying the Nosema disease cells. The second strategy explores deep learning and transfers learning. Several approaches were examined, including a convolutional neural network (CNN) classifier and several methods of transfer learning (AlexNet, VGG-16 and VGG-19), which were fine-tuned and applied to the object sub-images in order to identify the Nosema images from the other object images. The best accuracy was reached by the VGG-16 pre-trained neural network with 96.25%.

Sign in / Sign up

Export Citation Format

Share Document