Wyniki wyszukiwania - BazTech

1

Application of a deep-learning neural network for image reconstruction from a single-pixel infrared camera

Urbaś Sebastian, Więcek Bogusław

Opto-Electronics Review

|

2024

|

Vol. 32, No. 1

art. no. e148877

EN

The article presents the simulation results of a single-pixel infrared camera image reconstruction obtained by using a convolutional neural network (CNN). Simulations were carried out for infrared images with a resolution of 80 × 80 pixels, generated by a low-cost, low-resolution thermal imaging camera. The study compares the reconstruction results using the CNN and the ℓ₁ reconstruction algorithm. The results obtained using the neural network confirm a better quality of the reconstructed images with the same compression rate expressed by the peak signal-to-noise ratio (PSNR) and structural similarity index measure (SSIM).

2

Deep Learning-based Beamforming Approach Incorporating Linear Antenna Arrays

Bhalke Daulappa, Paikrao Pavan D., Anguera Jaume

Journal of Telecommunications and Information Technology

|

2024

|

nr 2

66--70

EN

This research delves into exploring machine learning and deep learning techniques relied upon in antenna design processes. First, the general concepts of machine learning and deep learning are introduced. Then, the focus shifts to various antenna applications, such as those relying on millimeter waves. The feasibility of employing antennas in this band is examined and compared with conventional methods, emphasizing the acceleration of the antenna design process, reduction in the number of simulations, and improved computational efficiency. The proposed method is a low-complexity approach which avoids the need for eigenvalue decomposition, the procedure for computing the entire matrix inversion, as well as incorporating signal and interference correlation matrices in the weight optimization process. The experimental results clearly demonstrate that the proposed method outperforms the compared beamformers by achieving a better signal-to-interference ratio.

3

Self-organized operational neural networks for the detection of atrial fibrillation

Zhang Junming, Dong Hao, Gao Jinfeng, Yao Ruxian, Li Gangqiang, Wu Haitao

Journal of Artificial Intelligence and Soft Computing Research

|

2024

|

Vol. 14, No. 1

63--75

EN

Atrial fibrillation is a common cardiac arrhythmia, and its incidence increases with age. Currently, numerous deep learning methods have been proposed for AF detection. However, these methods either have complex structures or poor robustness. Given the evidence from recent studies, it is not surprising to observe the limitations in the learning performance of these approaches. This can be attributed to their strictly homogenous conguration, which solely relies on the linear neuron model. The limitations mentioned above have been addressed by operational neural networks (ONNs). These networks employ a heterogeneous network configuration, incorporating neurons equipped with diverse nonlinear operators. Therefore, in this study, to enhance the detection performance while maintaining computational efficiency, a novel model named multi-scale Self-ONNs (MSSelf-ONNs) was proposed to identify AF. The proposed model possesses a significant advantage and superiority over conventional ONNs due to their self-organization capability. Unlike conventional ONNs, MSSelf -ONNs eliminate the need for prior operator search within the operator set library to find the optimal set of operators. This unique characteristic sets MSSelf -ONNs apart and enhances their overall performance. To validate and evaluate the system, we have implemented the experiments on the wellknown MIT-BIH atrial fibrillation database. The proposed model yields total accuracies and kappa coefficients of 98% and 0.95, respectively. The experiment results demonstrate that the proposed model outperform the state-of-the-art deep CNN in terms of both performance and computational complexity.

4

Convolutional neural networks in the SSI analysis for mine-induced vibrations

Zając Maciej, Kuźniar Krystyna

Computer Assisted Methods in Engineering and Science

|

2024

|

Vol. 31, no. 1

3--28

EN

Deep neural networks (DNNs) have recently become one of the most often used softcomputational tools for numerical analysis. The huge success of DNNs in the field of imageprocessing is associated with the use of convolutional neural networks (CNNs). CNNs,thanks to their characteristic structure, allow for the effective extraction of multi-layerfeatures. In this paper, the application of CNNs to one of the important soil-structureinteraction (SSI) problems, i.e., the analysis of vibrations transmission from the free-field next to a building to the building foundation, is presented in the case of mine-induced vibrations. To achieve this, the dataset from in-situ experimental measurements,containing 1D ground acceleration records, was converted into 2D spectrogram imagesusing either Fourier transform or continuous wavelet transform. Next, these images wereused as input for a pre-trained CNN. The output is a ratio of maximal vibration valuesrecorded simultaneously on the building foundation and on the ground. Therefore, the lastlayer of the CNN had to be changed from a classification to a regression one. The obtainedresults indicate the suitability of CNN for the analyzed problem.

5

Snoring Sound Recognition Using Multi-Channel Spectrograms

Ye Ziqiang, Peng Jianxin, Zhang Xiaowen, Song Lijuan

Archives of Acoustics

|

2024

|

Vol. 49, nr 2

169--178

EN

Obstructive sleep apnea-hypopnea syndrome (OSAHS) is a common and high-risk sleep-related breathing disorder. Snoring detection is a simple and non-invasive method. In many studies, the feature maps are obtained by applying a short-time Fourier transform (STFT) and feeding the model with single-channel input tensors. However, this approach may limit the potential of convolutional networks to learn diverse representations of snore signals. This paper proposes a snoring sound detection algorithm using a multi-channel spectrogram and convolutional neural network (CNN). The sleep recordings from 30 subjects at the hospital were collected, and four different feature maps were extracted from them as model input, including spectrogram, Mel-spectrogram, continuous wavelet transform (CWT), and multi-channel spectrogram composed of the three single-channel maps. Three methods of data set partitioning are used to evaluate the performance of feature maps. The proposed feature maps were compared through the training set and test set of independent subjects by using a CNN model. The results show that the accuracy of the multi-channel spectrogram reaches 94.18%, surpassing that of the Mel-spectrogram that exhibits the best performance among the single-channel spectrograms. This study optimizes the system in the feature extraction stage to adapt to the superior feature learning capability of the deep learning model, providing a more effective feature map for snoring detection.

6

Developing Convolutional Neural Network for Recognition of Bone Fractures in X-ray Images

Saad Aymen, Sheikh Usman Ullah, Moslim Mortada Sabri

Advances in Science and Technology. Research Journal

|

2024

|

Vol. 18, no 4

228--237

EN

In the domain of clinical imaging, the exact and quick identification proof of bone fractures assumes a crucial part in a pivotal role in facilitating timely and effective patient care. This research tends to this basic need by harnessing the force of profound learning, explicitly utilizing a Convolutional Neural Network (CNN) model as the foundation of our technique. The essential target of our study was to improve the mechanized recognition of bone fractures in X-ray images, utilizing the capacities of deep learning algorithms. The use of a CNN model permitted us to successfully capture and learn intricate patterns and features within the X-ray images, empowering the framework to make exact fracture detections. The training process included presenting the model to a various dataset, guaranteeing its versatility to an extensive variety of fracture types. The results of our research show the excellent performance of the CNN model in fracture detection, where our model has achieved an Average Precision 89.5%, Average Recall 87%, and the overall Accuracy 91%. These metrics assert the vigour of our methodology and highlight the capability of deep learning in medical image analysis.

7

Analysis of Wire Rolling Processes Using Convolutional Neural Networks

Capelin Matheus, Martinez Gustavo A.S., Xin Yutao, Siqueira Adriano F., Qian Wei-Liang

Advances in Science and Technology. Research Journal

|

2024

|

Vol. 18, no 2

103--114

EN

This study leverages machine learning to analyze the cross-sectional profiles of materials subjected to wire-rolling processes, focusing on the specific stages of these processes and the characteristics of the resulting microstructural profiles. The convolutional neural network (CNN), a potent tool for visual feature analysis and learning, is utilized to explore the properties and impacts of the cold plastic deformation technique. Specifically, CNNs are constructed and trained using 6400 image segments, each with a resolution of 120x90 pixels. The chosen architecture incorporates convolutional layers intercalated with polling layers and the “relu” activation function. The results, intriguingly, are derived from the observation of only a minuscule cropped fraction of the material’s cross-sectional profile. Following calibration and training of two distinct neural networks, we achieve training and validation accuracies of 97.4%/97% and 79%/75%, respectively. These accuracies correspond to identifying the cropped image’s location and the number of passes applied to the material. Further improvements in accuracy are reported upon integrating the two networks using a multiple-output setup, with the overall training and validation accuracies slightly increasing to 98.9%/79.4% and 94.6%/78.1%, respectively, for the two features. The study emphasizes the pivotal role of specific architectural elements, such as the rescaling parameter of the augmentation process, in attaining a satisfactory prediction rate. Lastly, we delve into the potential implications of our findings, which shed light on the potential of machine learning techniques in refining our understanding of wire-rolling processes and guiding the development of more efficient and sustainable manufacturing practices.

8

Travel time picking of ambient noise cross-correlation using a deep neural network combining convolutional neural networks and Transformer

Jin Chunwei, Ye Fang, Zhang Han, Bao Xue

Acta Geophysica

|

2024

|

Vol. 72, no. 1

97--114

EN

The travel time of ambient noise cross-correlation is widely used in geophysics, but traditional methods for picking the travel time of correlation are either difficult to be applied to data with low signal-to-noise ratio (SNR), or make some assumptions which fail to be achieved in many realistic situations, or require a lot of complex calculations. Here, we present a neural network based on convolutional neural networks (CNN) and Transformer for the travel time picking of ambient noise crosscorrelation. CNNs expand the dimension of the vector of each time step for the input of Transformer. Transformer focuses the model’s attention on the key parts of the sequence. Model derives the travel time according to the attention. 102,000 cross-correlations are used to train the network. Compared with traditional methods, the approach is easy to use and has a better performance, especially for the low SNR data. Then, we test our model on another ambient noise cross-correlation dataset, which contains cross-correlations from different regions and at different scales. The model has good performance on the test dataset. It can be seen from the experiment that the travel time of the cross-correlation function of ambient noise with an average SNR as low as 9.3 can be picked. 97.2% of the picked travel times are accurate, and the positive and negative travel time of most cross-correlations are identical (90.2%). Our method can be applied to seismic instrument performance verification, seismic velocity imaging, source location and other applications for its good ability to pick travel time accurately.

9

Detection of epileptic seizures with the use of convolutional neural networks

Wiszniewski Przemysław, Kołodziej Marcin, Majkowski Andrzej, Rysz Andrzej

Przegląd Elektrotechniczny

|

2023

|

R. 99, nr 2

51--55

EN

The purpose of the article is to investigate whether the implementation of a CNN consisting of several layers will allow the effective detection of epileptic seizures. For the research, a publicly available database registered for 4 dogs and 8 people was used. The 1-second iEEG recordings were marked by a neurophysiologist as interictal, early seizure, and seizure. A CNN was trained for each patient individually. Coefficients such as precision, AUC, sensitivity, and specificity were calculated, and the results were compared with the best algorithms published in one of the contests on the Kaggle platform. The average accuracy for the recognition of seizures using CNN is 0.921, the sensitivity is 0.850, and the specificity is 0.927. For early seizures these values are 0.825, 0.782, and 0.828, respectively.

PL

Celem artykułu było zbadanie czy zastosowanie sieci CNN, składającej się z kilku warstw umożliwi skuteczną detekcję napadów epileptycznych. Na użytek badań zastosowano ogólnodostępną bazę danych zarejestrowaną dla 4 psów oraz 8 ludzi. Jednosekundowe zapisy sygnału iEEG zostały oznaczone przez neurofizjologa jako: międzynapadowe, wczesnonapadowe oraz napadowe. Zaproponowano strukturę sieci CNN, a następnie wytrenowano ją dla każdego pacjenta indywidualnie. Zostały wyliczone współczynniki takie jak: trafność, AUC, czułość, specyficzność. Następnie wyniki zostały porównane do osiągniętych w najlepszych algorytmach opublikowanych w konkursie na platformie Kaggle. Średnia skuteczność rozpoznawania napadów z wykorzystaniem sieci CNN wynosi 0.921, czułość 0.850, a specyficzność 0.927. Dla okresów wczesnonapadowych wartości te wynoszą odpowiednio 0.825, 0.782 i 0.828.

10

Deep learning correction for image reconstruction in electrical impedance tomography using UNet model

Maciura Łukasz, Rymarczyk Tomasz, Wójcik Dariusz, Gauda Konrad, Kowalski Marcin

Przegląd Elektrotechniczny

|

2023

|

R. 99, nr 6

134--137

EN

This article was inspired by a similar Deep DBar algorithm, where a modified UNet convolutional model was used to correct the output of the DBar algorithm using the UNet model. However, instead of the DBar algorithm, another deterministic electrical impedance tomography reconstruction algorithm was used in this solution. The modified UNet model was used to successfully correct the initial reconstructions, which were computed using Kotre regularities using pseudo-inversion of the sensitivity matrix.

PL

Ten artykuł został inspirowany podobnym algorytmem Deep DBar, w którym zmodyfikowany model splotowy UNet został użyty do skorygowania danych wyjściowych algorytmu DBar przy użyciu modelu UNet. Jednak zamiast algorytmu DBar w tym rozwiązaniu zastosowano inny deterministyczny algorytm rekonstrukcji elektrycznej tomografii impedancyjnej. Zmodyfikowany model UNet został wykorzystany do skutecznej korekcji wstępnych rekonstrukcji, które zostały obliczone przy użyciu regularności Kotrego z wykorzystaniem pseudo-inwersji macierzy czułości.

11

Noise quantization simulation analysis of optical convolutional networks

Zhang Ye, Zhang Saining, Zhang Danni, Su Yanmei, Yi Junkai, Wang Pengfei, Wang Ruiting, Luo Guangzhen, Zhou Xuliang, Pan Jiaoqing

Optica Applicata

|

2023

|

Vol. 53, nr 3

483--493

EN

Optical neural network (ONN) has been regarded as one of the most prospective techniques in the future, due to its high-speed and low power cost. However, the realization of optical convolutional neural network (CNN) in non-ideal cases still remains a big challenge. In this paper, we propose an optical convolutional networks system for classification problems by applying general matrix multiply (GEMM) technology. The results show that under the influence of noise, this system still has good performance with low TOP-1 and TOP-5 error rates of 44.26% and 14.51% for ImageNet. We also propose a quantization model of CNN. The noise quantization model reaches a sufficient prediction accuracy of about 96% for MNIST handwritten dataset.

12

An intelligent compound gear-bearing fault identification approach using Bessel kernel-based time-frequency distribution

Andrews Athisayam, Manisekar Kondal

Metrology and Measurement Systems

|

2023

|

Vol. 30, nr 1

83--97

EN

The most crucial transmission components utilized in rotating machinery are gears and bearings. In a gearbox, the bearings support the force acting on the gears. Compound Faults in both the gears and bearings may cause heavy vibration and lead to early failure of components. Despite their importance, these compound faults are rarely studied since the vibration signals of the compound fault system are strongly dominated by noise. This work proposes an intelligent approach to fault identification of a compound gear-bearing system using a novel Bessel kernel-based Time-Frequency Distribution (TFD) called the Bessel transform. The Time-frequency images extracted using the Bessel transform are used as an input to the Convolutional Neural Network (CNN), which classifies the faults. The effectiveness of the proposed approach is validated with a case study, and a testing efficiency of 94% is achieved. Further, the proposed method is compared with the other TFDs and found to be effective.

13

Restoration of Remote Satellite Sensing Images using Machine and Deep Learning : a Survey

Abdellaoui Meriem, Benabdelkader Souad, Assas Ouarda

Machine Graphics and Vision

|

2023

|

Vol. 32, No. 2

147--167

EN

Remote sensing satellite images are affected by different types of degradation, which poses an obstacle for remote sensing researchers to ensure a continuous and trouble-free observation of our space. This degradation can reduce the quality of information and its effect on the reliability of remote sensing research. To overcome this phenomenon, the methods of detecting and eliminating this degradation are used, which are the subject of our study. The original aim of this paper is that it proposes a state of art of recent decade (2012-2022) on advances in remote sensing image restoration using machine and deep learning, identified by this survey, including the databases used, the different categories of degradation, as well as the corresponding methods. Machine learning and deep learning based strategies for remote sensing satellite image restoration are recommended to achieve satisfactory improvements.

14

Identifying selected diseases of leaves using deep learning and transfer learning models

Mimi Afsana, Zohura Sayeda Fatema Tuj, Ibrahim Muhammad, Haque Riddho Ridwanul, Farrok Omar, Jabid Taskeed, Ali Md Sawkat

Machine Graphics and Vision

|

2023

|

Vol. 32, No. 1

55--71

EN

Leaf diseases may harm plants in different ways, often causing reduced productivity and, at times, lethal consequences. Detecting such diseases in a timely manner can help plant owners take effective remedial measures. Deficiencies of vital elements such as nitrogen, microbial infections and other similar disorders can often have visible effects, such as the yellowing of leaves in Catharanthus roseus (bright eyes) and scorched leaves in Fragaria ×ananassa (strawberry) plants. In this work, we explore approaches to use computer vision techniques to help plant owners identify such leaf disorders in their plants automatically and conveniently. This research designs three machine learning systems, namely a vanilla CNN model, a CNN-SVM hybrid model, and a MobileNetV2-based transfer learning model that detect yellowed and scorched leaves in Catharanthus roseus and strawberry plants, respectively, using images captured by mobile phones. In our experiments, the models yield a very promising accuracy on a dataset having around 4000 images. Of the three models, the transfer learning-based one demonstrates the highest accuracy (97.35% on test set) in our experiments. Furthermore, an Android application is developed that uses this model to allow end-users to conveniently monitor the condition of their plants in real time.

15

Parallelization of Concise Convolutional Neural Networks for Plant Classification

Sembiring Arnes, Away Yuwaldi, Arnia Fitri, Muharar Rusdha

Journal of Ecological Engineering

|

2023

|

Vol. 24, nr 2

61--71

EN

Monitoring the agricultural field is the key to preventing the spread of disease and handling it quickly. The computer-based automatic monitoring system can meet the needs of large-scale and real-time monitoring. Plant classifiers that can work quickly in computer with limited resources are needed to realize this monitoring system. This study proposes convolutional neural network (CNN) architecture as a plant classifier based on leaf imagery. This architecture was built by parallelizing two concise CNN channels with different filter sizes using the addition operation. GoogleNet, SqueezeNet and MobileNetV2 were used to compare the performance of the proposed architecture. The classification performance of all these architectures was tested using the PlantVillage dataset which consists of 38 classes and 14 plant types. The experimental results indicated that the proposed architecture with a smaller number of parameters achieved nearly the same accuracy as the comparison architectures. In addition, the proposed architecture classified images 5.12 times faster than SqueezeNet, 8.23 times faster than GoogleNet, and 9.4 times faster than MobileNetV2. These findings suggest that when implemented in the agricultural field, the proposed architecture can be a reliable and faster plant classifier with fewer resources.

16

A Deep Transfer Learning Framework for the Multi-Class Classification of Vector Mosquito Species

Pise Reshma, Patil Kailas

Journal of Ecological Engineering

|

2023

|

Vol. 24, nr 9

183--191

EN

Mosquito borne diseases pose a substantial threat to public health. Vector surveillance and vector control approaches are critical to diminish the mosquito population. Quick and precise identification of mosquito species predominant in a geographic area is essential for ecological monitoring and devise effective vector control strategies in the targeted areas. There has been a growing interest in fine tuning the pretrained deep convolutional neural network models for the vision based identification of insect genera, species and gender. Transfer learning is a technique commonly applied to adapt a pre-trained model for a specific task on a different dataset especially when the new dataset has limited number of training images. In this research work, we investigate the capability of deep transfer learning to solve the multi-class classification problem of mosquito species identification. We train the pretrained deep convolutional neural networks in two transfer learning approaches: (i) Feature Extraction and (ii) Fine-tuning. Three state-of-the-art pretrained models including VGG-16, ResNet-50 and GoogLeNet were trained on a dataset of mobile captured images of three vector mosquito species: Aedes Aegypti , Anopheles Stephensi and Culex Quinquefasciatus. The results of the experiments show that GoogLeNet outperformed the other two models by achieving classification accuracy of 92.5% in feature extraction transfer learning and 96% with fine-tuning. Also, it was observed that fine-tuning the pretrained models improved the classification accuracy.

17

The concept of a mobile system for detection fire phenomena based on convolutional neural networks

Tatko Sebastian

Journal of Automation, Electronics and Electrical Engineering

|

2023

|

Vol. 5, No. 2

23--29

EN

The research problem taken up in the article is the development of an efficient, mobile and effective fire detection algorithm based on the architecture of artificial neural networks. Both the process of training and inference of CNNs is burdened with a high demand for computing power. In the case of desktop devices, equipped with powerful processors and graphics cards, this process is largely facilitated and does not cause great difficulties. Another situation, however, is the desire to create a detection algorithm that in its performance will not differ from the stationary version, nevertheless its additional feature will be mobility. The desire to supervise vast areas of critical infrastructure using an unmanned aerial vehicle, imposes peculiar hardware limitations, which mainly include weight and size. The creation of an algorithm that will carry out real-time fire detection under the above-mentioned assumptions will therefore be a task that will require the optimization of a trained neural network model, into a format supported by popular mobile systems such as the Raspberry Pi.

18

Feature map augmentation to improve scale invariance in convolutional neural networks

Kumar Dinesh, Sharma Dharmendra

Journal of Artificial Intelligence and Soft Computing Research

|

2023

|

Vol. 13, No. 1

51--74

EN

Introducing variation in the training dataset through data augmentation has been a popular technique to make Convolutional Neural Networks (CNNs) spatially invariant but leads to increased dataset volume and computation cost. Instead of data augmentation, augmentation of feature maps is proposed to introduce variations in the features extracted by a CNN. To achieve this, a rotation transformer layer called Rotation Invariance Transformer (RiT) is developed, which applies rotation transformation to augment CNN features. The RiT layer can be used to augment output features from any convolution layer within a CNN. However, its maximum effectiveness is shown when placed at the output end of final convolution layer. We test RiT in the application of scale-invariance where we attempt to classify scaled images from benchmark datasets. Our results show promising improvements in the networks ability to be scale invariant whilst keeping the model computation cost low.

19

A novel method for automatic detection of arrhythmias using the unsupervised convolutional neural network

Zhang Junming, Yao Ruxian, Gao Jinfeng, Li Gangqiang, Wu Haitao

Journal of Artificial Intelligence and Soft Computing Research

|

2023

|

Vol. 13. No. 3

181--196

EN

In recent years, various models based on convolutional neural networks (CNN) have been proposed to solve the cardiac arrhythmia detection problem and achieved saturated accuracy. However, these models are often viewed as “blackbox” and lack of interpretability, which hinders the understanding of cardiologists, and ultimately hinders the clinical use of intelligent terminals. At the same time, most of these approaches are supervised learning and require label data. It is a time-consuming and expensive process to obtain label data. Furthermore, in human visual cortex, the importance of lateral connection is same as feed-forward connection. Until now, CNN based on lateral connection have not been studied thus far. Consequently, in this paper, we combines CNNs, lateral connection and autoencoder (AE) to propose the building blocks of lateral connection convolutional autoencoder neural networks (LCAN) for cardiac arrhythmia detection, which learn representations in an unsupervised manner. Concretely, the LCAN contains a convolution layer, a lateral connection layer, an AE layer, and a pooling layer. The LCAN detects salient wave features through the lateral connection layer. The AE layer and competitive learning is used to update the filters of the convolution network—an unsupervised process that ensures similar weight distribution for all adjacent filters in each convolution layer and realizes the neurons’ semantic arrangement in the LCAN. To evaluate the performances of the proposed model, we have implemented the experiments on the well-known MIT–BIH Arrhythmia Database. The proposed model yields total accuracies and kappa coefficients of 98% and 0.95, respectively. The experiment results show that the LCAN is not only effective, but also a useful tool for arrhythmia detection.

20

FPGA Implementation of Neural Nets

Kumari B A Sujatha, Kulkarni Sudarshan Patil, Sinchana C. G.

International Journal of Electronics and Telecommunications

|

2023

|

Vol. 69, No. 3

599--604

EN

The field programmable gate array (FPGA) is used to build an artificial neural network in hardware. Architecture for a digital system is devised to execute a feed-forward multilayer neural network. ANN and CNN are very commonly used architectures. Verilog is utilized to describe the designed architecture. For the computation of certain tasks, a neural network’s distributed architecture structure makes it potentially efficient. The same features make neural nets suitable for application in VLSI technology. For the hardware of a neural network, a single neuron must be effectively implemented (NN). Reprogrammable computer systems based on FPGAs are useful for hardware implementations of neural networks.