Skocz do pozycji: 1.

Tytuł:

Noise quantization simulation analysis of optical convolutional networks

Autorzy:

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

Tematy:

optical neural network
convolutional neural network
noise
quantization

Pokaż więcej

Wydawca:

Politechnika Wrocławska. Oficyna Wydawnicza Politechniki Wrocławskiej

Powiązania:

https://bibliotekanauki.pl/articles/27310111.pdf  Link otwiera się w nowym oknie

Opis:

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.

Dostawca treści:

Biblioteka Nauki

/5

Artykuł

• na półce

Skocz do pozycji: 2.

Tytuł:

Grid Search of Convolutional Neural Network model in the case of load forecasting

Autorzy:

Tran, Thanh Ngoc

Tematy:

load forecasting
grid search
convolutional neural network

Pokaż więcej

Wydawca:

Polska Akademia Nauk. Czytelnia Czasopism PAN

Powiązania:

https://bibliotekanauki.pl/articles/1841362.pdf  Link otwiera się w nowym oknie

Opis:

The Convolutional Neural Network (CNN) model is one of the most effective models for load forecasting with hyperparameters which can be used not only to determine the CNN structure and but also to train the CNN model. This paper proposes a frame work for Grid Search hyperparameters of the CNN model. In a training process, the optimalmodels will specify conditions that satisfy requirement for minimum of accuracy scoresof Root Mean Square Error (RMSE), Mean Absolute Percentage Error (MAPE) and Mean Absolute Error (MAE). In the testing process, these optimal models will be used to evaluate the results along with all other ones. The results indicated that the optimal models have accuracy scores near the minimum values. Load demand data of Queensland (Australia) and Ho Chi Minh City (Vietnam) were utilized to verify the accuracy and reliability of the Grid Search framework.

Dostawca treści:

Biblioteka Nauki

/5

Artykuł

• na półce

Skocz do pozycji: 3.

Tytuł:

Application of visual classification algorithms for identification of underwater audio signals

Autorzy:

Gnyś, Piotr
Szczęsna, Gabriela
Domínguez-Brito, Antonio C.
Cabrera-Gámez, Jorge

Tematy:

audio processing
audio classification
convolutional neural network

Pokaż więcej

Wydawca:

Politechnika Gdańska

Powiązania:

https://bibliotekanauki.pl/articles/23956852.pdf  Link otwiera się w nowym oknie

Opis:

An audio processing and classification pipeline is presented in this work. The main focus is on the classification of sounds in a marine acoustic environment, however, the presented approach can be applied to other audio data. Audio samples from heterogeneous sources automatically spliced, normalized and transformed into spectrogram based visual representation are tagged on the pipeline input. The said representation is then used to train a convolutional neural network that can identify the presented categories in future recordings.

Dostawca treści:

Biblioteka Nauki

/5

Artykuł

• na półce

Skocz do pozycji: 4.

Tytuł:

Architecture optimization techniques for Convolutional Neural Networks: further experiments and insights

Autorzy:

Sobolewski, Artur
Szyc, Kamil

Tematy:

Model Compression
Convolutional Neural Network
computer vision

Pokaż więcej

Wydawca:

Polska Akademia Nauk. Czasopisma i Monografie PAN

Powiązania:

https://bibliotekanauki.pl/articles/58973028.pdf  Link otwiera się w nowym oknie

Opis:

In this paper, we have researched implementing convolutional neural network (CNN) models for devices with limited resources, such as smartphones and embedded computers. To optimize the number of parameters of these models, we studied various popular methods that would allow them to operate more efficiently. Specifically, our research focused on the ResNet-101 and VGG-19 architectures, which we modified using techniques specific to model optimization. We aimed to determine which approach would work best for particular requirements for a maximum accepted accuracy drop. Our contribution lies in the comprehensive ablation study, which presents the impact of different approaches on the final results, specifically in terms of reducing model parameters, FLOPS, and the potential decline in accuracy. We explored the feasibility of implementing architecture compression methods that can influence the model’s structure. Additionally, we delved into post-training methods, such as pruning and quantization, at various model sparsity levels. This study builds upon our prior research to provide a more comprehensive understanding of the subject matter at hand.

Dostawca treści:

Biblioteka Nauki

/5

Artykuł

• na półce

Skocz do pozycji: 5.

Tytuł:

Deep Learning-based Beamforming Approach Incorporating Linear Antenna Arrays

Autorzy:

Bhalke, Daulappa
Paikrao, Pavan D.
Anguera, Jaume

Tematy:

adaptive beamforming
antenna arrays
convolutional neural network

Pokaż więcej

Wydawca:

Instytut Łączności - Państwowy Instytut Badawczy

Powiązania:

https://bibliotekanauki.pl/articles/58906713.pdf  Link otwiera się w nowym oknie

Opis:

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.

Dostawca treści:

Biblioteka Nauki

/5

Artykuł

• na półce

Skocz do pozycji: 6.

Tytuł:

Competitive fitness analysis using convolutional neural network

Autorzy:

Fiok, Krzysztof
Antoł, Weronika
Prokop, Zofia
Palka, Joanna

Dostawca treści:

Repozytorium Uniwersytetu Jagiellońskiego

/5

Artykuł

• na półce

Skocz do pozycji: 7.

Tytuł:

Skin Lesion Analysis Toward Melanoma Detection Using Deep Learning Techniques

Autorzy:

Sherif, Fatma
Mohamed, Wael A.
Mohra, A.S.

Tematy:

melanoma
skin cancer
convolutional neural network
deep learning

Pokaż więcej

Wydawca:

Polska Akademia Nauk. Czytelnia Czasopism PAN

Powiązania:

https://bibliotekanauki.pl/articles/226719.pdf  Link otwiera się w nowym oknie

Opis:

In the last few years, a great attention was paid to the deep learning Techniques used for image analysis because of their ability to use machine learning techniques to transform input data into high level presentation. For the sake of accurate diagnosis, the medical field has a steadily growing interest in such technology especially in the diagnosis of melanoma. These deep learning networks work through making coarse segmentation, conventional filters and pooling layers. However, this segmentation of the skin lesions results in image of lower resolution than the original skin image. In this paper, we present deep learning based approaches to solve the problems in skin lesion analysis using a dermoscopic image containing skin tumor. The proposed models are trained and evaluated on standard benchmark datasets from the International Skin Imaging Collaboration (ISIC) 2018 Challenge. The proposed method achieves an accuracy of 96.67% for the validation set. The experimental tests carried out on a clinical dataset show that the classification performance using deep learning-based features performs better than the state-of-the-art techniques.

Dostawca treści:

Biblioteka Nauki

/5

Artykuł

• na półce

Skocz do pozycji: 8.

Tytuł:

Performance Analysis of LEACH with Deep Learning in Wireless Sensor Networks

Autorzy:

Prajapati, Hardik K.
Joshi, Rutvij

Tematy:

machine learning
Deep learning
Convolutional Neural Network (CNN)
LEACH

Pokaż więcej

Wydawca:

Polska Akademia Nauk. Czytelnia Czasopism PAN

Powiązania:

https://bibliotekanauki.pl/articles/2200710.pdf  Link otwiera się w nowym oknie

Opis:

Thousands of low-power micro sensors make up Wireless Sensor Networks, and its principal role is to detect and report specified events to a base station. Due to bounded battery power these nodes are having very limited memory and processing capacity. Since battery replacement or recharge in sensor nodes is nearly impossible, power consumption becomes one of the most important design considerations in WSN. So one of the most important requirements in WSN is to increase battery life and network life time. Seeing as data transmission and reception consume the most energy, it’s critical to develop a routing protocol that addresses the WSN’s major problem. When it comes to sending aggregated data to the sink, hierarchical routing is critical. This research concentrates on a cluster head election system that rotates the cluster head role among nodes with greater energy levels than the others.We used a combination of LEACH and deep learning to extend the network life of the WSN in this study. In this proposed method, cluster head selection has been performed by Convolutional Neural Network (CNN). The comparison has been done between the proposed solution and LEACH, which shows the proposed solution increases the network lifetime and throughput.

Dostawca treści:

Biblioteka Nauki

/5

Artykuł

• na półce

Skocz do pozycji: 9.

Tytuł:

A grinding surface roughness class recognition combining red and green information

Autorzy:

Huang, Jiefeng
Yi, Huaian
Fang, Runji
Song, Kun

Tematy:

roughness measurement
convolutional neural network
red and green information

Pokaż więcej

Wydawca:

Polska Akademia Nauk. Czasopisma i Monografie PAN

Powiązania:

https://bibliotekanauki.pl/articles/58973654.pdf  Link otwiera się w nowym oknie

Opis:

The current machine vision-based surface roughness measurement mainly relies on the design of feature indicators associated with roughness to measure the surface roughness. However, the process is tedious and complicated. Moreover, most existing deep learning methods for workpiece surface roughness measurement use a monochromatic light source to acquire images. In the case of surface roughness in a grinding process with low roughness and random texture characteristics, the feature information obtained by monochromatic light source acquisition is relatively small. It is difficult to extract the workpiece surface roughness features, which can easily cause problems for subsequent measurement. Based on the problems above, this paper proposes a grinding surface roughness measurement method combining red-green information and a convolutional neural network. The technique uses a particular red-green block to highlight the grinding surface texture features. Finally, it classifies the grinding surface roughness measurement with a classification detection technique of the convolutional neural network. Experimental results show that the accuracy of the grinding surface roughness measurement method combining red-green information and the convolutional neural network is significantly improved compared with that of the grinding surface roughness measurement method without using the red-green data.

Dostawca treści:

Biblioteka Nauki

/5

Artykuł

• na półce

Skocz do pozycji: 10.

Autorzy:

Omidi, Hamid
Sadeghi, Mohammad Hossein
Sina, Sedigheh
Farshchitabrizi, Amir Hossein
Alavi, Mehrosadat

Opis:

Ovarian cancer poses a major worldwide health issue, marked by high death rates and a deficiency in reliable diagnostic methods. The precise and prompt detection of ovarian cancer holds great importance in advancing patient outcomes and determining suitable treatment plans. Medical imaging techniques are vital in diagnosing ovarian cancer, but achieving accurate diagnoses remains challenging. Deep learning (DL), particularly convolutional neural networks (CNNs), has emerged as a promising solution to improve the accuracy of ovarian cancer detection. This systematic review explores the role of DL in improving the diagnostic accuracy for ovarian cancer. The methodology involved the establishment of research questions, inclusion and exclusion criteria, and a comprehensive search strategy across relevant databases. The selected studies focused on DL techniques applied to ovarian cancer diagnosis using medical imaging modalities, as well as tumour differentiation and radiomics. Data extraction, analysis, and synthesis were performed to summarize the characteristics and findings of the selected studies. The review emphasizes the potential of DL in enhancing the diagnosis of ovarian cancer by accelerating the diagnostic process and offering more precise and efficient solutions. DL models have demonstrated their effectiveness in categorizing ovarian tissues and achieving comparable diagnostic performance to that of experienced radiologists. The integration of DL into ovarian cancer diagnosis holds the promise of improving patient outcomes, refining treatment approaches, and supporting well-informed decision-making. Nevertheless, additional research and validation are necessary to ensure the dependability and applicability of DL models in everyday clinical settings.

Dostawca treści:

Repozytorium Uniwersytetu Jagiellońskiego

/5

Artykuł

• na półce

Skocz do pozycji: 11.

Tytuł:

Fringe pattern inpainting based on dual-exposure fused fringe guiding CNN denoiser prior

Autorzy:

Peng, Guangze
Chen, Wenjing

Tematy:

fringe projection profilometry
phase calculation
convolutional neural network
denoiser prior

Pokaż więcej

Wydawca:

Politechnika Wrocławska. Oficyna Wydawnicza Politechniki Wrocławskiej

Powiązania:

https://bibliotekanauki.pl/articles/2086757.pdf  Link otwiera się w nowym oknie

Opis:

The intensity of some pixels of the captured fringe will be saturated when fringe projection profilometry is used to measure objects with high reflectivity, which will significantly affect the reconstruction of the measured object. In this paper, we propose a fringe pattern inpainting method based on the convolutional neural network (CNN) denoiser prior guided by additional information from a fringe captured in short exposure time. First, a binary mask obtained by Otsu algorithm from the modulation information of the short exposure fringe is used to detect the high-saturation region in the normal exposure fringe. Then, the corrected gray-scales of the region of the short exposure fringe selected by the mask are inserted in the saturated region of the normal fringe to form an initial fringe for iteration. At last, fringe pattern inpainting is achieved by using a CNN denoiser prior. The correct phase can be reconstructed from the inpainted fringes. The computer simulation and experiments verify the effectiveness of the proposed method.

Dostawca treści:

Biblioteka Nauki

/5

Artykuł

• na półce

Skocz do pozycji: 12.

Tytuł:

Deep learning-based CNC milling tool wear stage estimation with multi-signal analysis

Autorzy:

Karabacak, Yunus Emre

Tematy:

wear stage estimation
milling
convolutional neural network
time-frequency analysis

Pokaż więcej

Wydawca:

Polska Akademia Nauk. Polskie Naukowo-Techniczne Towarzystwo Eksploatacyjne PAN

Powiązania:

https://bibliotekanauki.pl/articles/27312777.pdf  Link otwiera się w nowym oknie

Opis:

CNC milling machines are frequently used in the manufacturing of mechanical parts in the industry. One of the most important components of milling machines is the cutting tool. Monitoring the cutting tool wear is important for the reliability, continuity, and quality of production. Monitoring the tool and detecting the stage of wear are difficult processes. In this work, the convolutional neural network (CNN), which is a deep learning method in which the features are extracted by an inner process, was performed to detect the wear stages of the milling tool. These stages that define the total lifespan of the tool are known as initial wear (IW), steady-state wear (SSW), and accelerated wear (AW). Short Time Fourier Transform (STFT) was applied to signals, and signal spectrograms were used to train CNN models with different complex architectures. Vibration signals, acoustic emission signals, and motor current signals from The Nasa Ames Milling Dataset were used to obtain the spectrograms. Pre-trained CNNs (GoogleNet, AlexNet, ResNet-50, and EfficientNet-B0) detected the tool wear stage with varying accuracies. It has been seen that the time duration of model training increases as the size of the dataset grows and the network architecture becomes more complex. The recommended method has also been tested on the 2010 PHM Data Challenge Dataset. CNN shows promise for condition monitoring of milling operations and detecting tool wear stage.

Dostawca treści:

Biblioteka Nauki

/5

Artykuł

• na półce

Skocz do pozycji: 13.

Tytuł:

Selecting Update Blocks of Convolutional Neural Networks using Genetic Algorithm in Transfer Learning

Autorzy:

Hasan, Mehedi
Ibrahim, Muhammad
Ali, Sawkat

Tematy:

computer vision
transfer learning
convolutional neural network
EfficientNet
genetic algorithm

Pokaż więcej

Wydawca:

Szkoła Główna Gospodarstwa Wiejskiego w Warszawie. Instytut Informatyki Technicznej

Powiązania:

https://bibliotekanauki.pl/articles/59150199.pdf  Link otwiera się w nowym oknie

Opis:

The performance of convolutional neural networks (CNN) for computer vision problems depends heavily on their architectures. Transfer learning performance of a CNN strongly relies on selection of its trainable layers. Selecting the most effective update layers for a certain target dataset often requires expert knowledge on CNN architecture which many practitioners do not possess. General users prefer to use an available architecture (e.g. GoogleNet, ResNet, EfficientNet etc.) that is developed by domain experts. With the ever-growing number of layers, it is increasingly becoming difficult and cumbersome to handpick the update layers. Therefore, in this paper we explore the application of a genetic algorithm to mitigate this problem. The convolutional layers of popular pre-trained networks are often grouped into modules that constitute their building blocks. We devise a genetic algorithm to select blocks of layers for updating the parameters. By experimenting with EfficientNetB0 pre-trained on ImageNet and using three popular image datasets - namely Food-101, CIFAR-100 and MangoLeafBD - as target datasets, we show that our algorithm yields similar or better results than the baseline in terms of accuracy, and requires lower training and evaluation time due to learning a smaller number of parameters. We also devise a measure called block importance to measure each block’s efficacy as an update block and analyze the importance of the blocks selected by our algorithm.

Dostawca treści:

Biblioteka Nauki

/5

Artykuł

• na półce

Skocz do pozycji: 14.

Tytuł:

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

Autorzy:

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

Tematy:

convolutional neural network
arrhythmia detection
unsupervised learning
ECG classification

Pokaż więcej

Wydawca:

Społeczna Akademia Nauk w Łodzi. Polskie Towarzystwo Sieci Neuronowych

Powiązania:

https://bibliotekanauki.pl/articles/23944827.pdf  Link otwiera się w nowym oknie

Opis:

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.

Dostawca treści:

Biblioteka Nauki

/5

Artykuł

• na półce

Skocz do pozycji: 15.

Tytuł:

Deep networks for image super-resolution using hierarchical features

Autorzy:

Yang, Xin
Zhang, Yifan
Zhou, Dake

Tematy:

super-resolution
convolutional neural network
sub-pixel convolutional neural network
densely connected neural networks
super rozdzielczość
splotowa sieć neuronowa
subpikselowa splotowa sieć neuronowa
gęsto połączone sieci neuronowe

Pokaż więcej

Wydawca:

Polska Akademia Nauk. Czytelnia Czasopism PAN

Powiązania:

https://bibliotekanauki.pl/articles/2173634.pdf  Link otwiera się w nowym oknie

Opis:

To better extract feature maps from low-resolution (LR) images and recover high-frequency information in the high-resolution (HR) images in image super-resolution (SR), we propose in this paper a new SR algorithm based on a deep convolutional neural network (CNN). The network structure is composed of the feature extraction part and the reconstruction part. The extraction network extracts the feature maps of LR images and uses the sub-pixel convolutional neural network as the up-sampling operator. Skip connection, densely connected neural networks and feature map fusion are used to extract information from hierarchical feature maps at the end of the network, which can effectively reduce the dimension of the feature maps. In the reconstruction network, we add a 3×3 convolution layer based on the original sub-pixel convolution layer, which can allow the reconstruction network to have better nonlinear mapping ability. The experiments show that the algorithm results in a significant improvement in PSNR, SSIM, and human visual effects as compared with some state-of-the-art algorithms based on deep learning.

Dostawca treści:

Biblioteka Nauki

/5

Artykuł

• na półce

Skocz do pozycji: 16.

Tytuł:

Optimization of convolutional neural networks using the fuzzy gravitational search algorithm

Autorzy:

Poma, Yutzil
Melin, Patricia
González, Claudia I.
Martínez, Gabriela E.

Tematy:

neural networks
convolutional neural network
fuzzy gravitational search algorithm
deep learning

Pokaż więcej

Wydawca:

Sieć Badawcza Łukasiewicz - Przemysłowy Instytut Automatyki i Pomiarów

Powiązania:

https://bibliotekanauki.pl/articles/384794.pdf  Link otwiera się w nowym oknie

Opis:

This paper presents an approach to optimize a Convolutional Neural Network using the Fuzzy Gravitational Search Algorithm. The optimized parameters are the number of images per block that are used in the training phase, the number of filters and the filter size of the convolutional layer. The reason for optimizing these parameters is because they have a great impact on performance of the Convolutional Neural Networks. The neural network model presented in this work can be applied for any image recognition or classification applications; nevertheless, in this paper, the experiments are performed in the ORL and Cropped Yale databases. The results are compared with other neural networks, such as modular and monolithic neural networks. In addition, the experiments were performed manually, and the results were obtained (when the neural network is not optimized), and comparison was made with the optimized results to validate the advantage of using the Fuzzy Gravitational Search Algorithm.

Dostawca treści:

Biblioteka Nauki

/5

Artykuł

• na półce

Skocz do pozycji: 17.

Tytuł:

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

Autorzy:

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

Tematy:

convolutional neural network
operational Neural Networks
atrial fibrillation detection
ECG classification

Pokaż więcej

Wydawca:

Społeczna Akademia Nauk w Łodzi. Polskie Towarzystwo Sieci Neuronowych

Powiązania:

https://bibliotekanauki.pl/articles/59115797.pdf  Link otwiera się w nowym oknie

Opis:

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.

Dostawca treści:

Biblioteka Nauki

/5

Artykuł

• na półce

Skocz do pozycji: 18.

Tytuł:

Parallelization of Concise Convolutional Neural Networks for Plant Classification

Autorzy:

Sembiring, Arnes
Away, Yuwaldi
Arnia, Fitri
Muharar, Rusdha

Tematy:

parallelisation
concise CNN
plant classification
multi-scale CNN
convolutional neural network

Pokaż więcej

Wydawca:

Polskie Towarzystwo Inżynierii Ekologicznej

Powiązania:

https://bibliotekanauki.pl/articles/2202377.pdf  Link otwiera się w nowym oknie

Opis:

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.

Dostawca treści:

Biblioteka Nauki

/5

Artykuł

• na półce

Skocz do pozycji: 19.

Tytuł:

A two-step fall detection algorithm combining threshold-based method and convolutional neural network

Autorzy:

Xu, Tao
Se, Haifeng
Liu, Jiahui

Tematy:

wearable
fall detection
MPU6050
threshold-based method
convolutional neural network

Pokaż więcej

Wydawca:

Polska Akademia Nauk. Czytelnia Czasopism PAN

Powiązania:

https://bibliotekanauki.pl/articles/1848958.pdf  Link otwiera się w nowym oknie

Opis:

Falls are one of the leading causes of disability and premature death among the elderly. Technical solutions designed to automatically detect a fall event may mitigate fall-related health consequences by immediate medical assistance. This paper presents a wearable device called TTXFD based on MPU6050 which can collect triaxial acceleration signals. We have also designed a two-step fall detection algorithm that fuses threshold-based method (TBM) and machine learning (ML). The TTXFD exploits the TBM stage with low computational complexity to pick out and transmit suspected fall data (triaxial acceleration data). The ML stage of the two-step algorithm is implemented on a server which encodes the data into an image and exploits a fall detection algorithm based on convolutional neural network to identify a fall on the basis of the image. The experimental results show that the proposed algorithm achieves high sensitivity (97.83%), specificity (96.64%) and accuracy (97.02%) on the open dataset. In conclusion, this paper proposes a reliable solution for fall detection, which combines the advantages of threshold-based method and machine learning technology to reduce power consumption and improve classification ability.

Dostawca treści:

Biblioteka Nauki

/5

Artykuł

• na półce

Skocz do pozycji: 20.

Tytuł:

Convolutional neural networks for P300 signal detection applied to brain computer interface

Autorzy:

Riyad, Mouad
Khalil, Mohammed
Adib, Abdellah

Tematy:

deep learning
convolutional neural network
brain computer interface
P300
classification

Pokaż więcej

Wydawca:

Sieć Badawcza Łukasiewicz - Przemysłowy Instytut Automatyki i Pomiarów

Powiązania:

https://bibliotekanauki.pl/articles/2141900.pdf  Link otwiera się w nowym oknie

Opis:

A Brain‐Computer Interface (BCI) is an instrument capa‐ ble of commanding machine with brain signal. The mul‐ tiple types of signals allow designing many applications like the Oddball Paradigms with P300 signal. We propose an EEG classification system applied to BCI using the con‐ volutional neural network (ConvNet) for P300 problem. The system consists of three stages. The first stage is a Spatiotemporal convolutional layer which is a succession of temporal and spatial convolutions. The second stage contains 5 standard convolutional layers. Finally, a lo‐ gistic regression is applied to classify the input EEG sig‐ nal. The model includes Batch Normalization, Dropout, and Pooling. Also, It uses Exponential Linear Unit (ELU) function and L1‐L2 regularization to improve the lear‐ ning. For experiments, we use the database Dataset II of the BCI Competition III. As a result, we get an F1‐score of 53.26% which is higher than the BN3 model.

Dostawca treści:

Biblioteka Nauki

/5

Artykuł

• na półce

Skocz do pozycji: 21.

Tytuł:

A DHCR_ SmartNet: A smart Devanagari handwritten character recognition using level-wised CNN architecture

Autorzy:

Deore, Shalaka Prasad

Tematy:

convolutional neural network
VGG16
fine-tuned
handwritten script
Devanagari characters

Pokaż więcej

Wydawca:

Akademia Górniczo-Hutnicza im. Stanisława Staszica w Krakowie. Wydawnictwo AGH

Powiązania:

https://bibliotekanauki.pl/articles/27312907.pdf  Link otwiera się w nowym oknie

Opis:

Handwritten script recognition is a vital application of the machine-learning domain. Applications like automatic license plate detection, pin-code detection, and historical document management increases attention toward handwritten script recognition. English is the most widely spoken language in India; hence, there has been a lot of research into identifying a script using a machine. Devanagari is a popular script that is used by a large number of people on the Indian subcontinent. In this paper, a level-wised efficient transfer-learning approach is presented on the VGG16 model of a convolutional neural network (CNN) for identifying isolated Devanagari handwritten characters. In this work, a new dataset of Devanagari characters is presented and made accessible to the public. This newly created dataset is comprised of 5800 samples for 12 vowels, 36 consonants, and 10 digits. Initially, a simple CNN is implemented and trained on this new small dataset. During the next stage, a transfer-learning approach is implemented on the VGG16 model, and during the last stage, the efficient fine-tuned VGG16 model is implemented. The obtained accuracy of the fine-tuned model’s training and testing came to 98.16% and 96.47%, respectively.

Dostawca treści:

Biblioteka Nauki

/5

Artykuł

• na półce

Skocz do pozycji: 22.

Tytuł:

A distributed big data analytics model for traffic accidents classification and recognition based on SparkMlLib cores

Autorzy:

Mallahi, Imad El
Riffi, Jamal
Tairi, Hamid
Ez-Zahout, Abderrahmane
Mahraz, Mohamed Adnane

Tematy:

big data
machine learning
traffic accident
severity prediction
convolutional neural network

Pokaż więcej

Wydawca:

Sieć Badawcza Łukasiewicz - Przemysłowy Instytut Automatyki i Pomiarów

Powiązania:

https://bibliotekanauki.pl/articles/27314355.pdf  Link otwiera się w nowym oknie

Opis:

This paper focuses on the issue of big data analytics for traffic accident prediction based on SparkMllib cores; however, Spark’s Machine Learning Pipelines provide a helpful and suitable API that helps to create and tune classification and prediction models to decision-making concerning traffic accidents. Data scientists have recently focused on classification and prediction techniques for traffic accidents; data analytics techniques for feature extraction have also continued to evolve. Analysis of a huge volume of received data requires considerable processing time. Practically, the implementation of such processes in real-time systems requires a high computation speed. Processing speed plays an important role in traffic accident recognition in real-time systems. It requires the use of modern technologies and fast algorithms that increase the acceleration in extracting the feature parameters from traffic accidents. Problems with overclocking during the digital processing of traffic accidents have yet to be completely resolved. Our proposed model is based on advanced processing by the Spark MlLib core. We call on the real-time data streaming API on spark to continuously gather real-time data from multiple external data sources in the form of data streams. Secondly, the data streams are treated as unbound tables. After this, we call the random forest algorithm continuously to extract the feature parameters from a traffic accident. The use of this proposed method makes it possible to increase the speed factor on processors. Experiment results showed that the proposed method successfully extracts the accident features and achieves a seamless classification performance compared to other conventional traffic accident recognition algorithms. Finally, we share all detected accidents with details onto online applications with other users.

Dostawca treści:

Biblioteka Nauki

/5

Artykuł

• na półce

Skocz do pozycji: 23.

Tytuł:

Tomato disease detection model based on densenet and transfer learning

Autorzy:

Bakr, Mahmoud
Abdel-Gaber, Sayed
Nasr, Mona
Hazman, Maryam

Tematy:

leaf disease detection
convolutional neural network
deep learning
transfer learning

Pokaż więcej

Wydawca:

Polskie Towarzystwo Promocji Wiedzy

Powiązania:

https://bibliotekanauki.pl/articles/2097440.pdf  Link otwiera się w nowym oknie

Opis:

Plant diseases are a foremost risk to the safety of food. They have the potential to significantly reduce agricultural products quality and quantity. In agriculture sectors, it is the most prominent challenge to recognize plant diseases. In computer vision, the Convolutional Neural Network (CNN) produces good results when solving image classification tasks. For plant disease diagnosis, many deep learning architectures have been applied. This paper introduces a transfer learning based model for detecting tomato leaf diseases. This study proposes a model of DenseNet201 as a transfer learning-based model and CNN classifier. A comparison study between four deep learning models (VGG16, Inception V3, ResNet152V2 and DenseNet201) done in order to determine the best accuracy in using transfer learning in plant disease detection. The used images dataset contains 22930 photos of tomato leaves in 10 different classes, 9 disorders and one healthy class. In our experimental, the results shows that the proposed model achieves the highest training accuracy of 99.84% and validation accuracy of 99.30%.

Dostawca treści:

Biblioteka Nauki

/5

Artykuł

• na półce

Skocz do pozycji: 24.

Tytuł:

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

Autorzy:

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

Tematy:

convolutional neural network
transfer learning
leaf disease detection
image classification

Pokaż więcej

Wydawca:

Szkoła Główna Gospodarstwa Wiejskiego w Warszawie. Instytut Informatyki Technicznej

Powiązania:

https://bibliotekanauki.pl/articles/2204260.pdf  Link otwiera się w nowym oknie

Opis:

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.

Dostawca treści:

Biblioteka Nauki

/5

Artykuł

• na półce

Skocz do pozycji: 25.

Tytuł:

Scheduling using Convolutional Neural Network in GPU environment

Autorzy:

Świtalski, Piotr
Siwiak, Karolina

Tematy:

Job Shop Scheduling Problem
Convolutional Neural Network
optimization
genetic algorithm

Pokaż więcej

Wydawca:

Uniwersytet w Siedlcach

Powiązania:

https://bibliotekanauki.pl/articles/59123044.pdf  Link otwiera się w nowym oknie

Opis:

Graphics processing units (GPU) have become the foundation of artificial intelligence. Machine learning was slow, inaccurate, and inadequate for many of today’s applications. The inclusion and utilization of GPUs made a remarkable difference in large neural networks. The numerous core processors on a GPU allow machine learning engineers to train complex models using many files relatively quickly. The ability to rapidly perform multiple computations in parallel is what makes them so effective; with a powerful processor, the model can make statistical predictions about very large amounts of data. GPUs are widely used in machine learning because they offer more power and speed than CPUs. In this paper, we show the use of GPU for solving a scheduling problem. The results show that this idea is useful, especially for large optimization problems.

Dostawca treści:

Biblioteka Nauki

/5

Artykuł

• na półce

Skocz do pozycji: 26.

Tytuł:

FPGA Implementation of Neural Nets

Autorzy:

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

Tematy:

artificial neural network
Spartan-6
field programmable gate arrays (FPGAs)
convolutional neural network

Pokaż więcej

Wydawca:

Polska Akademia Nauk. Czasopisma i Monografie PAN

Powiązania:

https://bibliotekanauki.pl/articles/27311922.pdf  Link otwiera się w nowym oknie

Opis:

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.

Dostawca treści:

Biblioteka Nauki

/5

Artykuł

• na półce

Skocz do pozycji: 27.

Tytuł:

Feature map augmentation to improve scale invariance in convolutional neural networks

Autorzy:

Kumar, Dinesh
Sharma, Dharmendra

Tematy:

convolutional neural network
feature map augmentation
global features
scale-invariant
vision system

Pokaż więcej

Wydawca:

Społeczna Akademia Nauk w Łodzi. Polskie Towarzystwo Sieci Neuronowych

Powiązania:

https://bibliotekanauki.pl/articles/2201321.pdf  Link otwiera się w nowym oknie

Opis:

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.

Dostawca treści:

Biblioteka Nauki

/5

Artykuł

• na półce

Skocz do pozycji: 28.

Tytuł:

Classification of Parkinsons disease in brain MRI images using Deep Residual Convolutional Neural Network (DRCNN)

Autorzy:

Praneeth, Puppala
Sathvika, Majety
Kommareddy, Vivek
Sarath, Madala
Mallela, Saran
Vani, K. Suvarna
Chkrabarti, Prasun

Tematy:

Parkinson’s disease
Deep Residual Convolutional Neural Network
deep learning
health control

Pokaż więcej

Wydawca:

Polskie Towarzystwo Promocji Wiedzy

Powiązania:

https://bibliotekanauki.pl/articles/30148251.pdf  Link otwiera się w nowym oknie

Opis:

In our aging culture, neurodegenerative disorders like Parkinson's disease (PD) are among the most serious health issues. It is a neurological condition that has social and economic effects on individuals. It happens because the brain's dopamine-producing cells are unable to produce enough of the chemical to support the body's motor functions. The main symptoms of this illness are eyesight, excretion activity, speech, and mobility issues, followed by depression, anxiety, sleep issues, and panic attacks. The main aim of this research is to develop a workable clinical decision-making framework that aids the physician in diagnosing patients with PD influence. In this research, the authors propose a technique to classify Parkinson’s disease by MRI brain images. Initially, the input data is normalized using the min-max normalization method, and then noise is removed from the input images using a median filter. The Binary Dragonfly algorithm is then used to select features. In addition, the Dense-UNet technique is used to segment the diseased part from brain MRI images. The disease is then classified as Parkinson's disease or health control using the Deep Residual Convolutional Neural Network (DRCNN) technique along with the Enhanced Whale Optimization Algorithm (EWOA) to achieve better classification accuracy. In this work, the Parkinson's Progression Marker Initiative (PPMI) public dataset for Parkinson's MRI images is used. Indicators of accuracy, sensitivity, specificity and precision are used with manually collected data to evaluate the effectiveness of the proposed methodology.

Dostawca treści:

Biblioteka Nauki

/5

Artykuł

• na półce

Skocz do pozycji: 29.

Tytuł:

Detecting inflectional patterns for Croatian verb stems using class activation mappings

Autorzy:

Ševerdija, Domagoj
Čorić, Rebeka
Orešković, Marko
Šošić, Lucian

Tematy:

Croatian infinitive and present verb stems
convolutional neural network
class activation mapping

Pokaż więcej

Wydawca:

Polska Akademia Nauk. Instytut Podstaw Informatyki PAN

Powiązania:

https://bibliotekanauki.pl/articles/59112622.pdf  Link otwiera się w nowym oknie

Opis:

All verbal forms in the Croatian language can be derived from twobasic forms: the infinitive and the present stems. In this paper, wepresent a neural computation model that takes a verb in an infinitiveform and finds a mapping to a present form. The same model can beapplied vice-versa, i.e. map a verb from its present form to its infinitive form. Knowing the present form of a given verb, one can deduceits inflections using grammatical rules. We experiment with our modelon the Croatian language, which belongs to the Slavic group of lan-guages. The model learns a classifier through these two classification tasks and uses class activation mapping to find characters in verbs contributing to classification. The model detects patterns that follow established grammatical rules for deriving the present stem form from the infinitive stem form and vice-versa. If mappings can be found between such slots, the rest of the slots can be deduced using a rule-based system.

Dostawca treści:

Biblioteka Nauki

/5

Artykuł

• na półce

Skocz do pozycji: 30.

Tytuł:

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

Autorzy:

Urbaś, Sebastian
Więcek, Bogusław

Tematy:

single-pixel imaging
compressive sensing
thermal imaging
convolutional neural network
dataset augmentation

Pokaż więcej

Wydawca:

Polska Akademia Nauk. Stowarzyszenie Elektryków Polskich

Powiązania:

https://bibliotekanauki.pl/articles/59112910.pdf  Link otwiera się w nowym oknie

Opis:

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).

Dostawca treści:

Biblioteka Nauki

/5

Artykuł

• na półce

Skocz do pozycji: 31.

Tytuł:

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

Autorzy:

Pise, Reshma
Patil, Kailas

Tematy:

computer vision
convolutional neural network
mosquito classification
deep transfer learning
vector control

Pokaż więcej

Wydawca:

Polskie Towarzystwo Inżynierii Ekologicznej

Powiązania:

https://bibliotekanauki.pl/articles/59114124.pdf  Link otwiera się w nowym oknie

Opis:

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.

Dostawca treści:

Biblioteka Nauki

/5

Artykuł

• na półce

Skocz do pozycji: 32.

Tytuł:

Chronic liver disease detection using deep convolutional neural networks with MRI data : a deep learning approach

Autorzy:

Altınkaya, Emre
Gündoğdu, Elif
Zoroğlu Altınkaya, Elif
Emekli, Emre

Opis:

Purpose: Chronic liver disease (CLD) is a significant health issue, and detection is crucial for effective treatment. This study aimed to develop a deep learning based convolutional neural network (DeepCNN) to differentiate CLD from non-CLD patients using magnetic resonance imaging (MRI) images without segmentation, enhancing diagnostic accuracy and supporting timely intervention. Material and methods: A retrospective study was conducted using MRI data from 184 patients collected between 2018 and 2024, totaling 1112 images (460 normal, 652 CLD). Various MRI sequences, including axial T1, T2, and coronal, were used. The images were preprocessed with resizing, augmentation, and normalization techniques. The DeepCNN model was trained and compared against traditional machine learning (ML) algorithms, including logistic regression, k-nearest neighbor, support vector machines, and random forest. Model performance was evaluated using accuracy, precision, recall, F1-score, and confusion matrices. Results: The DeepCNN model achieved a 93% accuracy and an F1-score of 0.939. Precision and recall for CLD classification were 97% and 98%, respectively. In comparison, traditional ML algorithms performed with accuracies ranging from 72.31% to 83.16%, with random forest achieving the highest. The DeepCNN model significantly outperformed these methods, demonstrating its strength in medical image classification. Using axial-only images reduced accuracy to 86%, showing that coronal views contribute valuable information. Limitation of data constrained learning. Conclusions: The DeepCNN model provides superior accuracy in diagnosing CLD compared to traditional ML methods, using MRI images without segmentation. This approach offers a practical solution for improving CLD detection and paves the way for future enhancements using attention mechanisms and advanced deep learning architectures.

Dostawca treści:

Repozytorium Uniwersytetu Jagiellońskiego

/5

Artykuł

• na półce

Skocz do pozycji: 33.

Tytuł:

Transformation of PET raw data into images for event classification using convolutional neural networks

Autorzy:

Curceanu, Catalina
Chug, Neha
Kapłon, Łukasz
Niedźwiecki, Szymon
Coussat, Aurélien
Konieczka, Paweł
Czerwiński, Eryk
Skurzok, Magdalena
Kumar, Deepak
Kacprzak, Krzysztof
Korcyl, Grzegorz
Hiesmayr, Beatrix C.
Shivani, Shivani
Dulski, Kamil
Dadgar, Meysam
Baran, Jakub
Gajos, Aleksander
Raczyński, Lech
Shopa, Roman Y.
Stępień, Ewa
Fedoruk, Oleksandr
Klimaszewski, Konrad
Wiślicki, Wojciech
Parzych, Szymon
Tayefi Ardebili, Faranak
Sharma, Sushil
Krzemień, Wojciech
Moskal, Paweł
Kopka, Przemysław
Kozik, Tomasz
Perez del Rio, Elena

Opis:

In positron emission tomography (PET) studies, convolutional neural networks (CNNs) may be applied directly to the reconstructed distribution of radioactive tracers injected into the patient's body, as a pattern recognition tool. Nonetheless, unprocessed PET coincidence data exist in tabular format. This paper develops the transformation of tabular data into n-dimensional matrices, as a preparation stage for classification based on CNNs. This method explicitly introduces a nonlinear transformation at the feature engineering stage and then uses principal component analysis to create the images. We apply the proposed methodology to the classification of simulated PET coincidence events originating from NEMA IEC and anthropomorphic XCAT phantom. Comparative studies of neural network architectures, including multilayer perceptron and convolutional networks, were conducted. The developed method increased the initial number of features from 6 to 209 and gave the best precision results (79.8) for all tested neural network architectures; it also showed the smallest decrease when changing the test data to another phantom.

Dostawca treści:

Repozytorium Uniwersytetu Jagiellońskiego

/5

Artykuł

• na półce

Skocz do pozycji: 34.

Tytuł:

Plant disease detection using ensembled CNN framework

Autorzy:

Mondal, Subhash
Banerjee, Suharta
Mukherjee, Subinoy
Sengupta, Diganta

Tematy:

convolutional neural network
disease detection
ResNet-50
VGG-19
InceptionV3

Pokaż więcej

Wydawca:

Akademia Górniczo-Hutnicza im. Stanisława Staszica w Krakowie. Wydawnictwo AGH

Powiązania:

https://bibliotekanauki.pl/articles/27312905.pdf  Link otwiera się w nowym oknie

Opis:

Agriculture exhibits the prime driving force for the growth of agro-based economies globally. In agriculture, detecting and preventing crops from the attacks of pests is a primary concern in today’s world. The early detection of plant disease becomes necessary in order to avoid the degradation of the yield of crop production. In this paper, we propose an ensemble-based convolutional neural network (CNN) architecture that detects plant disease from the images of a plant’s leaves. The proposed architecture considers CNN architectures like VGG-19, ResNet-50, and InceptionV3 as its base models, and the prediction from these models is used as an input for our meta-model (Inception-ResNetV2). This approach helped us build a generalized model for disease detection with an accuracy of 97.9% under test conditions.

Dostawca treści:

Biblioteka Nauki

/5

Artykuł

• na półce

Skocz do pozycji: 35.

Tytuł:

A High-Accuracy of Transmission Line Faults (TLFs) Classification Based on Convolutional Neural Network

Autorzy:

Fuada, S.
Shiddieqy, H. A.
Adiono, T.

Tematy:

fault detection
fault classification
transmission lines
convolutional neural network
machine learning

Pokaż więcej

Wydawca:

Polska Akademia Nauk. Czytelnia Czasopism PAN

Powiązania:

https://bibliotekanauki.pl/articles/1844462.pdf  Link otwiera się w nowym oknie

Opis:

To improve power system reliability, a protection mechanism is highly needed. Early detection can be used to prevent failures in the power transmission line (TL). A classification system method is widely used to protect against false detection as well as assist the decision analysis. Each TL signal has a continuous pattern in which it can be detected and classified by the conventional methods, i.e., wavelet feature extraction and artificial neural network (ANN). However, the accuracy resulting from these mentioned models is relatively low. To overcome this issue, we propose a machine learning-based on Convolutional Neural Network (CNN) for the transmission line faults (TLFs) application. CNN is more suitable for pattern recognition compared to conventional ANN and ANN with Discrete Wavelet Transform (DWT) feature extraction. In this work, we first simulate our proposed model by using Simulink® and Matlab®. This simulation generates a fault signal dataset, which is divided into 45.738 data training and 4.752 data tests. Later, we design the number of machine learning classifiers. Each model classifier is trained by exposing it to the same dataset. The CNN design, with raw input, is determined as an optimal output model from the training process with 100% accuracy.

Dostawca treści:

Biblioteka Nauki

/5

Artykuł

• na półce

Skocz do pozycji: 36.

Tytuł:

VMD and CNN-Based Classification Model for Infrasound Signal

Autorzy:

Lu, Quanbo
Li, Mei

Tematy:

infrasound signal
variational mode decomposition
convolutional neural network
Fast Fourier Transform

Pokaż więcej

Wydawca:

Polska Akademia Nauk. Czasopisma i Monografie PAN

Powiązania:

https://bibliotekanauki.pl/articles/31339812.pdf  Link otwiera się w nowym oknie

Opis:

Infrasound signal classification is vital in geological hazard monitoring systems. The traditional classification approach extracts the features and classifies the infrasound events. However, due to the manual feature extraction, its classification performance is not satisfactory. To deal with this problem, this paper presents a classification model based on variational mode decomposition (VMD) and convolutional neural network (CNN). Firstly, the infrasound signal is processed by VMD to eliminate the noise. Then fast Fourier transform (FFT) is applied to convert the reconstructed signal into a frequency domain image. Finally, a CNN model is established to automatically extract the features and classify the infrasound signals. The experimental results show that the classification accuracy of the proposed classification model is higher than the other model by nearly 5%. Therefore, the proposed approach has excellent robustness under noisy environments and huge potential in geophysical monitoring.

Dostawca treści:

Biblioteka Nauki

/5

Artykuł

• na półce

Skocz do pozycji: 37.

Tytuł:

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

Autorzy:

Tatko, Sebastian

Tematy:

drone
fire detection
convolutional neural network
unmanned aerial vehicle
Coral AI

Pokaż więcej

Wydawca:

Uniwersytet Technologiczno-Humanistyczny im. Kazimierza Pułaskiego w Radomiu

Powiązania:

https://bibliotekanauki.pl/articles/59123659.pdf  Link otwiera się w nowym oknie

Opis:

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.

Dostawca treści:

Biblioteka Nauki

/5

Artykuł

• na półce

Skocz do pozycji: 38.

Tytuł:

Harnessing convolutional neural networks to find regions of interest on feet X-rays images

Autorzy:

Skomorowski, Marek
Skwirczyński, Maciej
Wojciechowski, Wadim

Opis:

X-rays images of the feet can vary in a number of ways, e.g. resolution of the image or spatial position of the foot on the image. This can be an obstacle when trying to build a software that, after a supervised learning, should be able to automatically search for and find elements of the X-rays images of feet allowing for diagnosis while also making it more efficient. With a small number of described examples, data coherence correlates strongly with the quality of the outcome. In this paper we are proposing a convolutional neural networks based solution, which automatically cuts and resizes feet images, so that not only the data is more consistent, but also the area, on which the artificial intelligence algorithm will work, is reduced.

Dostawca treści:

Repozytorium Uniwersytetu Jagiellońskiego

/5

Artykuł

• na półce

Skocz do pozycji: 39.

Tytuł:

Project of autonomous workstation feeding fledging birds
Projekt autonomicznego robota do karmienia podlotów

Autorzy:

Dwornicki, Dawid

Tematy:

fledging birds
convolutional neural network
computer vision
podloty
konwolucyjne sieci neuronowe
analiza obrazu

Pokaż więcej

Wydawca:

Uniwersytet Technologiczno-Humanistyczny im. Kazimierza Pułaskiego w Radomiu

Powiązania:

https://bibliotekanauki.pl/articles/2014198.pdf  Link otwiera się w nowym oknie

Opis:

The article describes project of autonomous workstation capable of feeding fledging birds. During the breeding season animal rescue centers are experiencing huge overload of patients and up to 20% of patients are birds. Despite small size they demand as much care as other animals – in case of fledging birds main need is frequent feeding which is impossible to cover by working staff. Designed workstation is meant to solve this problem and decrease mortality of sick or immature animals.
Artykuł opisuje projekt stanowiska służącego do automatycznego karmienia podlotów. W sezonie lęgowym ośrodki rehabilitacji dzikich zwierząt zmagają się ze zwiększoną liczbą pacjentów, z których nawet do 20% stanowią ptaki. Mimo małych rozmiarów wymagają tyle samo opieki co pozostałe zwierzęta – w przypadku podlotów głównym zadaniem jest regularne i częste karmienie co jest niemożliwe do zrealizowania przez ograniczony zespół. Zaprojektowany robot ma za zadanie rozwiązać ten problem, wspomóc pracowników i zmniejszyć śmiertelność młodych lub chorych ptaków.

Dostawca treści:

Biblioteka Nauki

/5

Artykuł

• na półce

Skocz do pozycji: 40.

Tytuł:

Using ADTS and CNN methods to effectively monitor CD, crosstalk, and OSNR in an optical network

Autorzy:

Mrozek, Tomasz
Perlicki, Krzysztof

Tematy:

optical performance monitoring
asynchronous delay-tap sampling
convolutional neural network
multi impairments monitoring

Pokaż więcej

Wydawca:

Polska Akademia Nauk. Stowarzyszenie Elektryków Polskich

Powiązania:

https://bibliotekanauki.pl/articles/59112940.pdf  Link otwiera się w nowym oknie

Opis:

The article presents the results of a method based on asynchronous delay-tap sampling (ADTS) and convolutional neural network (CNN) for determining simultaneously occurring disturbances described using the chromatic dispersion (CD), crosstalk and optical signal-to-noise ratio (OSNR) parameters. The ADTS method was used to generate training and test data for the convolutional network, which in turn was used to learn to recognize interference from said data. The tests were carried out for a transmission speed of 10 Gbit/s and for on-off keying (OOK) and differential phase shift keying (DPSK) modulation. Very good results were obtained in recognizing simultaneously occurring phenomena. Accuracy of over 99% was achieved for CD and crosstalk for DPSK modulation and over 98% for OOK modulation. In the case of amplified spontaneous emission (ASE) noise, slightly weaker results were obtained, above 95-96% for both modulations. Based on the conducted research, it was determined that the use of ADTS and CNN methods enables monitoring of simultaneously occurring CD, crosstalk, and ASE noise in the physical layer of the optical network, while maintaining the requirements for modern monitoring systems.

Dostawca treści:

Biblioteka Nauki

/5

Artykuł

• na półce

Skocz do pozycji: 41.

Tytuł:

Simultaneous monitoring of chromatic dispersion and optical signal to noise ratio in optical links using convolutional neural network and asynchronous delay-tap sampling

Autorzy:

Mrozek, Tomasz
Perlicki, Krzysztof
Jakubiak, Andrzej

Tematy:

deep learning
convolutional neural network
chromatic dispersion
OSNR
asynchronous delay-tap sampling

Pokaż więcej

Wydawca:

Politechnika Wrocławska. Oficyna Wydawnicza Politechniki Wrocławskiej

Powiązania:

https://bibliotekanauki.pl/articles/1835803.pdf  Link otwiera się w nowym oknie

Opis:

The article presents a method for image analysis using asynchronous delay-tap sampling (ADTS) technique and convolutional neural networks (CNNs), allowing simultaneous monitoring of many phenomena occurring in the physical layer of the optical network. The ADTS method makes it possible to visualize the course of the optical signal in the form of characteristics (so-called phase portraits), which change their shape under the influence of phenomena (including chromatic dispersion, amplified spontaneous emission noise and other). Using the VPI photonics software,a simulation model of the ADTS technique was built. After the simulation tests, 10000 images were obtained, which after proper preparation were subjected to further analysis using CNN algorithms. The main goal of the study was to train a CNN to recognize the selected impairment (distortion); then to test its accuracy and estimate the impairment for the selected set of test images. The input data consisted of processed binary images in the form of two-dimensional matrices, with the position of the pixel. This article focuses on the analysis of images containing simultaneously the phenomena of chromatic dispersion and optical signal to noise ratio.

Dostawca treści:

Biblioteka Nauki

/5

Artykuł

• na półce

Skocz do pozycji: 42.

Tytuł:

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

Autorzy:

Andrews, Athisayam
Manisekar, Kondal

Tematy:

compound gear-bearing faults
Bessel transform
time-frequency distribution
convolutional neural network

Pokaż więcej

Wydawca:

Polska Akademia Nauk. Czytelnia Czasopism PAN

Powiązania:

https://bibliotekanauki.pl/articles/2203369.pdf  Link otwiera się w nowym oknie

Opis:

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.

Dostawca treści:

Biblioteka Nauki

/5

Artykuł

• na półce

Skocz do pozycji: 43.

Tytuł:

A convolutional neural network machine learning based navigation of underwater vehicles under limited communication

Autorzy:

Sahoo, Sarada Prasanna
Pati, Bibhuti Bhusan
Das, Bikramaditya

Tematy:

Autonomous Underwater Vehicle
AUV
machine learning
hierarchical clustering
Convolutional Neural Network
CNN

Pokaż więcej

Wydawca:

Polska Akademia Nauk. Czasopisma i Monografie PAN

Powiązania:

https://bibliotekanauki.pl/articles/59111252.pdf  Link otwiera się w nowym oknie

Opis:

This paper proposes navigation of multiple autonomous underwater vehicles (AUVs) by employing machine learning approach for wide area surveys in underwater environment. Wide area survey in underwater environment is affected by low data rate. We consider two AUVs moving in formation through clustering followed by selection of optimal path that is affected by low data rate and limited acoustical underwater communication. A state compression approach using machine learning based acoustical localization and communication (ML-ALOC) is proposed to overcome the low data rate issue in which AUV states are approximated by Hierarchical clustering followed by an optimal selection approach using Convolutional Neural Network (CNN). The performance of the proposed state compression algorithm is compared with particle state compression algorithm based on K-Means clustering at each iteration followed by Akaike information criterion (AIC) pursuing extensive simulations, in which two AUVs navigate through trajectory. It is observed from the simulations that the proposed ML-ALOC system provides better estimates when compared with acoustical localization and communication (ALOC) system using particle clustering for state compression scheme.

Dostawca treści:

Biblioteka Nauki

/5

Artykuł

• na półce

Skocz do pozycji: 44.

Tytuł:

Segmentation of cancer masses on breast ultrasound images using modified U-net
Segmentacja mas nowotworowych na obrazach ultrasonografii piersi z użyciem zmodyfikowanego modelu U-net

Autorzy:

Khallassi, Ihssane
El Yousfi Alaoui, My Hachem
Jilbab, Abdelilah

Tematy:

convolutional neural network
segmentation
u-net
residual neural network
konwolucyjna sieć neuronowa
segmentacja
rezydualna sieć neuronowa

Pokaż więcej

Wydawca:

Politechnika Lubelska. Wydawnictwo Politechniki Lubelskiej

Powiązania:

https://bibliotekanauki.pl/articles/27315434.pdf  Link otwiera się w nowym oknie

Opis:

Breast cancer causes a huge number of women’s deaths every year. The accurate localization of a breast lesion is a crucial stage. The segmentation of breast ultrasound images participates in the improvement of the process of detection of breast anomalies. An automatic approach of segmentation of breast ultrasound images is presented in this paper, the proposed model is a modified u-net called Attention Residual U-net, designed to help radiologists in their clinical examination to determine adequately the limitation of breast tumors. Attention Residual U-net is a combination of existing models (Convolutional Neural Network U-net, the Attention Gate Mechanism and the Residual Neural Network). Public breast ultrasound images dataset of Baheya hospital in Egypt is used in this work. Dice coefficient, Jaccard index and Accuracy are used to evaluate the performance of the proposed model on the test set. Attention residual u-net can significantly give a dice coefficient = 90%, Jaccard index = 76% and Accuracy = 90%. The proposed model is compared with two other breast segmentation methods on the same dataset. The results show that the modified U-net model was able to achieve accurate segmentation of breast lesions in breast ultrasound images.
Każdego roku rak piersi powoduje ogromną liczbę zgonów kobiet. Dokładna lokalizacja zmiany piersi jest kluczowym etapem. Segmentacja obrazów ultrasonograficznych piersi przyczynia się do poprawy procesu wykrywania nieprawidłowości piersi. W tym artykule przedstawiono automatyczne podejście do segmentacji obrazów ultrasonograficznych piersi, proponowany model to zmodyfikowany U-net, nazwany Attention Residual U-net, zaprojektowany w celu wspomagania radiologów podczas badania klinicznego, w celu odpowiedniego określenia zasięgu guzów piersiowych. Attention Residual U-net jest połączeniem istniejących modeli (konwolucyjną siecią neuronową U-net, Attention Gate Mechanism i Residual Neural Network). W tym badaniu wykorzystano publiczny zbiór danych obrazów ultrasonograficznych piersi szpitala Baheya w Egipcie. Do oceny wydajności zaproponowanego modelu na zbiorze testowym wykorzystano współczynnik Dice'a, indeks Jaccarda i dokładność. Attention Residual U-net może znacznie przyczynić się do uzyskania współczynnika Dice'a równego 90%, indeksu Jaccarda równego 76% i dokładności równiej 90%. Proponowany model został porównany z dwoma innymi metodami segmentacji piersi na tym samym zbiorze danych. Wyniki pokazują, że zmodyfikowany model U-net był w stanie osiągnąć dokładną segmentację zmian piersiowych na obrazach ultrasonograficznych piersi.

Dostawca treści:

Biblioteka Nauki

/5

Artykuł

• na półce

Skocz do pozycji: 45.

Tytuł:

Intelligent fault diagnosis of rolling bearings based on continuous wavelet transform-multiscale feature fusion and improved channel attention mechanism

Autorzy:

Zhang, Jiqiang
Kong, Xiangwei
Cheng, Liu
Qi, Haochen
Yu, Mingzhu

Tematy:

deep learning
continuous wavelet transform
improved channel attention mechanism
multi-conditions
convolutional neural network

Pokaż więcej

Wydawca:

Polska Akademia Nauk. Polskie Naukowo-Techniczne Towarzystwo Eksploatacyjne PAN

Powiązania:

https://bibliotekanauki.pl/articles/24200817.pdf  Link otwiera się w nowym oknie

Opis:

Accurate fault diagnosis is critical to operating rotating machinery safely and efficiently. Traditional fault information description methods rely on experts to extract statistical features, which inevitably leads to the problem of information loss. As a result, this paper proposes an intelligent fault diagnosis of rolling bearings based on a continuous wavelet transform(CWT)-multiscale feature fusion and an improved channel attention mechanism. Different from traditional CNNs, CWT can convert the 1-D signals into 2-D images, and extract the wavelet power spectrum, which is conducive to model recognition. In this case, the multiscale feature fusion was implemented by the parallel 2-D convolutional neural networks to accomplish deeper feature fusion. Meanwhile, the channel attention mechanism is improved by converting from compressed to extended ways in the excitation block to better obtain the evaluation score of the channel. The proposed model has been validated using two bearing datasets, and the results show that it has excellent accuracy compared to existing methods.

Dostawca treści:

Biblioteka Nauki

/5

Artykuł

• na półce

Skocz do pozycji: 46.

Tytuł:

Denseformer for single image deraining

Autorzy:

Wang, Tianming
Wang, Kaige
Li, Qing

Tematy:

artificial intelligence
convolutional neural network
image deraining
sztuczna inteligencja
sieć neuronowa konwolucyjna
obraz pojedynczy

Pokaż więcej

Wydawca:

Uniwersytet Zielonogórski. Oficyna Wydawnicza

Powiązania:

https://bibliotekanauki.pl/articles/24987759.pdf  Link otwiera się w nowym oknie

Opis:

Image is one of the most important forms of information expression in multimedia. It is the key factor to determine the visual effect of multimedia software. As an image restoration task, image deraining can effectively restore the original information of the image, which is conducive to the downstream task. In recent years, with the development of deep learning technology, CNN and Transformer structures have shone brightly in computer vision. In this paper, we summarize the key to success of these structures in the past, and on this basis, we introduce the concept of a layer aggregation mechanism to describe how to reuse the information of the previous layer to better extract the features of the current layer. Based on this layer aggregation mechanism, we build the rain removal network called DenseformerNet. Our network strengthens feature promotion and encourages feature reuse, allowing better information and gradient flow. Through a large number of experiments, we prove that our model is efficient and effective, and expect to bring some illumination to the future rain removal network.

Dostawca treści:

Biblioteka Nauki

/5

Artykuł

• na półce

Skocz do pozycji: 47.

Tytuł:

Advanced diabetic retinopathy detection with the R-CNN: A unified visual health solution

Autorzy:

Sri Sravya, Valluri
Naga Srinivasu, Parvathaneni
Shafi, Jana
Hołubowski, Waldemar
Zielonka, Adam

Tematy:

disease diagnosis
diabetic retinopathy
convolutional neural network
diagnostyka choroby
retinopatia cukrzycowa
sieć neuronowa konwolucyjna

Pokaż więcej

Wydawca:

Uniwersytet Zielonogórski. Oficyna Wydawnicza

Powiązania:

https://bibliotekanauki.pl/articles/59123772.pdf  Link otwiera się w nowym oknie

Opis:

Diabetic retinopathy (DR) can cause blindness and vision impairment. This degenerative eye condition may lead to an irreversible vision loss. The prevalence of vision impairment and blindness caused by DR emphasizes the critical need for better screening and therapy measures. DR aetiology involves persistent hyperglycemia-induced microvascular abnormalities, oxidative stress, inflammatory reactions, and retinal blood flow changes. Common screening methods for retinal issues include fundus photography, OCT, and fluorescein angiography. For those with diabetic macular edema (DME), it is a common cause of vision loss. Our goal is to develop an automated, cost-effective method for identifying diabetic retinal disease specimens. This study introduces a faster R-CNN method for detecting and classifying DR lesions in retinal images. Those are classified across five different classes. An extensive analysis of 88,704 images from a Kaggle dataset indicates the efficiency of the proposed model, with a reasonable accuracy of 98.38%. The proposed method is robust in disease localization and classification tasks and it has outperformed other existing studies in DR recognition. On evaluating cross-datasets in Kaggle and APTOS, the model has yield better results during training and testing phases.

Dostawca treści:

Biblioteka Nauki

/5

Artykuł

• na półce

Skocz do pozycji: 48.

Tytuł:

Feasibility of the fat-suppression image-subtraction method using deep learning for abnormality detection on knee MRI

Autorzy:

Terada, Hitoshi
Nakatsuka, Tomoya
Inaoka, Tsutomu
Wada, Akihiko
Nakagawa, Koichi
Kasuya, Shusuke

Opis:

Purpose: To evaluate the feasibility of using a deep learning (DL) model to generate fat-suppression images and detect abnormalities on knee magnetic resonance imaging (MRI) through the fat-suppression image-subtraction method. Material and methods: A total of 45 knee MRI studies in patients with knee disorders and 12 knee MRI studies in healthy volunteers were enrolled. The DL model was developed using 2-dimensional convolutional neural networks for generating fat-suppression images and subtracting generated fat-suppression images without any abnormal findings from those with normal/abnormal findings and detecting/classifying abnormalities on knee MRI. The image qualities of the generated fat-suppression images and subtraction-images were assessed. The accuracy, average precision, average recall, F-measure, sensitivity, and area under the receiver operator characteristic curve (AUROC) of DL for each abnormality were calculated. Results: A total of 2472 image datasets, each consisting of one slice of original T1WI, original intermediate-weighted images, generated fat-suppression (FS)-intermediate-weighted images without any abnormal findings, generated FS-intermediate-weighted images with normal/abnormal findings, and subtraction images between the generated FS-intermediate-weighted images at the same cross-section, were created. The generated fat-suppression images were of adequate image quality. Of the 2472 subtraction-images, 2203 (89.1%) were judged to be of adequate image quality. The accuracies for overall abnormalities, anterior cruciate ligament, bone marrow, cartilage, meniscus, and others were 89.5-95.1%. The average precision, average recall, and F-measure were 73.4-90.6%, 77.5-89.4%, and 78.4-89.4%, respectively. The sensitivity was 57.4-90.5%. The AUROCs were 0.910-0.979. Conclusions: The DL model was able to generate fat-suppression images of sufficient quality to detect abnormalities on knee MRI through the fat-suppression image-subtraction method.

Dostawca treści:

Repozytorium Uniwersytetu Jagiellońskiego

/5

Artykuł

• na półce

Skocz do pozycji: 49.

Tytuł:

SpeakerNet for Cross-lingual Text-Independent Speaker Verification

Autorzy:

Habib, Hafsa
Tauseef, Huma
Fahiem, Muhammad Abuzar
Farhan, Saima
Usman, Ghousia

Tematy:

convolutional neural network
deep learning
Siamese network
speaker verification
text-independent
binary operation
Urdu speaker recognition

Pokaż więcej

Wydawca:

Polska Akademia Nauk. Czasopisma i Monografie PAN

Powiązania:

https://bibliotekanauki.pl/articles/1953543.pdf  Link otwiera się w nowym oknie

Opis:

Biometrics provide an alternative to passwords and pins for authentication. The emergence of machine learning algorithms provides an easy and economical solution to authentication problems. The phases of speaker verification protocol are training, enrollment of speakers and evaluation of unknown voice. In this paper, we addressed text independent speaker verification using Siamese convolutional network. Siamese networks are twin networks with shared weights. Feature space can be learnt easily by training these networks even if similar observations are placed in proximity. Extracted features from Siamese then can be classified using difference or correlation measures. We have implemented a customized scoring scheme that utilizes Siamese’ capability of applying distance measures with the convolutional learning. Experiments made on cross language audios of multi-lingual speakers confirm the capability of our architecture to handle gender, age and language independent speaker verification. Moreover, our designed Siamese network, SpeakerNet, provided better results than the existing speaker verification approaches by decreasing the equal error rate to 0.02.

Dostawca treści:

Biblioteka Nauki

/5

Artykuł

• na półce

Skocz do pozycji: 50.

Tytuł:

Zastosowanie metod sztucznej inteligencji do generowania zawartości w grach komputerowych
Application of artificial intelligence methods to generate content in video games.

Autorzy:

Gunia, Zuzanna

Opis:

Głównym celem pracy było badanie zastosowań metod sztucznej inteligencji w kontekście generowania zawartości w grach komputerowych. Praca dyplomowa podzielona została na dwie główne części: teoretyczną oraz projektową. Pierwsza część pracy poświęcona została zagadnieniom teoretycznym związanym zarówno ze sztuczną inteligencją jak i jej zastosowaniem w grach komputerowych, ze szczególnym uwzględnieniem zastosowań w kontekście generowania zawartości. Autorka przedstawiła w niej zarówno historię dotyczącą stosowania metod sztucznej inteligencji w kontekście gier, niezbędne definicje, obszary zastosowania oraz opisałą metody wykorzystywane w dalszej, projektowej części pracy. W drugiej części pracy opisane zostały etapy tworzenia projektu, który autorka wykonała przy użyciu silnika Unity. Projekt miał postać pojedynczego poziomu gry platformowej stworzonej w technice 2.5D. Głównym elementem projektu było wykorzystanie transferu stylu neuronowego do generowania efektów wizualnych, które miałby być jednocześnie jedną z przeszkód występujących w trakcie rozgrywki. W projekcie odnaleźć można także inne metody sztucznej inteligencji, które posłużyły do generowania zawartości. W końcowej części pracy przedstawione zostały rezultaty projektu, wnioski oraz możliwości rozwoju.
The main purpose of the thesis was research about application of artificial intelligence methods in the field of content generation in video games. The thesis was divided into two main parts: theoretical and practical. The first part of the thesis was devoted to theoretical issues related to both artificial intelligence and its application in computer games, with particular emphasis on applications in the context of content generation. The author presented the history of the use of artificial intelligence methods in the context of games, the necessary definitions, areas of application and described the methods used in the further, practical part of the work. The second part of the thesis describes the stages of creating a project that the author made using the Unity engine. The project was created as a single level of a 2.5D platform game. The main element of the project was to use Neural Style Transfer to generate visual effects that would also be one of the obstacles occurring during the game. The project includes also other artificial intelligence methods that were used to generate content. The final part of the work presents the results of the project, conclusions and development opportunities.

Dostawca treści:

Repozytorium Uniwersytetu Jagiellońskiego

/5

Inne

• na półce