检索结果-内蒙古大学图书馆

Optimizing 3D U-Net-based Brain Tumor Segmentation with Integer-arithmetic Deep Learning Accelerators

ACM JOURNAL ON EMERGING TECHNOLOGIES IN COMPUTING SYSTEMS 2022年第2期18卷 25-25页

作者： Wang, Weijia Lin, Bill Univ Calif San Diego Dept Elect & Comp Engn La Jolla CA 92093 USA

While gliomas have become the most common cancerous brain tumors, manual diagnoses from 3D MRIs are time-consuming and possibly inconsistent when conducted by different radiotherapists, which leads to the pressing demand for automatic segmentation of brain tumors. State-of-the-art approaches employ FCNs to automatically segment the MRI scans. In particular, 3D U-Net has achieved notable performance and motivated a series of subsequent works. However, their significant size and heavy computation have impeded their actual deployment. Although there exists a body of literature on the compression of CNNs using low-precision representations, they either focus on storage reduction without computational improvement or cause severe performance degradation. In this article, we propose a CNN training algorithm that approximates weights and activations using non-negative integers along with trained affine mapping functions. Moreover, our approach allows the dot-product operations to be performed in an integer-arithmetic manner and defers the floating-point decoding and encoding phases until the end of layers. Experimental results on BraTS 2018 show that our trained affine mapping approach achieves near full-precision dice accuracy with 8-bit weights and activations. In addition, we achieve a dice accuracy within 0.005 and 0.01 of the full-precision counterparts when using 4-bit and 2-bit precisions, respectively.

关键词： Semantic segmentation volumetric segmentation fully convolutional network quantization training algorithm

来源：评论

学校读者我要写书评

暂无评论

Assessment of ANN algorithms for the Concentration Prediction of Indoor Air Pollutants in Child Daycare Centers

引用

ENERGIES 2022年第7期15卷 2654页

作者： Kim, Jeeheon Hong, Yongsug Seong, Namchul Kim, Daeung Danny Gachon Univ Ecosyst Res Ctr Seongnam 13120 South Korea Daejin Univ Div Human Architectural Engn 1007 Hoguk Ro Pochon 11159 South Korea Kangwon Natl Univ Dept Architectural Engn Samcheok Si 25913 South Korea King Fahd Univ Petr & Minerals KFUPM Architectural Engn Dept Dhahran 31261 Saudi Arabia

As the time spent by people indoors continues to significantly increase, much attention has been paid to indoor air quality. While many IAQ studies have been conducted through field measurements, the use of data-driven techniques such as machine learning has been increasingly used for the prediction of indoor air pollutants. For the present study, the concentrations of indoor air pollutants such as CO2, PM2.5, and VOCs in child daycare centers were predicted by using an artificial neural network model with three different training algorithms including Levenberg-Marquardt, Bayesian regularization, and Broyden-Fletcher-Goldfarb-Shanno quasi-Newton methods. For training and validation, data of indoor pollutants measured in child daycare facilities over a 1-month period were used. The results showed all the models produced a good performance for the prediction of indoor pollutants compared with the measured data. Among the models, the prediction by the LM model met the acceptable criteria of ASHRAE guideline 14 under all conditions. It was observed that the prediction performance decreased as the number of hidden layers increased. Moreover, the prediction performance was differed by the type of indoor pollutant. This was caused by patterns observed in the measured data. Considering the outcomes of the study, better prediction results can be obtained through the selection of suitable prediction models for time series data as well as the adjustment of training algorithms.

关键词： indoor air pollutants ANN model training algorithm child daycare center

来源：评论

学校读者我要写书评

暂无评论

Enhancing anomaly detection: A comprehensive approach with MTBO feature selection and TVETBO-Optimized - Optimized Quad-LSTM classification

引用

COMPUTERS & ELECTRICAL ENGINEERING 2024年第PartA期119卷

作者： Reddy, N. V. Raja Sekhar Divya, N. Sree Jagadesh, B. Gandikota, Ramu Lella, Kranthi Kumar Pydala, Bhasha Vatambeti, Ramesh MLR Inst Technol Dept Informat Technol Hyderabad 500043 India Mahatma Gandhi Inst Technol Dept Informat Technol Hyderabad India VIT AP Univ Sch Comp Sci & Engn Vijayawada India Narsimha Reddy Engn Coll Dept Comp Sci & Engn Hyderabad 500075 India Mohan Babu Univ Dept Data Sci Tirupati India

The rapid growth of computer networks has heightened system vulnerabilities, challenging traditional Machine Learning (ML) techniques in network anomaly detection due to the abundance of data, imbalanced attack classes, and large-scale datasets. This paper introduces a novel approach leveraging the AWID3 dataset, utilizing advanced preprocessing techniques such as data splitting and normalization. We propose the Mountaineering Team-Based Optimization (MTBO) algorithm for robust feature selection, inspired by social behavior and human collaboration. Additionally, we deploy a Quad-LSTM framework, integrating ConvLSTM, CNN-LSTM, E3DLSTM, and SAConvLSTM models, fine-tuned by the Technical and Vocational Education and training-Based Optimizer (TVETBO). Our approach achieves a remarkable 99.91 % accuracy, significantly surpassing existing models, thus enhancing network security by effectively addressing the complexities of anomaly detection.

关键词： Anomaly detection Network security Mountaineering team-based optimization Long short term memory Technical and vocational education and training algorithm

来源：评论

学校读者我要写书评

暂无评论

Jednoduché rozpoznávání písma

Jednoduché rozpoznávání písma

引用

作者： Duba, Nikolas Brno University of Technology

Tato práce je zaměřená na rozpoznávání textu a jeho následné zpracování. Cílem je umožnit jednoduché sledování denních výdavků. Může to slouž... 详细信息

Tato práce je zaměřená na rozpoznávání textu a jeho následné zpracování. Cílem je umožnit jednoduché sledování denních výdavků. Může to sloužit pro jednotlivce nebo pro firmu jako kontrola spotřeby. Princip je založen na uživatelsky jednoduchém ovládání, jak je to len možné. Aplikace ze vstupního hardvéru jako je například skener anebo digitální fotoaparát načítá pokladniční blok, který potom následně upraví pro další zpracování. Pomocí různých metod pro rozpoznávání písma analyzuje doklad a rozparsuje obsah. Metody jsou detailně rozebírány v práci. Výstup aplikace je databáze, ve které jsou uloženy zpracované pokladniční bloky. Součást řešení je taky informační systém, který slouží jako výstup pro zobrazení vyhodnocených výsledků.

关键词： Rozpoznávání písma práhování histogram neurální sítě parsování zpracování obrazu Hough trénování pokladniční blok perceptron OCR character recognition optical scanning image processing Hough neuron network training algorithm histogram

来源：评论

学校读者我要写书评

暂无评论

Enhanced Heart Rate Prediction Model Using Damped Least-Squares algorithm

引用

SENSORS 2022年第24期22卷 9679-9679页

作者： An, Angela Al-Fawa'reh, Mohammad Kang, James Jin Sch Informat Technol Deakin Univ Burwood Vic 3125 Australia Edith Cowan Univ Sch Sci Comp & Secur Joondalup WA 6027 Australia Natl Taiwan Univ Int Coll Taipei 10617 Taiwan

Monitoring a patient's vital signs is considered one of the most challenging problems in telehealth systems, especially when patients reside in remote locations. Companies now use IoT devices such as wearable devices to participate in telehealth systems. However, the steady adoption of wearables can result in a significant increase in the volume of data being collected and transmitted. As these devices run on limited battery power, they can run out of power quickly due to the high processing requirements of the device for data collection and transmission. Given the importance of medical data, it is imperative that all transmitted data adhere to strict integrity and availability requirements. Reducing the volume of healthcare data and the frequency of transmission can improve a device's battery life via an inference algorithm. Furthermore, this approach creates issues for improving transmission metrics related to accuracy and efficiency, which are traded-off against each other, with increasing accuracy reducing efficiency. This paper demonstrates that machine learning (ML) can be used to overcome the trade-off problem. The damped least-squares algorithm (DLSA) is used to enhance both metrics by taking fewer samples for transmission whilst maintaining accuracy. The algorithm is tested with a standard heart rate dataset to compare the metrics. The results showed that the DLSA provides the best performance, with an efficiency of 3.33 times for reduced sample data size and an accuracy of 95.6%, with similar accuracies observed in seven different sampling cases adopted for testing that demonstrate improved efficiency. This proposed method significantly improve both metrics using ML without sacrificing one metric over the other compared to existing methods with high efficiency.

关键词： inference algorithm data accuracy data efficiency healthcare damped least-squares algorithm (DLSA) machine learning neural networks training algorithm

来源：评论

学校读者我要写书评

暂无评论

Food Cooking Process Modeling With Neural Networks

引用

IEEE ACCESS 2024年 12卷 175866-175881页

作者： Fananas-Anaya, Javier Lopez-Nicolas, Gonzalo Sagues, Carlos Llorente, Sergio Univ Zaragoza Inst Invest Ingn Aragon I3A Zaragoza 50018 Spain BSH Home Appliances Grp Prod Div Cookers Res & Dev Dept Induct Technol Zaragoza 50016 Spain

Food cooking process are complex dynamical systems to model. In the state of the art we find that a good solution consists of physics-based finite element models (FEM). FEM models, although being very accurate, have a high computational cost making them unfeasible for real-time applications. To solve this problem, we consider neural networks (NN) trained from FEM simulations. Specifically, we propose a Nonlinear AutoRegressive with eXogenous inputs Neural Network (NARX-NN). The main novelty is that we define a novel training algorithm adapted to the modeling of real-time dynamical systems, allowing a NARX-NN with a simple structure to obtain a negligible error compared to the results of the original FEM model. The NARX-NN trained with the proposed training algorithm obtains an R-squared of more than 0.99 in the test simulations, while the same NARX-NN trained with the standard training algorithm obtains an R-squared of 0.78 in the same tests. The proposed NARX-NN achieves a speedup of 8 orders of magnitude compared to the original FEM model. Moreover, the developed NN is able to predict the complete cooking of the food in a few milliseconds without the need of external sensors. Alternatively, our approach can also be used in real time with information captured with sensors. The presented methodology is highly scalable and could be adapted to different types of food and cooking processes, as well as to other dynamical systems in general.

关键词： Food modeling neural network real-time sensors simulation training algorithm Food modeling neural network real-time sensors simulation training algorithm

来源：评论

学校读者我要写书评

暂无评论

Design and Implementation of a Spiking Neural Network with Integrate-and-Fire Neuron Model for Pattern Recognition

引用

INTERNATIONAL JOURNAL OF NEURAL SYSTEMS 2021年第3期31卷 2050073页

作者： Rashvand, Parvaneh Ahmadzadeh, Mohammad Reza Shayegh, Farzaneh Isfahan Univ Technol Dept Elect & Comp Engn Digital Signal Proc Res Lab Esfahan *** Iran

In contrast to the previous artificial neural networks (ANNs), spiking neural networks (SNNs) work based on temporal coding approaches. In the proposed SNN, the number of neurons, neuron models, encoding method, and learning algorithm design are described in a correct and pellucid fashion. It is also discussed that optimizing the SNN parameters based on physiology, and maximizing the information they pass leads to a more robust network. In this paper, inspired by the "center-surround" structure of the receptive fields in the retina, and the amount of overlap that they have, a robust SNN is implemented. It is based on the Integrate-and-Fire (IF) neuron model and uses the time-to-first-spike coding to train the network by a newly proposed method. The Iris and MNIST datasets were employed to evaluate the performance of the proposed network whose accuracy, with 60 input neurons, was 96.33% on the Iris dataset. The network was trained in only 45 iterations indicating its reasonable convergence rate. For the MNIST dataset, when the gray level of each pixel was considered as input to the network, 600 input neurons were required, and the accuracy of the network was 90.5%. Next, 14 structural features were used as input. Therefore, the number of input neurons decreased to 210, and accuracy increased up to 95%, meaning that an SNN with fewer input neurons and good skill was implemented. Also, the ABIDE1 dataset is applied to the proposed SNN. Of the 184 data, 79 are used for healthy people and 105 for people with autism. One of the characteristics that can differentiate between these two classes is the entropy of the existing data. Therefore, Shannon entropy is used for feature extraction. Applying these values to the proposed SNN, an accuracy of 84.42% was achieved by only 120 iterations, which is a good result compared to the recent results.

关键词： Spiking neural networks training algorithm neural coding image recognition IF neuron model

来源：评论

学校读者我要写书评

暂无评论

Artificial neural network modeling for steam ejector design

引用

APPLIED THERMAL ENGINEERING 2022年 204卷 1页

作者： Zhang, Kun Zhang, Zhen Han, Yuning Gu, Yinggang Qiu, Qinggang Zhu, Xiaojing Dalian Ocean Univ Sch Ocean & Civil Engn Dalian 116023 Peoples R China Nucl Power Inst China Chengdu 610213 Peoples R China Dalian Univ Technol Minist Educ Key Lab Ocean Energy Utilizat & Energy Conservat Dalian 116024 Peoples R China

An artificial neural network (ANN) model for a steam-centered ejector was established and the effect of different training algorithms on the prediction effectiveness of the ANN model was discussed, which found that the ANN model produces better results than the conventional thermodynamic model on the fitting and prediction of experimental data. The Levenberg-Marquardt(LM) trained model yielded the best results among three chosen ANN models, with the experimental accordance improvement of 68% and the prediction error within 15% under given operating conditions. The LM model made the prediction for a steam ejector in a certain system that the outlet area ratio exhibits a smaller effect on the system operation, compared with the entrainment ratio and throat area ratio, which assists to optimize system design and maintain operation stability.

关键词： Steam ejector Artificial neural network training algorithm System stability

来源：评论

学校读者我要写书评

暂无评论

Research on road extraction of remote sensing image based on convolutional neural network

引用

EURASIP JOURNAL ON IMAGE AND VIDEO PROCESSING 2019年第1期2019卷 1页

作者： Jiang, Yuantao Shanghai Maritime Univ Sch Econ & Management Shanghai Peoples R China

Road is an important kind of basic geographic information. Road information extraction plays an important role in traffic management, urban planning, automatic vehicle navigation, and emergency management. With the development of remote sensing technology, the quality of high-resolution satellite images is improved and more easily obtained, which makes it possible to use remote sensing images to locate roads accurately. Therefore, it is an urgent problem to extract road information from remote sensing images. To solve this problem, a road extraction method based on convolutional neural network is proposed in this paper. Firstly, convolutional neural network is used to classify the high-resolution remote sensing images into two classes, which can distinguish the road from the non-road and extract the road information initially. Secondly, the convolutional neural network is optimized and improved from the training algorithm. Finally, because of the influence of natural scene factors such as house and tree shadow, the non-road noise still exists in the road results extracted by the optimized convolutional neural network method. Therefore, this paper uses wavelet packet method to filter these non-road noises, so as to accurately present the road information in remote sensing images. The simulation results show that the road information of remote sensing image can be preliminarily distinguished by convolutional neural network;the road information can be distinguished effectively by optimizing convolutional neural network;and the wavelet packet method can effectively remove noise interference. Therefore, the proposed road extraction method based on convolutional neural network has good road information extraction effect.

关键词： Road extraction Remote sensing image Convolutional neural network training algorithm Wavelet packet method

来源：评论

学校读者我要写书评

暂无评论

Nonlinear size-dependent aerodynamics of axially reinforced doubly curved micropanel with GPLs: Application of innovative artificial neural network model

引用

MECHANICS OF ADVANCED MATERIALS AND STRUCTURES 2024年第26期31卷 8136-8160页

作者： Han, Shaoyong Ye, Qianqian Almadhor, Ahmad Abbas, Mohamed Changzhou Vocat Inst Ind Technol Sch Informat Engn & Technol Changzhou Jiangsu Peoples R China Bank Zhengzhou Postdoctoral Sci Res Workstn Zhengzhou Henan Peoples R China Wenzhou Univ Technol Sch Data Sci & Artificial Intelligence Wenzhou Zhejiang Peoples R China Jouf Univ Coll Comp & Informat Sci Dept Comp Engn & Networks Sakaka Saudi Arabia King Khalid Univ Coll Engn Elect Engn Dept Abha Saudi Arabia

Recent years have seen significant research on oscillations in aircraft structures. A self-excited oscillation known as flutter is known to wear down aircraft equipment. So, in this work, the graphene nanoplatelets reinforced composite (GPLRC) micropanel's nonlinear aerodynamics are modeled in this paper, and the nonlinear aerodynamics properties are validated using an artificial neural network technique. The axial direction and the material characteristics of the current system are graded. The relationship between the input layers of composite weight fraction, mode number, and slenderness ratio and the output layer of nonlinear vibrations derived from mathematical simulation has been estimated using an artificial neural network model. As a training algorithm, Levenberg-Marquardt back-propagation is employed. This study is new in that it attempts for the first time to estimate the nonlinear vibration properties of micropanels built of axially graded GPLRC material using the aforementioned input layer. Without the need to solve any differential equations or go through time-consuming experimental procedures, the suggested artificial neural network model can forecast linear or nonlinear natural frequencies. The findings demonstrate that axially graded GPLRC micropanel linear/nonlinear vibration problems can be effectively addressed using artificial intelligence techniques. Future vibration analyses of the microstructures for engineering reasons can use the provided mode.

关键词： Nonlinear vibrations training algorithm aerodynamics Levenberg-Marquardt back-propagation method mathematical simulation

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：