检索结果-内蒙古大学图书馆

Noise robust automatic speech recognition: review and analysis

International Journal of Speech Technology 2023年第2期26卷 475-519页

作者： Dua, Mohit Akanksha Dua, Shelza Department of Computer Engineering National Institute of Technology Kurukshetra India Department of Electronics and Communication Engineering National Institute of Technology Kurukshetra India

Automatic Speech Recognition (ASR) system is an emerging technology used in various fields such as robotics, traffic controls, and healthcare, etc. The leading cause of ASR performance degradation is mismatch between the training and testing environments. The main reason for this mismatch is the presence of noise during the testing phase of an ASR system. Various techniques have been used by different researchers in front and backend phases of ASR, to detect and handle the noise. However, a very few review papers have considered noise as a criterion to present the comparison among the existing research works. Hence, the objective of this survey is to analyze and review all the effective methods proposed by different scientists and researchers to boost the noise robustness of an ASR system. Initially, the paper discusses the basic architecture of an ASR system, the factors affecting the its performance, and noise problem formulation. Secondly, the work analysis existing state of the art noise robust ASR methods in terms of front end feature extraction techniques and backend classification model. Then, a detailed review in terms of various speech databases, that are used by these methods, is given. Finally, an analysis in terms of performance metrics of all these noise-resistant ASR techniques is presented. Also, the paper discusses various feature extraction techniques, backend classification methods, different speech databases and performance metrics in detail, while presenting the analysis. The paper also discusses the existing challenges, and describes future research directions in the area of building noise-resistant ASR systems. © 2023, The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature.

关键词： Feature extraction

来源：评论

学校读者我要写书评

暂无评论

An Ingenious IoT Based Crop Prediction System Using ML and EL

引用

computers, Materials & Continua 2024年第4期79卷 183-199页

作者： Shabana Ramzan Yazeed Yasin Ghadi Hanan Aljuaid Aqsa Mahmood Basharat Ali Department of Computer Science&IT Government Sadiq CollegeWomen UniversityBahawalpur63100Pakistan Department of Computer Science/Software Engineering Al Ain UniversityAl Ain64141UAE Department of Computer Sciences College of Computer and Information SciencesPrincess Nourah bint Abdulrahman University(PNU)Riyadh11671Saudi Arabia Agronomic Research Station Bahawalpur63100Pakistan

Traditional farming procedures are time-consuming and expensive as based on manual labor. Farmers haveno proper knowledge to select which crop is suitable to grow according to the environmental factors and soilcharacteristics. This is the main reason for the low yield of crops and the economic crisis in the agricultural sectorof the different countries. The use of modern technologies such as the Internet of Things (IoT), machine learning,and ensemble learning can facilitate farmers to observe different factors such as soil electrical conductivity (EC),and environmental factors like temperature to improve crop yield. These parameters play a vital role in suggestinga suitable crop to cope the food scarcity. This paper proposes a systemcomprised of twomodules, first module usesstatic data and the second module takes hybrid data collection (IoT-based real-time data and manual data) withmachine learning and ensemble learning algorithms to suggest the suitable crop in the farm to maximize the *** is used to train the model that predicts the crop. This system proposed an intelligent and low-cost solutionfor the farmers to process the data and predict the suitable *** implemented the proposed system in the *** efficiency and accuracy of the proposed system are confirmed by the generated results to predict the crop.

关键词： Machine learning Internet of Things sensors ensemble learning

来源：评论

学校读者我要写书评

暂无评论

Modified Buffalo Optimization with Big Data Analytics Assisted Intrusion Detection Model

引用

computer Systems Science & engineering 2023年第8期46卷 1415-1429页

作者： R.Sheeba R.Sharmila Ahmed Alkhayyat Rami Q.Malik Department of Computer Science and Engineering K.Ramakrishnan College of EngineeringTiruchirappalli621112India Department of Computer Applications Dhanalakshmi Srinivasan Engineering CollegePerambalur621212India College of Technical Engineering The Islamic UniversityNajafIraq Medical Instrumentation Techniques Engineering Department Al-Mustaqbal University CollegeBabylonIraq

Lately,the Internet of Things(IoT)application requires millions of structured and unstructured data since it has numerous problems,such as data organization,production,and *** address these shortcomings,big data analytics is the most superior technology that has to be *** though big data and IoT could make human life more convenient,those benefits come at the expense of *** manage these kinds of threats,the intrusion detection system has been extensively applied to identify malicious network traffic,particularly once the preventive technique fails at the level of endpoint IoT *** cyberattacks targeting IoT have gradually become stealthy and more sophisticated,intrusion detection systems(IDS)must continually emerge to manage evolving security *** study devises Big Data Analytics with the Internet of Things Assisted Intrusion Detection using Modified Buffalo Optimization Algorithm with Deep Learning(IDMBOA-DL)*** the presented IDMBOA-DL model,the Hadoop MapReduce tool is exploited for managing big *** MBOA algorithm is applied to derive an optimal subset of features from picking an optimum set of feature ***,the sine cosine algorithm(SCA)with convolutional autoencoder(CAE)mechanism is utilized to recognize and classify the intrusions in the IoT network.A wide range of simulations was conducted to demonstrate the enhanced results of the IDMBOA-DL *** comparison outcomes emphasized the better performance of the IDMBOA-DL model over other approaches.

关键词： Big data analytics internet of things security intrusion detection deep learning

来源：评论

学校读者我要写书评

暂无评论

Efficient Deep Learning Approach for Arthritis Prediction 2

Efficient Deep Learning Approach for Arthritis Prediction

引用

2nd International Conference on Artificial Intelligence, Blockchain, and Internet of Things, AIBThings 2024

作者： Khalil, Kasem Niloy, Samiul Islam Mohaidat, Tamador Khan, Md Rahat Kader Bayoumi, Magdy University of Mississippi Department of Electrical and Computer Engineering MS United States University of Louisiana at Lafayette Department of Electrical and Computer Engineering LA United States

ISBN: (纸本)9798331529529

Arthritis prediction allows for early diagnosis and timely intervention, significantly improving patient outcomes and quality of life. Using advanced imaging techniques and machine learning (ML), accurate prediction can reduce the burden on healthcare systems and enable personalized treatment plans. This paper presents a novel method for predicting arthritis using a combination of pre-processing, Discrete Fourier Transform (DFT), and Convolutional Neural Network (CNN) techniques. The pre-processing stage standardizes and enhances X-ray images through normalization and noise reduction using Gaussian filtering. The DFT stage transforms these pre-processed images into the frequency domain, enabling more effective feature extraction. The CNN stage, comprising five convolutional layers, further processes these features to predict arthritis with high accuracy. The proposed method uses a dataset from Kottayam Medical College in Kerala, comprising 564 grayscale MRI scans of the hands and wrists. Experimental results demonstrate a prediction accuracy of 97%. Implementation on the Altera 10 GX FPGA using VHDL shows efficient resource utilization. The total power consumption is measured at 650 mW. The proposed method demonstrates significant efficiency in hardware implementation, making it a viable solution for real-time arthritis detection and analysis in clinical settings. © 2024 IEEE.

关键词： Discrete Fourier transforms

来源：评论

学校读者我要写书评

暂无评论

Optimal Wavelet Neural Network-Based Intrusion Detection in Internet of Things Environment

引用

computers, Materials & Continua 2023年第5期75卷 4467-4483页

作者： Heba G.Mohamed Fadwa Alrowais Mohammed Abdullah Al-Hagery Mesfer Al Duhayyim Anwer Mustafa Hilal Abdelwahed Motwakel Department of Electrical Engineering College of EngineeringPrincess Nourah bint Abdulrahman UniversityP.O.Box84428Riyadh 11671Saudi Arabia Department of Computer Sciences College of Computer and Information SciencesPrincess Nourah bint Abdulrahman UniversityP.O.Box84428Riyadh 11671Saudi Arabia Department of Computer Science College of ComputerQassim UniversitySaudi Arabia Department of Computer Science College of Computer Engineering and SciencesPrince Sattam bin Abdulaziz UniversityAl-Kharj16273Saudi Arabia Department of Computer and Self Development Preparatory Year DeanshipPrince Sattam bin Abdulaziz UniversityAlKharjSaudi Arabia

As the Internet of Things(IoT)endures to develop,a huge count of data has been *** IoT platform is rather sensitive to security challenges as individual data can be leaked,or sensor data could be used to cause *** typical intrusion detection system(IDS)studies can be frequently designed for working well on databases,it can be unknown if they intend to work well in altering network *** learning(ML)techniques are depicted to have a higher capacity at assisting mitigate an attack on IoT device and another edge system with reasonable *** article introduces a new Bird Swarm Algorithm with Wavelet Neural Network for Intrusion Detection(BSAWNN-ID)in the IoT *** main intention of the BSAWNN-ID algorithm lies in detecting and classifying intrusions in the IoT *** BSAWNN-ID technique primarily designs a feature subset selection using the coyote optimization algorithm(FSS-COA)to attain ***,to detect intrusions,the WNN model is *** last,theWNNparameters are optimally modified by the use of *** experiment is performed to depict the better performance of the BSAWNNID *** resultant values indicated the better performance of the BSAWNN-ID technique over other models,with an accuracy of 99.64%on the UNSW-NB15 dataset.

关键词： Internet of things wavelet neural network security intrusion detection machine learning

来源：评论

学校读者我要写书评

暂无评论

Early detection of stroke disease using patients previous medical data instil with deep learning

引用

Multimedia Tools and Applications 2025年第16期84卷 16853-16881页

作者： Diwan, Tausif Gajbhiye, Saurav M. Goydani, Purva R. Gannarpwar, Vedant R. Khandait, Harshal R. Tembhurne, Jitendra V. Sahare, Parul Department of Computer Science and Engineering Indian Institute of Information Technology Maharashtra Nagpur441108 India Department of Electronics and Communication Engineering Indian Institute of Information Technology Maharashtra Nagpur441108 India

Early detection of any disease and starting its treatment in this early stage are the most important steps in case of any life-threatening disease. Stroke is not an exception in this regard which is one of the leading causes of death and disability worldwide. We develop a simple but efficient deep neural network for the stroke prediction that accurately evaluates the probability of occurrence of stroke disease by treating this as a binary classification problem on one of the standard datasets named as Stroke Prediction Dataset available on Kaggle. With the help of effective pre-processing techniques such as SMOTE and FastICA on the noisy and imbalanced dataset, we could achieve improved performance for the binary classification of stroke prediction by employing a deep dense neural network. The architectural hyper-parameters are critically designed with the help of Keras Tuner and consisting of just ten layers in the deep dense neural network. With the help of this light weight deep dense model, we report an accuracy of 95%, AUC Score of 100% and F1-Score of 95% on a testing set. Moreover, we also present comparative illustration of various machine learning models for the aforesaid task along with the comparative illustration of various state of the arts with our proposed model for the stroke prediction. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2024.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

Research on Indoor Positioning Technology of WSN based on T-RL Partition Path Model

引用

Journal of Communications and Information Networks 2024年第3期9卷 219-232页

作者： Wei Wang Xinlin Wang Yutong Liu Yulin Ren Maozhen Li Asoke K.Nandi School of Information and Communication Engineering North University of ChinaTaiyuan 030051China Department of Electronic and Computer Engineering Brunel University LondonUxbridge UB83PHUK

To address the issues of unstable received signal strength indicator(RSSI)and low indoor positioning accuracy caused by walls and obstacles,the propagation conditions of the wireless communication system are categorized into two distinct environments:line-of-sight(LOS)and non-line-of-sight(NLOS).In the LOS environment,the traditional logarithmic path loss model is *** the NLOS environment,the impact of walls on signal transmission is considered,leading to the development of a multi-wall path loss model based on the T-RL method,with improvements made to the key parameter,the Fresnel coefficient *** breakpoint value d=2.3 m in the partitioned model is determined,and the positional coordinates of the unknown nodes are calculated using the trilateration *** results indicate that the T-RL based multi-wall model improves localization accuracy by 47%in NLOS environments compared to the traditional logarithmic path loss *** average localization error using the T-RL partitioned path loss model is 0.7021 m,representing a 55.9%improvement over the logarithmic path loss model and a 16.8%enhancement over the T-RL attenuation multi-wall model,thereby providing better environmental adaptability.

关键词： WSN indoor positioning loss model reflection transmission RSSI

来源：评论

学校读者我要写书评

暂无评论

A Comprehensive Real-World Evaluation of 5G Improvements Over 4G in Low- and Mid-Bands

引用

IEEE Transactions on Cognitive Communications and Networking 2025年第3期11卷 1427-1441页

作者： Rochman, Muhammad Iqbal Ye, Wei Zhang, Zhi-Li Ghosh, Monisha University of Notre Dame Department of Electrical Engineering Notre DameIN46556 United States University of Minnesota Twin Cities Department of Computer Science and Engineering MinneapolisMN55455 United States

With the proliferation of 5G and developments of 6G technologies already underway, understanding the real-world performance of various 5G enhancements such as higher modulation, beamforming, and MIMO of deployed 5G over 4G is vital. This work addresses this knowledge gap by conducting extensive 4G/5G measurements in low- ( © 2015 IEEE.

关键词： Beamforming

来源：评论

学校读者我要写书评

暂无评论

Child Detection by Utilizing Touchscreen Behavioral Biometrics on Mobile 7th

Child Detection by Utilizing Touchscreen Behavioral Biometr...

引用

7th International Conference on Inventive Communication and Computaional Technologies, ICICCT 2023

作者： Mane, Sunil Mandhan, Kartik Bapna, Mudit Terwadkar, Atharv Department of Computer Engineering College of Engineering Pune India

ISBN: (纸本)9789819951659

Studies show that young children are exposed to smart devices. Early stages of education, children’s internet safety, and children’s interactions with computers are all significantly impacted by mobile usage. With the significant increase in average daily smartphone use, preschoolers’, school-going children’s, and teenagers’ health is ruptured both mentally and physically. This research presents an extensive point of reference for investigating the operation of artificial neural networks and several machine learning algorithms in postulating the classification problem regarding the age of a smartphone user based on their touch. Hence, our work will become a reliable component to forecast the age range and effectively help parents control the time spent on smartphones by their children. Out of a variety of models that were examined, we propose the top five for simulating the behavioral differences between children and adults when using mobile devices. The first approach is based on using artificial neural networks for classification. Next, we used gradient boosting algorithms like XgBoost classifier and LightGBM classifier. Finally, we also analyzed support vector machines and K-nearest neighbors model. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Smartphones

来源：评论

学校读者我要写书评

暂无评论

Unsafe-Net: YOLO v4 and ConvLSTM based computer vision system for real-time detection of unsafe behaviours in workplace

引用

Multimedia Tools and Applications 2024年 1-27页

作者： Önal, Oğuzhan Dandıl, Emre Department of Electronic and Automation Vocational School Bilecik Seyh Edebali University Bilecik Turkey Department of Computer Engineering Faculty of Engineering Bilecik Seyh Edebali University Bilecik Turkey

Unsafe behaviour is a leading cause of death or injury in the workplace, including many accidents. Despite regular safety inspections in workplaces, many accidents occur as a result of breaches of occupational health and safety protocols. In these environments, despite efforts to prevent accidents and losses in hazardous environments, human error cannot be completely eliminated. In particular, in computer-based solutions, automated behaviour detection has low accuracy, is very costly, not real-time and requires a lot of time. In this study, we propose Unsafe-Net, a hybrid computer vision approach using deep learning models for real-time classification of unsafe behaviours in workplace. For the Unsafe-Net, a dataset is first specifically created by capturing 39 days of video footage from a factory. Using this dataset, YOLO v4 and ConvLSTM methods are combined for object detection and video understanding to achieve fast and accurate results. In the experimental studies, the classification accuracy of unsafe behaviours using the proposed Unsafe-Net method is 95.81% and the average time for action recognition from videos is 0.14 s. In addition, the Unsafe-Net has increased the real-time detection speed by reducing the average video duration to 1.87 s. In addition, the system is installed in a real-time working environment in the factory and employees are immediately alerted by the system, both audibly and visually, when unsafe behaviour occurs. As a result of the installation of the system in the factory environment, it has been determined that the recurrence rate of unsafe behaviour has been reduced by approximately 75%. © The Author(s) 2024.

关键词： Object detection

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：