检索结果-内蒙古大学图书馆

2024 International Conference on Advancements in Power, communication and Intelligent Systems, APCI 2024

作者： Tummalapalli, Geetamma Prasad, Ch. Babji Kousik, Jami GMR Institute of Technology Electronics and Communication Engineering Vizianagaram India Electronics and Communication Engineering India

ISBN: (纸本)9798350363289

Tuberculosis (TB) is an infectious disease caused by the elusive Mycobacterium tuberculosis, leading to a persistent lung condition due to bacterial infection. The identification of tuberculosis (TB) is crucial, and chest X-ray images are routinely employed;nevertheless, manual interpretation is time-consuming and prone to errors. This research looked at;we introduced an advanced deep learning approach for classifying TB using chest X-rays. Our method involves leveraging deep learning algorithms to explore and identify optimal hypotheses for accurate predictions of TB. To address challenges like vanishing gradients and maximize feature reuse, we adopted the DenseNet architecture, well-known for its capabilities. This proposed model builds upon previous research, enhancing the accuracy of TB prediction using DenseNet, a powerful deep learning technique. The developed classifier, utilizing deep learning, achieved an impressive accuracy rate of 96%. This advancement holds promise for more efficient and reliable TB diagnosis through automated analysis of chest X-ray images. © 2024 IEEE.

关键词： Lung cancer

来源：评论

学校读者我要写书评

暂无评论

Optimized Convolutional Neural Network at the IoT edge for image detection using pruning and quantization

引用

Multimedia Tools and Applications 2025年第9期84卷 5435-5455页

作者： Naveen, Soumyalatha Kounte, Manjunath R Department of Computer Science and Engineering Manipal Institute of Technology Bengaluru Manipal Academy of Higher Education Manipal India School of Electronics and Communication Engineering REVA University Bangalore India

Most real-time computer vision applications heavily rely on Convolutional Neural Network (CNN) based models, for image classification and recognition. Due to the computationally and memory-intensive nature of the CNN model, it’s challenging to deploy on resource-constrained Internet of Things (IoT) devices to enable Edge intelligence for real-time decision-making. Edge intelligence requires minimum inference latency, memory footprint, and energy-efficient model. This work aims to develop an energy-efficient deep learning accelerator using a 3-stage pipeline: Training, Weight-pruning, and Quantization to reduce the model size and optimize the resources. First, we employ YOLOv3, a CNN architecture to detect objects in an image on the trained data. In addition, a sparse network of YOLO has been created by using pruning, which helps to improve the network’s performance and efficiency by reducing the computational requirements. Finally, we utilize 8-bit quantization to reduce the precision of the weights and activations, in a neural network. The evaluation of our proposed model shows that combining pruning and 8-bit quantization improves the efficiency and performance of the model. While pruning shows a decline of 80.39% in model parameters. The combination of 8-bit quantization results in an improvement in inference latency by 22.72% compared to existing SQuantization approach and a reduction of energy consumption by 29.41%. © The Author(s) 2024.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Dental Radiography Analysis and Diagnosis using YOLOv8 9

Dental Radiography Analysis and Diagnosis using YOLOv8

引用

9th International Conference on Smart Computing and communications, ICSCC 2023

作者： George, James Hemanth, T.S. Raju, Joshua Mattapallil, Johan George Naveen, N. Electronics and Communication Engineering Rajagiri School of Engineering and Technology Cochin India Rajagiri School of Engineering and Technology Dept. of Electronics and Communication Engineering Cochin India

ISBN: (纸本)9798350314090

Panoramic dental radiography is a widely used diagnostic tool in dentistry that provides valuable information about a patient's oral health. These two-dimensional images of the entire upper and lower jaw offer a comprehensive view of the oral cavity and allow for the diagnosis of various conditions, including tooth decay, periodontal diseases, and jaw abnormalities. The use of Gaussian and Laplacian pyramids for image enhancement is a common technique employed in this field. The GVF Snake technique, a computer vision approach, is utilized for dental radiograph analysis. It has proven effective in tasks such as tooth segmentation, root canal detection, and tooth movement analysis, improving the accuracy and efficiency of dental diagnosis and treatment planning. Manual annotation, model training, and deployment are facilitated by the computer vision platform Roboflow. The YOLOv8 training model, chosen for this research, has shown promising results with a precision of 82.36% in detecting and classifying dental diseases such as cavities, periodontal disease, and oral cancers. By training the model on a large dataset of dental radiographs, it can accurately identify different types of dental diseases, leading to early detection and more effective treatment. Overall, the integration of deep learning algorithms, image enhancement techniques, computer vision, and machine learning methods holds great potential in improving patient outcomes by enabling early detection, accurate diagnosis, and better management of dental conditions. © 2023 IEEE.

关键词： Diseases

来源：评论

学校读者我要写书评

暂无评论

A Fuzzy inference anisotropic diffusion based high boost filter for denoising and enhancement of brain MR image

引用

Multimedia Tools and Applications 2024年 1-43页

作者： Kumar, Vinay Kumar, Abhinav Srivastava, Subodh Department of Electronics and Communication Engineering National Institute of Technology Patna Patna800005 India

Magnetic resonance imaging is a well-proven clinical tool that emphasizes brain abnormalities in terms of tumours and clots. However, during the image formation process, multiplicative Rician noise affects the MR images. Moreover, the presence of noise reduces the readable features of the image, which potentially leads an expert to falsely interpret the disease. To address these challenges, a fuzzy inference anisotropic diffusion-based high boost filter i.e., FIADHBF has been proposed. The proposed FIADHBF is an enhanced high boost filter that incorporates two advancements. The first advancement replaces the HBF's smoothing filter with a fuzzy inference-based anisotropic diffusion filter. In lieu of manual and user-opted values, the second advancement employs a fuzzy inference system to automate the calculation of conventional anisotropic diffusion coefficients. The T1, T2, and PD-weighted MR images with 3% to 9% noise variance of Brain Web datasets have been utilized for the development of the proposed method. The performance has been assessed in terms of comparative qualitative and quantitative analysis with respect to existing methodologies. The assessment takes into account the human visual system, full and no reference parameters. The metrics MSE, PSNR CP, NAE, UQI, SSIM, EME, EMEE, SDME, AG, and PSI yield the following values for the 9% noise variance of the T1 modality: 0.0321, 65.2917, 0.9215, 0.0948, 0.9612, 0.9502, 5.0798, 1.6898, 51.2726, 22.5499, and 0.6576, respectively. The results show that the proposed method is the best suited approach for removing Rician noise in a single framework, along with edge preservation, fine detail restoration, and abnormality enhancement. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2024.

关键词： Magnetic resonance imaging

来源：评论

学校读者我要写书评

暂无评论

Progress on Medical Implant:A Review and Prospects

引用

Journal of Bionic engineering 2023年第2期20卷 470-494页

作者： Ankur Pandey Swagatadeb Sahoo Department of Electronics and Communication Engineering National Institute of Technology JamshedpurJamshedpurJharkhand831014India

Medical implant from different materials such as metals,ceramics,polymers and composites have gained a lot of research attraction due to wide applications in medical industry for treatment,surgical operations and preparing artificial body *** this work,we highlight a comprehensive review of medical implant mechanism,various types of implant materials,factors affecting the performance of implant and different characterization *** review provides an overall summary of the state-of-the-art progress on various interesting and promising material-based medical ***,few new prospects are explained from the established theoretical and experimental results for real-life *** study is expected to promote extended interest of scientists and engineers in recent trend of modern biomaterials based medical implant.

关键词： Implant material Biocompatibility Hydroxyapatite Dielectric constant

来源：评论

学校读者我要写书评

暂无评论

Development of innovative cloud-based IoMD architecture for elder population monitoring

引用

International Journal of Cloud Computing 2023年第2-4期12卷 324-339页

作者： Almurisi, Nasr Musaed S. Tadisetty, Srinivasulu Department of Electronics & Communication Engineering College of Engineering and Technology Kakatiya University Warangal India

The rapid evolution in IoT technology has facilitated the development of smart applications in different sectors. Nowadays, IoT paradigm allows smart objects to be interconnected with medical devices to provide healthcare services. Recently, many healthcare systems (HS) have been proposed for integrating IoT with medical things. However, the performance of existing architecture is decreased when medical data is increased. Therefore, this article proposes a novel internet of medical device (IoMD) architecture to overcome the limitations of conventional HS. Since medical data are reaching the cloud, hence we propose a neural network method called deep LSTM that runs in the cloud platform and capable of processing huge data in a fast and efficient manner. A real dataset from UCL Repository is used to train and validate the proposed method. However, the analysis of the results confirms that our method can classify the activities with an accuracy of 98.60%. Copyright © 2023 Inderscience Enterprises Ltd.

关键词： Long short-term memory

来源：评论

学校读者我要写书评

暂无评论

Continuous feature learning representation to XGBoost classifier on the aggregation of discriminative Features using DenseNet-121 architecture and ResNet 18 architectures towards Apraxia Recognition in the Child Speech Therapy

引用

International Journal of Speech technology 2024年第1期27卷 187-199页

作者： Ashwini, P. Bharathi, S.H. Research Scholar School of Electronics and Communication Engineering REVA University Karnataka Bangalore560064 India School of Electronics and Communication REVA University Yelahanka Bangalore India

Due to the peculiar cell growth, apraxia is one of the common types of speech stuttering seen in youngsters. Machine learning-based architecture has been used to automatically categorize the disfluency based on the feature and its properties, however, these models use more energy, and processing time, and will have worse scalability and dependability. Deep learning architecture has been used to overcome these restrictions as it is more effective and efficient at differentiating speech characteristics. Utilizing ResNet -18 convolutional neural networks and DenseNet-121 convolution network architectures, continuous feature learning representation has been developed as a feature extraction technique. The combined extracted feature is given to the XGBoost Classifier for Apraxia identification. Utilizing Ultra Suite Repository, the suggested model's experimental findings have been assessed in Python. When compared to traditional methods, the suggested model's performance analysis produced 98 percent accuracy, which is evaluated. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2024.

关键词： Speech recognition

来源：评论

学校读者我要写书评

暂无评论

Energy-Efficient Design for Logic Circuits Using a Leakage Control Configuration in FinFET technology

引用

Journal of The Institution of Engineers (India): Series B 2024年第4期105卷 903-911页

作者： Ul Haq, Shams Sharma, Vijay Kumar School of Electronics and Communication Engineering Shri Mata Vaishno Devi University Katra India

The multigate technology adopted in the latest processors is based on the Fin-shaped field effect transistor (FinFET). In the pursuit of increasing the computational power of the metal oxide semiconductor field effect transistor (MOSFET), ceaseless downscaling drove the channel length of MOSFETs to be comparable to their depletion widths around drain and source terminals. This led to a sequence of issues, which include drain-induced barrier lowering (DIBL), roll-off of threshold voltage, leakage power, hot carrier effects, mobility reduction, and an increase in reverse leakage current. The use of MOSFET was questioned at lower technology nodes because short channel effects (SCEs) lead to the loss of gate terminal command on the channel. This research work proposes a leakage power control technique based on the reverse body bias effect. The transistor configuration has been proposed to put a curb on the leakage power. The results have been approved on some fundamental logic gates and a C17 ((ISCAS-85) benchmark circuit. The shorted-gate (SG) FinFET-based proposed inverter confirms a leakage power and power delay product (PDP) reduction of 51.09% and 25.90%, respectively. The power optimization and PDP optimization of 45.47% and 49.12% have been observed in the low-power NAND gate. The proposed 2-input NOR gate depicts a leakage power and PDP mitigation by 69.60% and 70.58%, respectively. Similarly, the SG FinFET-based proposed C17 benchmark circuit shows leakage power and PDP reduction of 56.17% and 43.20%, respectively, as compared to the conventional circuit. Monte Carlo (MC) analysis is completed with a ± 10% digression in operating conditions to calculate reliability. © The Institution of Engineers (India) 2024.

关键词： Timing circuits

来源：评论

学校读者我要写书评

暂无评论

Performance of MIMO-GFDM scheme for Future Wireless communications

Performance of MIMO-GFDM scheme for Future Wireless Communic...

引用

2024 International Conference on Knowledge engineering and communication Systems, ICKECS 2024

作者： Kumar, R. Anil Kumar, R. Prakash Patchala, Sarala Sreenivasulu, U. Aditya College of Engineering and Technology Electronics and Communication Engineering Surampalem India Cvr College of Engineering Electronics and Communication Engineering Hyderabad India Srkr Engineering College Electronics and Communication Engineering Bhimavaram India Kkr & Ksr Institute of Technology and Sciences Electronics and Communication Engineering Guntur India

ISBN: (纸本)9798350359688

Wireless technology became more popular from couple of decades. Modern Technologies such as Mobile Broad Band communications, Ultra Low Latency communications and Internet of Things purely depends on Wireless technologies. The use of set of antennas at User Equipment and/or Base Station improves the User Capacity, Bit Error Rate, Spectral Efficiency and Data Rates. Current research on Generalized Frequency Division Multiplexing Technique combined with Multi-Input Multi-Output (MIMO Systems) gives good resource allocation and better signal strengths over the existing Orthogonal Frequency Division Multiplexing (OFDM). This paper explains the performance of MIMO-GFDM in an ultradense multipath environment with affordable complexity by considering Uplink scenario at User Equipment (UE). The simulation results such as Spectral Efficiency, User Capacity, Bit Error Rate and Data rates are calculated for different set of antenna systems. © 2024 IEEE.

关键词： Orthogonal frequency division multiplexing

来源：评论

学校读者我要写书评

暂无评论

Capacity enhancement of WDM visible light communication system employing 3-SOPs/channel/LD color

引用

Journal of Optical communications 2025年第1期46卷 107-113页

作者： Pahuja, Hunny Sachdeva, Shippu Sindhwani, Manoj Department of Electronics and Communication Engineering KIET Group of Institutions Delhi-NCR Ghaziabad India School of Electronics and Electrical Engineering Lovely Professional University Phagwara India

Nowadays, polarization division multiplexing (PDM) employing different states of polarizations (SOP) is a prominent candidate to offer high capacity and improved performance. However, the incorporation of PDM in indoor wireless optical communication (WOC) systems is limited to horizontal and vertical SOPs only. Therefore, in this research article, a cost-effective and high capacity 90 Gbps non-return to zero (NRZ) line coding based 18 channels VLC system is presented by incorporating red-green-blue laser diodes (RGB LDs). Performance investigations of different SOPs such as SOP 0°, SOP 45°, and SOP 90° are carried out at diverse VLC link distances in terms of log BER and Q factor. Results revealed that the proposed VLC system can successfully cover the 6 m distance and red LD together with SOP0 is found to be the highest performing. © 2022 Walter de Gruyter GmbH, Berlin/Boston.

关键词： Polarization

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：