检索结果-内蒙古大学图书馆

Convolutional neural network for classification of SiO2 scanning electron microscope images

International Journal of Business Intelligence and Data Mining 2022年第1期21卷 118-128页

作者： Jayaram, Kavitha Prakash, G. Jayaram, V. Department of Computer Science and Engineering Amrita School of Engineering Amrita Vishwa Vidyapeetham Bengaluru India Solid State and Structural Chemistry Unit Indian Institute of Science Bengaluru India

The recent development in deep learning has made image and speech classification and recognition tasks possible with better accuracy. An attempt was made to automatically extract required sections from literature published in journals to analyse and classify them according to their application. This paper presents high-temperature materials classification into four categories according to their wide applications such as electronic, high temperature, semiconductors, and ceramics. The challenging act is to extract SEM images’ unique features as they are microscopic with different resolutions. A total of 10,000 scanning electron microscope (SEM) images are classified into two labelled categories namely crystalline and amorphous structure. The image classification and recognition process of SiO2 was implemented using convolutional neural network (CNN) deep learning framework. Our algorithm successfully classified with a precision of 96% and accuracy of 95.5% of the test dataset of SEM images. Copyright © 2022 Inderscience Enterprises Ltd.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

MLSTM: A Novel Modified Long Short-Term Memory Networks for Melody Synthesis Melody Generation

MLSTM: A Novel Modified Long Short-Term Memory Networks for ...

引用

International Conference on Advanced Communications and Machine Intelligence, MICA 2023

作者： Ganguly, Tanudeep Sharma, Sushmita Deb, Suman Ezugwu, Absalom El-Shamir Saha, Apu Kumar Department of Computer Science and Engineering National Institute of Technology Agartala Tripura Agartala India Department of Computer Science and Engineering Indian Institute of Information Technology Agartala Tripura Agartala India Unit for Data Science and Computing North-West University Potchefstroom2520 South Africa

ISBN: (纸本)9789819762217

This research elaborates a melody synthesis model utilizing a Modified Long Short-Term Memory (mLSTM) neural network, trained on Kern dataset, preprocessed into Musical Instrument Digital Interface (MIDI) files. After receiving a seed melody as input 01, the model predicts the note that will come next by looking at the notes that have come before it. The model can learn the structures and patterns seen in classical music by using MIDI files as the training dataset. Categorical cross-entropy loss function is used to train the model over several epochs with early halting. To regulate how random the predictions are, the model makes use of a temperature parameter. The assessment demonstrates that the produced tunes have a variety of traits and are aesthetically pleasing to human ears. To justify the performance of the proposed method some criteria were considered and expert evaluation was done. The evaluation shows that the model has generated the melodies that are pleasing to the ear and have a similar structure to the training dataset. The model has several uses, including prediction and music composition. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Long short-term memory

来源：评论

学校读者我要写书评

暂无评论

Improving Lung CT Analysis through Fuzzy Dilated Convolution Attention 3

Improving Lung CT Analysis through Fuzzy Dilated Convolution...

引用

3rd IEEE Applied Signal Processing Conference, ASPCON 2023

作者： Das, Rangan Bhattacharya, Samayan Jana, Swadesh Maulik, Ujjwal Bandyopadhyay, Sanghamitra Jadavpur University Department of Computer Science and Engineering Kolkata India Indian Statistical Institute Machine Intelligence Unit Kolkata India

ISBN: (纸本)9798350325881

Computed Tomography (CT) is a versatile imaging modality that can detect and evaluate a wide range of diseases and conditions affecting organs in the thoracic cavity. computer vision methods, particularly, deep learning architectures using attention mechanisms, have been extensively used for analysis of CT images. However, there is still a need for increased accuracy when it comes to pulmonary CT analysis owing to the complex structures visible in the images. In this study, we propose a novel convolution layer, called fuzzy dilated convolution and develop an attention module. We build an encoder-decoder network using the proposed module and train it on two large publicly available lung CT datasets - LUNA16 and LIDC-IDRI. The performance of the proposed network surpasses state-of-the-art segmentation models on both the datasets. The implementation is available at ***/Swadesh13/Fuzzy-Dilated-Convolution. © 2023 IEEE.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

Robust graph‐based localization for industrial Internet of things in the presence of flipping ambiguities

引用

CAAI Transactions on Intelligence Technology 2023年第4期8卷 1140-1149页

作者： Mian Imtiaz ul Haq Ruhul Amin Khalil Muhannad Almutiry Ahmad Sawalmeh Tanveer Ahmad Nasir Saeed Faculty of Electrical Engineering Swedish College of Engineering and TechnologyWah CanttPakistan Department of Electrical Engineering Faculty of Electrical and Computer EngineeringUniversity of Engineering and TechnologyPeshawarPakistan Department of Electrical Engineering Remote Sensing UnitNorthern Border UniversityArarSaudi Arabia Data Science and Artificial Intelligence Department‐College of Science and Information Technology Irbid National UniversityIrbadJordan Innovation Education and Research Center for On‐Device AI Software(Bk21) Department of Computer Science and EngineeringChungnam National UniversityDaejeonRepublic of Korea Department of Electrical and Communication Engineering United Arab Emirates University(UAEU)Al AinUnited Arab Emirates

Localisation of machines in harsh Industrial Internet of Things(IIoT)environment is necessary for various ***,a novel localisation algorithm is proposed for noisy range measurements in IIoT *** position of an unknown machine device in the network is estimated using the relative distances between blind machines(BMs)and anchor machines(AMs).Moreover,a more practical and challenging scenario with the erroneous position of AM is considered,which brings additional uncertainty to the final position ***,the AMs selection algorithm for the localisation of BMs in the IIoT network is *** those AMs will participate in the localisation process,which increases the accuracy of the final location ***,the closed‐form expression of the proposed greedy successive anchorization process is derived,which prevents possible local convergence,reduces computation,and achieves Cramér‐Rao lower bound accuracy for white Gaussian measurement *** results are compared with the state‐of‐the‐art and verified through numerous simulations.

关键词： Cramér‐Rao lower bound greedy successive anchorization industrial internet of things localization

来源：评论

学校读者我要写书评

暂无评论

Low-Depth Flag-Style Syndrome Extraction for Small Quantum Error-Correction Codes 4

Low-Depth Flag-Style Syndrome Extraction for Small Quantum E...

引用

4th IEEE International Conference on Quantum Computing and engineering, QCE 2023

作者： Bhatnagar, Dhruv Steinberg, Matthew Elkouss, David Almudever, Carmen G. Feld, Sebastian Delft University of Technology Quantum & Computer Engineering Department Delft2628 CD Netherlands Oist Graduate University Networked Quantum Devices Unit Okinawa Japan Technical University of Valencia Computer Engineering Department Valencia Spain

ISBN: (纸本)9798350343236

Flag-style fault-tolerance has become a linchpin in the realization of small fault-tolerant quantum-error correction experiments. The flag protocol's utility hinges on low qubit over-head, which is typically much smaller than in other approaches. However, as in most fault-tolerance protocols, the advantages of flag-style error correction come with a tradeoff: fault tolerance can be guaranteed, but such protocols involve high-depth circuits, due to the need for repeated stabilizer measurements. Here, we demonstrate that a dynamic choice of stabilizer measurements, based on past syndromes, and the utilization of elements from the full stabilizer group, leads to flag protocols with lower-depth syndrome-extraction circuits for the [[5], [1], [3]] code, as well as for the Steane code when compared to the standard methods in flag fault tolerance. We methodically prove that our new protocols yield fault-tolerant lookup tables, and demonstrate them with a pseudothreshold simulation, showcasing large improvements for all protocols when compared to previously-established methods. This work opens the dialogue on exploiting the properties of the full stabilizer group for reducing circuit overhead in fault-tolerant quantum-error correction. © 2023 IEEE.

关键词： Extraction

来源：评论

学校读者我要写书评

暂无评论

Deep Learning Framework for Facial Reconstruction Outcome Prediction: Integrating Image Inpainting and Depth Estimation for computer-Assisted Surgical Planning

SSRN

引用

SSRN 2025年

作者： Manni, Guido Bini, Fabiano Marinozzi, Franco Unit of Computer Systems and Bioinformatics Department of Engineering University Campus Bio-Medico of Rome Italy Department of Mechanical and Aerospace Engineering Sapienza University of Rome Rome00184 Italy

Background: Facial reconstructive surgery requires preoperative planning to optimize functional and aesthetic outcomes. While current imaging technologies like CT and MRI provide detailed anatomical assessment, they do not offer visualization of expected post-surgical appearance, limiting surgical planning capabilities in facial reconstruction procedures like maxillofacial surgery, otolaryngology, and ophthalmic ***: To develop a deep learning framework that predicts surgical outcomes for facial reconstruction by integrating facial inpainting and monocular depth estimation models, enabling both 2D and 3D surgical planning from standard clinical ***: The framework integrates two key components: a deep learning facial inpainting model for predicting reconstructed facial features and a deep learning monocular depth estimation model for generating three-dimensional representations. We evaluated three state-of-the-art inpainting architectures (LaMa, LGNet, MAT) and two state-of-the-art monocular depth estimation approaches (ZoeDepth, DepthAnything) using the FFHQ dataset for inpainting and C3I-SynFace dataset for depth estimation. Models were evaluated using comprehensive quantitative metrics for both reconstruction quality and depth accuracy. Results: LGNet demonstrated optimal performance for facial reconstructions with missing regions covering 10-25% of the facial area, achieving superior scores in MSE (63.8104), PSNR (31.4162), and SSIM (0.9587). For depth estimation, ZoeDepth consistently outperformed DepthAnything, showing significantly better accuracy in absolute relative difference (0.8123 vs 1.6065) and RMSE-log (0.6721 vs 0.9512).Conclusion: This novel framework addresses a critical gap in surgical planning by providing a comprehensive preoperative tool that enables both 2D and 3D visualization of potential outcomes from standard clinical photographs, supporting various applications from maxillofacial reconstruction to orbital and

关键词： Data visualization

来源：评论

学校读者我要写书评

暂无评论

Endpoint Detection in Breast Images for Automatic Classification of Breast Cancer Aesthetic Results 1st

Endpoint Detection in Breast Images for Automatic Classifi...

引用

1st Deep Breast Workshop on AI and Imaging for Diagnostic and Treatment Challenges in Breast Care, Deep-Breath 2024

作者： Freitas, Nuno Veloso, Carlos Mavioso, Carlos Cardoso, Maria J. Oliveira, Hélder P. Cardoso, Jaime S. Faculdade de Engenharia Universidade do Porto Porto Portugal Institute for Systems and Computer Engineering Technology and Science Porto Portugal Breast Unit Champalimaud Foundation Lisbon Portugal

ISBN: (纸本)9783031777882

Breast cancer is the most common type of cancer in women worldwide. Because of high survival rates, there has been an increased interest in patient Quality of Life after treatment. Aesthetic results play an important role in this aspect, as these treatments can leave a mark on a patient’s self-image. Despite that, there are no standard ways of assessing aesthetic outcomes. Commonly used software such as *** or BAT require the manual annotation of keypoints, which makes them time-consuming for clinical use and can lead to result variability depending on the user. Recently, there have been attempts to leverage both traditional and Deep Learning algorithms to detect keypoints automatically. In this paper, we compare several methods for the detection of Breast Endpoints across two datasets. Furthermore, we present an extended evaluation of using these models as input for full contour prediction and aesthetic evaluation using the *** software. Overall, the YOLOv9 model, fine-tuned for this task, presents the best results considering both accuracy and usability, making this architecture the best choice for this application. The main contribution of this paper is the development of a pipeline for full breast contour prediction, which reduces clinician workload and user variability for automatic aesthetic assessment. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Lung cancer

来源：评论

学校读者我要写书评

暂无评论

Cross-Modality Calibration in Multi-Input Network for Axillary Lymph Node Metastasis Evaluation

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2024年第9期5卷 4823-4836页

作者： Gravina, Michela Santucci, Domiziana Cordelli, Ermanno Soda, Paolo Sansone, Carlo Naples80125 Italy University of Rome Campus Bio-Medico Department of Radiology Rome00128 Italy Umea University Department of Diagnostics and Intervention Radiation Physics Biomedical Engineering Umea90187 Sweden University of Rome Campus Bio-Medico Unit of Computer Systems and Bioinformaticss Department of Engineering Rome00128 Italy

The use of deep neural networks (DNNs) in medical images has enabled the development of solutions characterized by the need of leveraging information coming from multiple sources, raising the multimodal deep learning. DNNs are known for their ability to provide hierarchical and high-level representations of input data. This capability has led to the introduction of methods performing data fusion at an intermediate level, preserving the distinctiveness of the heterogeneous sources in modality-specific paths, while learning the way to define an effective combination in a shared representation. However, modeling the intricate relationships between different data remains an open issue. In this article, we aim to improve the integration of data coming from multiple sources. We introduce between layers belonging to different modality-specific paths a transfer module (TM) able to perform the cross-modality calibration of the extracted features, reducing the effects of the less discriminative ones. As case of study, we focus on the axillary lymph nodes (ALNs) metastasis evaluation in malignant breast cancer (BC), a crucial prognostic factor, affecting patient's survival. We propose a multi-input single-output 3-D convolutional neural network (CNN) that considers both images acquired with multiparametric magnetic resonance and clinical information. In particular, we assess the proposed methodology using four architectures, namely BasicNet and three ResNet variants, showing the improvement of the performance obtained by including the TM in the network configuration. Our results achieve up to 90% and 87% of accuracy and area under ROC curve, respectively when the ResNet10 is considered, surpassing various fusion strategies proposed in the literature. © 2020 IEEE.

关键词： Medical imaging

来源：评论

学校读者我要写书评

暂无评论

Caching Strategies for the Metaverse: Taxonomy, Open Challenges, and Future Research Directions

International Journal of Interactive Mobile Technologies

引用

International Journal of Interactive Mobile Technologies 2024年第14期18卷 4-27页

作者： Mohammed, Alaa Hamid Alaidaros, Hashem Salleh, Farizan Sajat, Mohd Samsu Ramaha, Nehad Habbal, Adib Department of Medical Instruments Techniques Engineering Al-Maarif University College Ramadi Iraq Department of Computer Engineering Faculty of Engineering Karabuk University Karabuk Turkey Cybersecurity Department Dar Al-Hekma University Jeddah Saudi Arabia Unit for Instructional Development & Multimedia Politeknik Sultan Idris Shah Selangor Malaysia School of Computing Universiti Utara Malaysia Kedah Malaysia

The metaverse, which is considered to be the next evolutionary stage of the Internet, has captured the attention of both academia and industry. Its primary goal is to establish a shared 3D virtual space that interconnects all virtual worlds through the Internet. In this shared space, users are represented as digital avatars, enabling them to communicate, interact with each other, and engage with the virtual environment as if they were in the physical world. However, realizing the full potential of the metaverse poses significant challenges, such as the requirement for higher throughput compared to current social VR platforms and the need to minimize latency to just a few milliseconds to uphold a truly immersive user experience. Caching is a critical aspect of optimizing data access on the current Internet, and it is equally crucial for addressing similar challenges in Web 3.0 and the metaverse. This paper explores different caching strategies suggested to address these challenges on the current Internet and assesses their potential relevance to the metaverse. Caching strategies are categorized into three groups: web caching, mobile caching, and Internet of Things (IoT) caching. Recent solu-tions are then examined to determine their relevance to the metaverse. Finally, the paper discusses open research challenges and potential future research directions in this domain. © 2024 by the authors of this article.

关键词： Virtual reality

来源：评论

学校读者我要写书评

暂无评论

Compound attention embedded dual channel encoder-decoder for ms lesion segmentation from brain MRI

引用

Multimedia Tools and Applications 2024年 1-33页

作者： Ghosal, Palash Roy, Abhijit Agarwal, Rohit Purkayastha, Kunal Sharma, Aaditya Lochan Kumar, Amish Department of Information Technology Sikkim Manipal Institute of Technology Sikkim Manipal University Majhitar Sikkim Rangpo737136 India Department of Computer Science and Engineering National Institute of Technology West Bengal Durgapur713209 India Department of Computer Science and Engineering Sikkim Manipal Institute of Technology Sikkim Manipal University Majhitar Sikkim Rangpo737136 India Computer Vision and Pattern Recognition Unit Indian Statistical Institute West Bengal Kolkata700108 India

Multiple Sclerosis (MS) lesions’ segmentation is difficult due to their variegated sizes, shapes, and intensity levels. Besides this, the class imbalance problem and the availability of limited annotated data samples obstruct the building of highly efficient deep learning-based models. Though researchers have made many attempts to design efficient deep learning-based models, the maximum Dice Coefficient achieved by their models is fairly below the acceptable level of 0.70. The possible reason may be due to the inability to capture sufficient local and global features of the lesions required for accurate segmentation. In this paper, we present a new deep-learning architecture based on compound attention for MS lesion segmentation from magnetic resonance images that handles the challenges of capturing the local and global variable features of the MS lesions. The proposed model is equipped with a dual-channel CNN encoder-decoder structure employing residual connections in one channel and residual channel and spatial attention in the other. The residual connections alleviate the vanishing gradient problem and pass the fine-grained information through the channels, which is crucial for pixel-wise prediction. The attention mechanism used in a channel helps to capture long-range dependencies. Thus, the complete model leverages rich global and local information through the two channels for lesion segmentation. The problem of data imbalance is handled by using the Focal Tversky loss function. Through rigorous evaluation using 3-fold cross-validation on the MICCAI 2016 challenge dataset, our model demonstrates superior performance, achieving a Dice Coefficient of 0.73, surpassing state-of-the-art models in both qualitative and quantitative assessments. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2024.

关键词：

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：