检索结果-内蒙古大学图书馆

23rd IEEE International Conference on High Performance Computing and Communications, 7th IEEE International Conference on Data Science and Systems, 19th IEEE International Conference on Smart City and 7th IEEE International Conference on Dependability in Sensor, Cloud and Big Data Systems and Applications, HPCC-DSS-SmartCity-DependSys 2021

作者： Liu, Siping Li, Renfa Tu, Xiaohan Xie, Guoqi Xu, Cheng College of Computer Science and Electronic Engineering Hunan University Changsha410082 China Railway Police College Department of Image and Network Investigation Zhengzhou450053 China

ISBN: (纸本)9781665494571

Monocular depth estimation (MDE) has been a basic task in many fields. Currently, researchers usually improve the accuracy of MDE by designing deep neural networks, including complex branches or other components. Although complex networks achieve increased precision, they bring slow speed and high memory access cost (MAC) in embedded systems. To balance the speed, accuracy, and MAC for embedded systems, we propose a directly connected model (DCM), which relies on existing networks, for fast inference in embedded systems. DCM balances the speed, precision, and MAC with structural reparameterization that decouples its training and inference networks. To further reduce the MAC and latency of DCM for embedded systems, we construct a network optimization frame-work based on recent template-based optimization methods. In the framework, we respectively adopt the machine learning LightGBM and contextual simulated annealing algorithm as the cost model and optimizer to complete rapid optimization in embedded systems. Our comprehensive experiments confirm that our methods achieve a better trade-off among precision, latency, and MAC. The accuracy of our original DCM inference networks reaches at least 6.13% better than the state of the arts. The optimized DCM inference networks decrease latency, main MAC, and power consumption by 46.49%, 90.97%, and 65.31% on Jetson TX2 GPU, without unchanged accuracy. © 2021 IEEE.

关键词： Training Embedded systems Costs Power demand Smart cities Graphics processing units Estimation

来源：评论

学校读者我要写书评

暂无评论

A Dual-View Fusion network for Automatic Spinal Keypoint Detection in Biplane X-ray Images

A Dual-View Fusion Network for Automatic Spinal Keypoint Det...

引用

2023 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2023

作者： Zhou, Dandan Guo, Lijun Zhang, Rong He, Xiuchao Wang, Qiang Wang, Jianhua Ningbo University The Faculty of Electrical Engineering and Computer Science Ningbo China The Zhejiang Key Laboratory of Mobile Network Application Technology Ningbo China The Affiliated Hospital of Medicine School of Ningbo University The Department of Radiology Ningbo China

ISBN: (纸本)9798350337488

Accurate keypoint detection in medical images of the spine is critical for the assessment, diagnosis, treatment planning, and clinical investigation of spinal deformities. However, due to severe occlusions of spinal structures in lateral X-ray images, accurate keypoint detection can be hardly achieved in lateral X-ray images based on single-view information. Thus, methods based on both the anterior-posterior (AP) and lateral (LAT) X-ray image views have been proposed to alleviate occlusion problems and achieve better keypoint detection performance. Although some progress has been made with these dual-view methods, they do not effectively exploit a priori knowledge of the spine and hence cannot adequately account for the structural correlation of the vertebrae across views. In this paper, a new dual-view fusion network (DVFNet) framework is proposed for keypoint detection in spinal X-ray images. This framework obtains structural correlations between AP and LAT views of the spine based on a priori spine knowledge represented by high-level semantic features. Meanwhile, the proposed framework combines local and global features extracted respectively by a local subnetwork and a global subnetwork. On the one hand, the local subnetwork is constructed as an enhanced codec structure based on both the AP and LAT views. This subnetwork is trained to output local features that contain both joint semantic features of the two views and independent fine-grained features of each individual view. This scheme leads to accurate keypoint estimation locally. On the other hand, the global subnetwork utilizes a self-attention mechanism to extract view-specific global features based on either the AP view or the LAT view in order to eliminate ambiguity, and reduce confusion on keypoint locations. Further, we propose a weighted feature fusion (WFF) module for adaptive fusion of the local and global features. We evaluated the DVFNet model on a private dataset and found that our proposed metho

关键词： biplane X-ray images dual-view feature fusion spinal keypoint detection

来源：评论

学校读者我要写书评

暂无评论

IIESC-Net: Incorporating Implicit and Explicit Structural Constraints for Hip Joint Landmark Detection in pelvic X-ray

IIESC-Net: Incorporating Implicit and Explicit Structural Co...

引用

2023 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2023

作者： Zhang, Jing Guo, Lijun Ni, Lixin He, Xiuchao Wang, Jianhua Zhang, Rong Ningbo University The Faculty of Electrical Engineering and Computer Science Ningbo China The Zhejiang Key Laboratory of Mobile Network Application Technology Ningbo China The Affiliated Hospital of Medicine School of Ningbo University The Department of Radiology Ningbo China

ISBN: (纸本)9798350337488

The hip joint, a vital weight-bearing joint in the human body, is susceptible to various hip-related diseases. The accurate identification of anatomical landmarks in the hip joint is essential for both disease diagnosis and surgical planning. However, these landmarks are often inconspicuous in X-ray images, where irrelevant background interference increases detection difficulty. In this study, we proposed an IIESC-Net model that combined local and global features, enabling a hierarchical understanding of the implicit structural characteristics of the hip joint. Additionally, drawing from domain expertise based on the explicit physiological structure of the hip joint, we designed a Hip Morphology-Aware loss function to constrain large landmark errors through the application of high-confidence landmarks with robust distinctive identification, thereby achieving accuracy in automatic detection. Furthermore, we constructed a dataset comprising of 843 pelvic X-ray images. The experimental results demonstrated a substantial enhancement in hip joint landmark detection accuracy attributed to the proposed IIESC-Net. This innovation established state-of-the-art performance, notably excelling in attaining heightened successful detection rates under stringent error tolerance. This achievement has profound practical implications for clinical applications. © 2023 IEEE.

关键词： Global Features Hip joint Hip Morphology-Aware Loss Landmark Detection Local Features

来源：评论

学校读者我要写书评

暂无评论

Traffic Flow Analysis in Digital Forensics: Unveiling Patterns and Anomalies

Traffic Flow Analysis in Digital Forensics: Unveiling Patter...

引用

International Conference on Computation System and Information Technology for Sustainable Solutions (CSITSS)

作者： Malik Zaib Alam Haneef Khan Adeel Ahmad Sarfaraz Ahamd Mohammad Rafeek Khan Shams Tabrez Siddiqui Department of Computer and Network Engineering College of CS & IT Jazan University KSA Department of Computer Science College of CS & IT Jazan University KSA

The rapid advancement of digital technology and the increasing interconnection of systems has resulted in a complex network of digital activities. In this scene, examining organization traffic stream has arisen as a vital device in computerized criminology. This paper dives into the “network Traffic Flow Analysis in the Field of Digital Forensics and Cybersecurity: Revealing Patterns and Deviations” domain. The objective of the study is to investigate the way in which traffic stream examination fills in as a strong procedure for unraveling the multifaceted computerized effects left behind by clients, gadgets, and potential danger entertainers. Forensic Data analysts acquire the ability to detect security breaches, data exfiltration, or unauthorized access by closely examining examples of information transmission, source-objective connections, and correspondence conventions within an organization. This analysis allows them to identify both regular patterns and irregular deviations. Methodologies for capturing and recording network packets, the role of flow data in summarizing complex interactions, and the use of intrusion detection systems (IDS) and intrusion prevention systems (IPS) to monitor real-time network activities are all examined in this paper. Also, the review dives into utilizing AI and man-made reasoning methods for prescient investigation and inconsistency identification inside network traffic.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Laryngeal Leukoplakia Classification Via Dense Multiscale Feature Extraction in White Light Endoscopy Images 48

Laryngeal Leukoplakia Classification Via Dense Multiscale Fe...

引用

48th IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2023

作者： You, Zhenzhen Yan, Yan Shi, Zhenghao Zhao, Minghua Yan, Jing Liu, Haiqin Hei, Xinhong Ren, Xiaoyong Xi'An University of Technology Shaanxi Key Laboratory for Network Computing and Security Technology School of Computer Science and Engineering Xi'an China Xi'An Jiaotong University Second Affiliated Hospital of Medical College Department of Otorhinolaryngology Xi'an China

ISBN: (纸本)9781728163277

Laryngeal leukoplakia classification is challenging using white light endoscopy images. Relevant research focus on normal tissues versus non normal tissues, cancer versus non cancer classification. The objective of this paper is to classify laryngeal leukoplakia in white light endoscopy images into six classes: normal tissues, inflammatory keratosis, mild dysplasia, moderate dysplasia, severe dysplasia and squamous cell carcinoma. We proposed a dense multiscale convolutional neural network including parallel multiscale convolution, dense convolution and recurrent convolution in favor of extracting dense multiscale features of laryngeal leukoplakia for fine classification. The proposed network achieved an overall accuracy of 0.8958 for the six-class classification. It has high sensitivity and specificity for each class which are, respectively, 1.0000 and 0.9394 for normal tissues, 0.6667 and 1.0000 for inflammatory keratosis, 0.8889 and 0.9744 for mild dysplasia and moderate dysplasia, 0.7500 and 1.0000 for severe dysplasia, 1.0000 and 0.9767 for squamous cell carcinoma. The experimental results show that our proposed model is superior to the state-of-the-art deep learning-based models. © 2023 IEEE.

关键词： Endoscopy

来源：评论

学校读者我要写书评

暂无评论

A multimodal emotion recognition model based on deep neural network with cross-media data feature fusion

A multimodal emotion recognition model based on deep neural ...

引用

2021 International Conference on Electronic Information engineering and computer Communication, EIECC 2021

作者： Liu, Sheng-Tong Chen, Zhi-Hong Li, Peng-Hai Lu, Jia Shao, Xin-Wang Gao, Xiao-Qi Zhang, Hai-Wei Yang, Xiao-Ping Department of Optoelectronic Information Science and Engineering Institute of Integrated Circuit Science and Engineering Tianjin University of Technology Tianjin China Institute of Brain-computer Interaction and Industrial Intelligent Network Tianjin University of Technology Tianjin China

ISBN: (数字)9781510653207

ISBN: (纸本)9781510653191

Affective computing is an interdisciplinary research area that includes machine learning and pattern recognition, psychology, and cognitive science. The aim is to research and develop theories, methods and systems that can recognize, interpret, process and simulate human emotions. In this article we propose a neural network model for multimodal emotion recognition based on cross-media data-feature fusion. Multimodal data fusion can effectively improve the accuracy of emotion recognition. We extract features from EEG data and facial images using a deep double-stream neural network and then merge them in a medium-term feature layer to identify three categories of emotions (sadness, calm, and happiness). The experimental results show that the detection accuracy can reach over 95%. Compared to the traditional single-modal emotion recognition method, the accuracy rate of emotion recognition based on EEG data and facial images has been significantly improved. It also proves that the multimodal medium-term feature layer fusion method has good applicability for emotion recognition. © COPYRIGHT SPIE. Downloading of the abstract is permitted for personal use only.

关键词： Emotion Recognition

来源：评论

学校读者我要写书评

暂无评论

Developing Real-World Applications with Algorithmic Intelligence in Reinforcement Learning for Robotics

Developing Real-World Applications with Algorithmic Intellig...

引用

Trends in Quantum Computing and Emerging Business Technologies (TQCEBT), International Conference on

作者： M. Bhanumathi M. Sangeetha F. Mary Harin Fernandez A. Ganesh Ramachandran S.K. Saravanan Department of Computer Science and Engineering Vels Institute of Science Technology and Advanced Studies (VISTAS) Chennai India Department of Computer Engineering University of Technology and Applied Sciences Nizwa Sultanate of Oman Department of Computer Science and Engineering Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Saveetha University Chennai India Department of Computer Science Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Saveetha University Chennai India Department of Computers Science and Engineering BigData and Network Security Saveetha School of Engineering SIMATS Chennai Tamil Nadu

ISBN: (数字)9798350384277

ISBN: (纸本)9798350384284

In the study, three leading reinforcement learning algorithms Deep Q-network (DQN), Proximal Policy Optimization (PPO) and Asynchronous Advantage Actor – Critic (A3Cc are applied in a number of real-world robotics tasks. The focus of our analysis is on performance indicators, computational efficiencies across multiple environments as well with respect to customer satisfaction rating. The DQN demonstrated good performance, with the success rate of up to 85% in object manipulation tasks and confirmed its reliability under controlled lab conditions where it made only a 5 percent error. PPO excelled in autonomous navigation (82%) and was the most energy-efficient option, requiring just 10 hours for training and using only twelve KWH of power. Compared with A3C, the method performed to a higher level of adaptability and robustness, especially in object manipulation (90% success rate) and dynamic environments (10% error rates applicable for outdoor variable weather conditions). But A3C was also the most automatable, which required 15 hours to train and consumed an astonishing amount of electricity – 18 kWh. Regarding customer satisfaction, PPO performs the best in end-user rating on user interface; A3C shows a preference with engineers due to its advanced features and flexibility. The focus of this study is significant trade-offs between performance, efficiency and user preference in real robotics applications that advocate for individual tailoring selections as well as deployments based on task specifics nature.

关键词： Training Error analysis Heuristic algorithms Customer satisfaction Reinforcement learning Robustness Computational efficiency

来源：评论

学校读者我要写书评

暂无评论

Transformer-based Image and Video Inpainting: Current Challenges and Future Directions

arXiv

引用

arXiv 2024年

作者： Elharrouss, Omar Damseh, Rafat Belkacem, Abdelkader Nasreddine Badidi, Elarbi Lakas, Abderrahmane Department of Computer Science and Software Engineering College of Information Technology United Arab Emirates University United Arab Emirates Department of Computer and Network Engineering College of Information Technology United Arab Emirates University United Arab Emirates

Image inpainting is currently a hot topic within the field of computer vision. It offers a viable solution for various applications, including photographic restoration, video editing, and medical imaging. Deep learning advancements, notably convolutional neural networks (CNNs) and generative adversarial networks (GANs), have significantly enhanced the inpainting task with an improved capability to fill missing or damaged regions in an image or video through the incorporation of contextually appropriate details. These advancements have improved other aspects, including efficiency, information preservation, and achieving both realistic textures and structures. Recently, visual transformers have been exploited and offer some improvements to image or video inpainting. The advent of transformer-based architectures, which were initially designed for natural language processing, has also been integrated into computer vision tasks. These methods utilize self-attention mechanisms that excel in capturing long-range dependencies within data;therefore, they are particularly effective for tasks requiring a comprehensive understanding of the global context of an image or video. In this paper, we provide a comprehensive review of the current image or video inpainting approaches, with a specific focus on transformer-based techniques, with the goal to highlight the significant improvements and provide a guideline for new researchers in the field of image or video inpainting using visual transformers. We categorized the transformer-based techniques by their architectural configurations, types of damage, and performance metrics. Furthermore, we present an organized synthesis of the current challenges, and suggest directions for future research in the field of image or video inpainting. © 2024, CC BY-NC-SA.

关键词： computer vision

来源：评论

学校读者我要写书评

暂无评论

Algorithmic Innovations in Machine Learning for Drug Discovery and Pharmaceutical Innovations

Algorithmic Innovations in Machine Learning for Drug Discove...

引用

Trends in Quantum Computing and Emerging Business Technologies (TQCEBT), International Conference on

作者： M. Sangeetha F. Mary Harin Fernandez A. Ganesh Ramachandran S. K. Saravanan M. Bhanumathi Department of Computer Engineering University of Technology and Applied Sciences Nizwa Sultanate of Oman Department of Computer Science and Engineering Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Saveetha University Chennai India Department of Computer Science Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Saveetha University Chennai India Department of Computers Science and Engineering BigData and Network Security Saveetha School of Engineering SIMATS Chennai Tamil Nadu Department of Computer Science and Engineering Vels Institute of Science Technology and Advanced Studies (VISTAS) Chennai India

ISBN: (数字)9798350384277

ISBN: (纸本)9798350384284

Algorithmic advancements in machine learning have transformed drug discovery and pharmaceutical innovation. This study examines how machine learning might speed medication development and pharmaceutical innovation. Data from genomes, chemistry, and clinical trials is collected to start the research. Careful data cleansing, normalization, and feature engineering prepare the data for model building. Random Forest, SVM, NN, Decision Tree, and Logistic Regression are tested for prediction. ROC AUC results show that Random Forest is the most accurate model at 86%. Ensemble approaches have drug discovery potential. Beyond accuracy, models are assessed for precision, false positive, and false negative rates. Drug discovery relies on precision to measure positive predictions. The Random Forest model makes the most accurate positive predictions at 0.87. However, false positives and negatives also matter in drug discovery. The Decision Tree model contains the falsest positives (35), underlining the need to reduce them to save resources. The Neural network approach reduces false negatives to 12, ensuring good medication candidates are not overlooked. The results show that different machine learning models have trade-offs, underscoring the necessity for a personalized strategy depending on objectives and restrictions. Algorithmic breakthroughs in pharmaceutical innovation reduce medication development time by 40% and cost by 25%, according to the study.

关键词： Support vector machines Drugs Technological innovation Logistic regression Machine learning algorithms Genomics Prediction algorithms

来源：评论

学校读者我要写书评

暂无评论

Learning Features Using an optimized Artificial Neural network for Breast Cancer Diagnosis

引用

SN computer Science 2022年第3期3卷 229页

作者： AlShourbaji, Ibrahim Kachare, Pramod Zogaan, Waleed Muhammad, L.J. Abualigah, Laith Department of Computer and Network Engineering Jazan University Jazan 45142 Saudi Arabia Department of Electronics and Telecommunication Engineering Ramrao Adik Institute of Technology Nerul Maharashtra Navi Mumbai 400706 India Department of Computer Science Jazan University Jazan 45142 Saudi Arabia Department of Computer Science Faculty of Science Federal University of Kashere P.M.B. 0182 Gombe Nigeria Faculty of Computer Sciences and Informatics Amman Arab University Amman 11953 Jordan

Breast cancer (BC) has been one of the significant causes of death worldwide, and its early detection can play a vital role in increasing the survival rate of this disease. This paper suggests a novel feature learning method for BC diagnosis using the artificial neural network (ANN) by optimizing the hidden layers. It uses a systematic hyper-parameter search method to optimize number of hidden layers, neurons in the hidden layers, activation functions, learning rate, batch size, and training epochs of the ANN. The number of hidden layers and the number of neurons in the hidden layer are investigated using t-distributed stochastic neighbor embedding (t-SNE) technique to specify the best hyper-parameters that achieved the highest testing accuracy while avoiding model overfitting. Comparative analysis showed that the proposed method achieved higher performance than the ANN-based models that utilizes particle swarm optimization (PSO), multi-verse optimizer (MVO), BAT algorithm, and firefly algorithm (FFA) as feature selection (FS) methods, with the best classification accuracy of 0.9948, sensitivity of 0.9815, and specificity of 0.9882 for 10 iterations of holdout cross-validation. The performance of the proposed method is also comparable with several earlier reported state-of-the-art models in the literature. © 2022, The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd.

关键词： ANN Brest cancer Classification Feature selection Machine learning

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：