检索结果-内蒙古大学图书馆

Emsd-detr:efficient small object detection for UAV aerial images based on enhanced RT-DETR model

The Journal of Supercomputing 2025年第9期81卷 1-33页

作者： Zhang, Chaofeng Yang, Jingmin School of Computer Science Minnan Normal University Zhangzhou China Key Laboratory of Data Science and Intelligence Application Minnan Normal University Zhangzhou China Department of Electronic Engineering National Taipei University of Technology Taipei China

UAV aerial image object detection studies are affected by imbalances in large fields of view, changes in scale and viewing angle, object aggregation occlusion, and changes in lighting and weather conditions, resulting in poor performance of small object detection algorithms. In this paper, we propose an Efficient Multi-scale Drone DETR (EMSD-DETR), an object detection model for UAV aerial images that aims to overcome specific challenges. To mitigate the loss of information about small objects due to the increase in the number of layers of the convolutional neural networks, the FRE-Block is proposed. Furthermore, we propose a small object sensitive pyramid for cross-scale feature fusion that effectively fuses local spatial coordinate information with global contextual information to avoid missing details of different feature layers. Finally, the Focaler-Shape-IoU is introduced that incorporates consideration of the shape and scale of the bounding box to expedite convergence and enhance detection precision for difficult samples. Experiments on the VisDrone dataset demonstrate that the EMSD-DETR model improves 1.8%, 2.9%, 2.7%, and 1.2% with regard to precision, recall, mAP $$_{50}$$ , and mAP $$_{50:95}$$ , respectively, while decreasing the number of parameters by 8.2%, as compared to RT-DETR. In addition, generalizability experiments on the TinyPerson and NWPU VHR-10 dataset prove the validity of the EMSD-DETR model.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Skinned motion retargeting with dense geometric interaction perception 24

Skinned motion retargeting with dense geometric interaction ...

引用

Proceedings of the 38th International Conference on Neural Information Processing Systems

作者： Zijie Ye Jia-Wei Liu Jia Jia Shikun Sun Mike Zheng Shou Department of Computer Science and Technology BNRist Tsinghua University and Key Laboratory of Pervasive Computing Ministry of Education Show Lab National University of Singapore

ISBN: (纸本)9798331314385

Capturing and maintaining geometric interactions among different body parts is crucial for successful motion retargeting in skinned characters. Existing approaches often overlook body geometries or add a geometry correction stage after skeletal motion retargeting. This results in conflicts between skeleton interaction and geometry correction, leading to issues such as jittery, interpenetration, and contact mismatches. To address these challenges, we introduce a new retargeting framework, MeshRet, which directly models the dense geometric interactions in motion retargeting. Initially, we establish dense mesh correspondences between characters using semantically consistent sensors (SCS), effective across diverse mesh topologies. Subsequently, we develop a novel spatio-temporal representation called the dense mesh interaction (DMI) field. This field, a collection of interacting SCS feature vectors, skillfully captures both contact and non-contact interactions between body geometries. By aligning the DMI field during retargeting, MeshRet not only preserves motion semantics but also prevents self-interpenetration and ensures contact preservation. Extensive experiments on the public Mixamo dataset and our newly-collected ScanRet dataset demonstrate that MeshRet achieves state-of-the-art performance. Code available at https://***/abcyzj/MeshRet.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Analysis of Different Methods for Forecasting Transport Carbon Dioxide Emissions in the Philippines

引用

Transportation Research Procedia 2025年 82卷 3768-3789页

作者： Aaron Michael Salang Virgilio Ma. Ramos Arse John Salison John Justine Villar Intelligent Transportation Systems Laboratory National Center for Transportation Studies University of the Philippines Diliman Quezon City Philippines Department of Logistics and Information Engineering Tokyo University of Marine Science and Technology Tokyo Japan Scientific Computing Laboratory Department of Computer Science University of the Philippines Diliman Quezon City Philippines

Emission forecasts can be an important way of creating awareness among the public and decision-makers on solving environmental problems. The main goal of this study is to forecast and compare the transport CO2 emissions of the Philippines using four different forecasting models. We use the models of Holt-Winters Exponential Smoothing, Autoregressive Integrated Moving Average (ARIMA), Vector Autoregressive (VAR), and the Artificial Neural Network (ANN). The performance of the different forecasting methods was compared using the coefficient of determination (R2) and the root mean squared error (RMSE) values. Several economic variables from 1990 to 2019 and the transport Carbon Dioxide (CO2) emissions in the Philippines were utilized in this study. The result show that all four methods exhibit goodness of fit and accuracy results according to the statistical measures. In comparison, the multivariate methods (ANN & VAR) performed better than univariate methods (ARIMA & Holt-Winters).

关键词： Transport Emissions Forecasting Time-series modeling ARIMA ANN VAR Holt-Winters

来源：评论

学校读者我要写书评

暂无评论

A Comparative Study of Sentiment Analysis Methods for Detecting Fake Reviews in E-Commerce

HighTech and Innovation Journal

引用

HighTech and Innovation Journal 2023年第2期4卷 349-363页

作者： Puttarattanamanee, Maneerat Boongasame, Laor Thammarak, Karanrat KMITL-Digital Analytics and Intelligence Center Faculty of Science King Mongkut’s Institute of Technology Ladkrabang Bangkok10520 Thailand Department of Mathematics Faculty of Science King Mongkut’s Institute of Technology Ladkrabang Bangkok10520 Thailand Business Innovation and Investment Laboratory: B2I-Lab School of Science King Mongkut’s Institute of Technology Ladkrabang Bangkok10520 Thailand Department of Computer Engineering and Electronics School of Engineering and Technology Walailak University Nakhon Si Thammarat80160 Thailand

The popularity of the e-commerce system has increased, especially under the COVID scenario. Consumer product reviews from the past have had a significant impact on influencing consumers' purchasing decisions. Fake reviews—those written by humans and computers that engage in dishonest behavior—are consequently generated to increase product sales. The fake reviews hurt consumers and are dishonest. The goal of this research is to examine and evaluate the performance of various methods for identifying fake reviews. The well-known and widely-used Amazon Review Data (2018) dataset was used for this research. The first 10 product categories on *** with favorable feedback will be provided in the data section. After that, perform fundamental data preparation procedures such as special character trimming, bag of words, TF-IDF, etc. The models are trained to create a dataset for detecting fake reviews. This research compares the performance of four different models: GPT-2, NBSVM, BiLSTM, and RoBERTa. The hyperparameters of the models are also tuned to find the optimal values. The research concludes that the RoBERTa model performs the best overall, with an accuracy of 97%. GPT-2 has an overall accuracy of 82%, NBSVM has an overall accuracy of 95%, and BiLSTM has an overall accuracy of 92%. The research also calculates the Area Under the Curve (AUC) for each model and finds that RoBERTa has an AUC of 0.9976, NBSVM has an AUC of 0.9888, BiLSTM has an AUC of 0.9753, and GPT-2 has an AUC of 0.9226. It can be observed that the RoBERTa model has the highest AUC value, which is close to 1. Therefore, it can be concluded that this model provides the most accurate prediction for detecting fake reviews, which is the main focus of this research. © 2023, Ital Publication. All rights reserved..

关键词： Fake detection

来源：评论

学校读者我要写书评

暂无评论

Deep Multi-Task Models for Misogyny Identification and Categorization on Arabic Social Media 13

Deep Multi-Task Models for Misogyny Identification and Categ...

引用

Working Notes of FIRE - 13th Forum for Information Retrieval Evaluation, FIRE-WN 2021

作者： El Mahdaouy, Abdelkader El Mekki, Abdellah Oumar, Ahmed Mousannif, Hajar Berrada, Ismail School of Computer Sciences Mohammed VI Polytechnic University Morocco LISI Laboratory Computer Science Department FSSM Cadi Ayyad University Morocco

The prevalence of toxic content on social media platforms, such as hate speech, offensive language, and misogyny, presents serious challenges to our interconnected society. These challenging issues have attracted widespread attention in Natural Language Processing (NLP) community. In this paper, we present the submitted systems to the first Arabic Misogyny Identification shared task. We investigate three multi-task learning models as well as their single-task counterparts. In order to encode the input text, our models rely on the pre-trained MARBERT language model. The overall obtained results show that all our submitted models have achieved the best performances (top three ranked submissions) in both misogyny identification and categorization tasks. © 2021 Forum for Information Retrieval Evaluation, December 13-17, 2021, India.

关键词： Learning systems

来源：评论

学校读者我要写书评

暂无评论

Progress in research on ultrasound radiomics for predicting the prognosis of breast cancer

引用

Cancer Innovation 2023年第4期2卷 283-289页

作者： Xuantong Gong Xuefeng Liu Xiaozheng Xie Yong Wang Department of Ultrasound National Cancer Center/National Clinical Research Center for Cancer/Cancer HospitalChinese Academy of Medical Sciences and Peking Union Medical CollegeBijingChina State Key Laboratory of Virtual Reality Technology and Systems School of Computer Science and EngineeringBejjing Advanced Innovation Center for Big Data and Brain Computing(BDBC)Beihang UniversityBeijingChina School of Computer and Communication Engi neering.University of Science and Technology Beijing BeijingChina

Breast cancer is the most common malignant tumor and the leading cause of cancer-related deaths in women *** means of predicting the prognosis of breast cancer are very helpful in guiding treatment and improving patients'*** extracted by radiomics reflect the genetic and molecular characteristics of a tumor and are related to its biological behavior and the patient's ***,radiomics provides a new approach to noninvasive assessment of breast cancer *** is one of the commonest clinical means of examining breast *** recent years,some results of research into ultrasound radiomics for diagnosing breast cancer,predicting lymph node status,treatment response,recurrence and survival times,and other aspects,have been *** this article,we review the current research status and technical challenges of ultrasound radiomics for predicting breast cancer *** aim to provide a reference for radiomics researchers,promote the development of ultrasound radiomics,and advance its clinical application.

关键词： breast cancer deeplearning prognosisprediction radiomics ultrasound

来源：评论

学校读者我要写书评

暂无评论

Recent developments in application of single-cell RNA sequencing in the tumour immune microenvironment and cancer therapy

引用

Military Medical Research 2023年第3期10卷 383-402页

作者： Pei-Heng Li Xiang-Yu Kong Ya-Zhou He Yi Liu Xi Peng Zhi‑Hui Li Heng Xu Han Luo Jihwan Park Department of Thyroid and Parathyroid Surgery Laboratory of Thyroid and Parathyroid DiseaseFrontiers Science Centre for Disease‑Related Molecular NetworkDepartment of General SurgeryWest China HospitalSichuan UniversityChengdu 610044China Department of Oncology West China School of Public Health and West China Fourth HospitalSichuan UniversityChengdu 610044China Department of Rheumatology and Immunology Rare Diseases CentreWest China HospitalSichuan UniversityChengdu 610044China College of Computer Science Sichuan UniversityChengdu 610065China State Key Laboratory of Biotherapy and Cancer Centre West China HospitalSichuan University and Collaborative Innovation CentreChengdu 610044China School of Life Sciences Gwangju Instituteof Science and Technology(GIST)Gwangju 61005Republic of Korea

The advent of single-cell RNA sequencing(scRNA-seq)has provided insight into the tumour immune microenvironment(TIME).This review focuses on the application of scRNA-seq in investigation of the *** time,scRNA-seq methods have evolved,and components of the TIME have been deciphered with high *** this review,we first introduced the principle of scRNA-seq and compared different sequencing *** cell types in the TIME,a continuous transitional state,and mutual intercommunication among TIME components present potential targets for prognosis prediction and treatment in ***,we concluded novel cell clusters of cancerassociated fibroblasts(CAFs),T cells,tumour-associated macrophages(TAMs)and dendritic cells(DCs)discovered after the application of scRNA-seq in *** also proposed the development of TAMs and exhausted T cells,as well as the possible targets to interrupt the *** addition,the therapeutic interventions based on cellular interactions in TIME were also *** decades,quantification of the TIME components has been adopted in clinical practice to predict patient survival and response to therapy and is expected to play an important role in the precise treatment of *** the current findings,we believe that advances in technology and wide application of single-cell analysis can lead to the discovery of novel perspectives on cancer therapy,which can subsequently be implemented in the ***,we propose some future directions in the field of TIME studies that can be aided by scRNA-seq technology.

关键词： Single-cell RNA sequencing(scRNA-seq) Tumour immune microenvironment(TIME) Trajectory Cellular interactions Therapeutic targets

来源：评论

学校读者我要写书评

暂无评论

Multiuser Communication Aided by Movable Antenna 24

Multiuser Communication Aided by Movable Antenna

引用

24th IEEE International Workshop on Signal Processing Advances in Wireless Communications, SPAWC 2023

作者： Zhu, Lipeng Ma, Wenyan Ning, Boyu Zhang, Rui University of Electronic Science and Technology of China National Key Laboratory of Science and Technology on Communications 611731 China The Chinese University of HongKong Shenzhen Research Institute of Big Data Shenzhen518172 China National University of Singapore Department of Electrical and Computer Engineering Singapore117583 Singapore

ISBN: (纸本)9781665496261

This paper investigates the multiuser communication aided by movable antenna (MA) via antenna position optimization, where the uplink transmission from multiple user terminals (UTs) each equipped with a single MA to a base station (BS) with a fixed-position antenna (FPA) array is considered. Specifically, we formulate an optimization problem for minimizing the total transmit power of UTs, subject to a minimum-achievable-rate requirement for each UT, where the positions of MAs and the transmit powers of UTs, as well as the receive combining matrix at the BS are jointly optimized. To solve this non-convex optimization problem, we first express the combining matrix of the BS and the total transmit power of UTs as a function of the MAs' position vectors, which are then optimized by using the gradient descent method in an iterative manner. Simulation results demonstrate that compared to conventional FPA systems, the proposed MA-enhanced multiple access systems can significantly decrease the total transmit power of UTs by moving antennas within small regions of several-wavelength size. © 2023 IEEE.

关键词： Convex optimization

来源：评论

学校读者我要写书评

暂无评论

Smart-PKI: A Blockchain-based Distributed Identity Validation Scheme for IoT Devices

Smart-PKI: A Blockchain-based Distributed Identity Validatio...

引用

IEEE International Conference on Communications (ICC)

作者： Han Bao Xiaoping Zhang Gaoyuan Wang Renrui Tian Jinrong Duan Youjian Zhao Department of Computer Science and Technology BNRist Tsinghua University Beijing China Zhongguancun Laboratory Beijing China

Internet of Things (IoT) devices have achieved rapid development but most of them are vulnerable to spoofing attacks and spoofing-related attacks. It is crucial to verify source identity at the near-source end to defend against attacks, save network forwarding resources, and relieve the authentication pressure on the receiver end. In this paper, we propose Smart-PKI, a blockchain-based distributed identity validation scheme for IoT Devices. In the architecture of Smart-PKI, near-source forwarders can verify the authenticity of the source identity of packets and can filter spoofed packets. Besides, we apply Merkle Patricia Trie (MPT) to the Smart-PKI blockchain to enable lightweight blockchain copy storage and efficient retrieval and verification of identity information on forwarders. Meanwhile, Smart-PKI proposes an identity restoration mechanism and enables solutions for the attacks caused by public and private key compromise. Furthermore, we implement Smart-PKI on Network Simulator Version 3 (NS3) and evaluate its performance against reflection denial-of-service (DDoS) attacks. The simulation results demonstrate the effectiveness and efficiency of Smart-PKI and it outperforms existing blockchain-based PKI solutions for IoT devices in terms of network latency for verifying certificates.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Predictive model for daily risk alerts in sepsis patients in the ICU:visualization and clinical analysis of risk indicators

引用

Precision Clinical Medicine 2025年第1期8卷 28-39页

作者： Hao Yang Jiaxi Li Chi Zhang Alejandro Pazos Sierra Bairong Shen Information Center West China Hospital of Sichuan UniversityChengdu 610041China Department of Computer Science and Information Technologies Research Center for Information and Communications TechnologiesUniversity of A CoruñaBiomedical Research Institute of a CoruñaA Coruña 15071Spain Department of Clinical Laboratory Medicine Jinniu Maternity and Child Health Hospital of ChengduChengdu 610031China Joint Laboratory of Artificial Intelligence for Critical Care Medicine Department of Critical Care Medicine and Institutes for Systems GeneticsFrontiers Science Center for Disease-related Molecular NetworkWest China HospitalSichuan UniversityChengdu 610041China

This study introduces a novel Transformer-based time-series framework designed to revolutionize risk stratification in Intensive Care Units(ICUs)by predicting patient outcomes with high temporal *** sequential data from the eICU database,our two-stage architecture dynamically captures evolving health trajectories throughout a patient’s ICU stay,enabling real-time identification of high-risk individuals and actionable insights for personalized *** model demonstrated exceptional predictive power,achieving a progressive AUC increase from 0.87(±0.021)on admission day to 0.92(±0.009)by day 5,reflecting its capacity to assimilate longitudinal physiological *** external validation across geographically diverse cohorts—including an 81.8%accuracy on Chinese sepsis data(AUC=0.73)and 76.56%accuracy on MIMIC-IV-3.1(AUC=0.84)—confirmed robust ***,SHAP-derived temporal heatmaps unveiled mortality-associated feature dynamics over time,bridging the gap between model predictions and clinically interpretable *** findings establish a new paradigm for ICU prognostics,where data-driven temporal modeling synergizes with clinician expertise to optimize triage,reduce diagnostic latency,and ultimately improve survival outcomes in critical care.

关键词： sepsis Transformer time-series visualization predicting mortality

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：