检索结果-内蒙古大学图书馆

arXiv 2022年

作者： Cao, Jian Qian, Chen Huang, Yihui Chen, Dicheng Gao, Yuncheng Dong, Jiyang Guo, Di Qu, Xiaobo Department of Electronic Science Biomedical Intelligent Cloud R&D Center Fujian Provincial Key Laboratory of Plasma and Magnetic Resonance Xiamen University Xiamen361005 China School of Computer and Information Engineering Fujian Engineering Research Center for Medical Data Mining and Application Xiamen University of Technology Xiamen361024 China

Implicit regularization is an important way to interpret neural networks. Recent theory starts to explain implicit regularization with the model of deep matrix factorization (DMF) and analyze the trajectory of discrete gradient dynamics in the optimization process. These discrete gradient dynamics are relatively small but not infinitesimal, thus fitting well with the practical implementation of neural networks. Currently, discrete gradient dynamics analysis has been successfully applied to shallow networks but encounters the difficulty of complex computation for deep networks. In this work, we introduce another discrete gradient dynamics approach to explain implicit regularization, i.e. landscape analysis. It mainly focuses on gradient regions, such as saddle points and local minima. We theoretically establish the connection between saddle point escaping (SPE) stages and the matrix rank in DMF. We prove that, for a rank-R matrix reconstruction, DMF will converge to a second-order critical point after R stages of SPE. This conclusion is further experimentally verified on a low-rank matrix reconstruction problem. This work provides a new theory to analyze implicit regularization in deep learning. Copyright © 2022, The Authors. All rights reserved.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Improved hybrid query tree anti-collision algorithm 8th

Improved hybrid query tree anti-collision algorithm

引用

8th International Conference on computer engineering and Networks, CENet 2018

作者： Deng, Hongwei Li, Lang Liang, Xiaoman School of Computer Science and Technology Hengyang Normal University HengyangHunan421002 China School of Information Science and Engineering Central South University ChangshaHunan410083 China Hunan Provincial Key Laboratory of Intelligent Information Processing and Application Hengyang421002 China

ISBN: (纸本)9783030146795

An improved algorithm based on the hybrid query tree (HQT) algorithm is proposed in this work. Tags are categorized according to the combined information of the highest bit of collision and second-highest bit of collision. Then, these tags are used to decide the postponement of several delayed-response time slots, to diminish the collision probability. The performance analysis results show that this algorithm is superior to the query tree (QT) and HQT algorithms. It reduces the query time and system communication traffic and improves the tag identification efficiency. © Springer Nature Switzerland AG 2020.

关键词： Trees (mathematics)

来源：评论

学校读者我要写书评

暂无评论

Vision-Based UAV Self-Positioning in Low-Altitude Urban Environments

arXiv

引用

arXiv 2022年

作者： Dai, Ming Zheng, Enhui Feng, Zhenhua Zhuang, Jiedong Yang, Wankou Qi, Lei The School of Automation Southeast University Nanjing210096 China The Unmanned System Application Technology Research Institute China Jiliang University Hangzhou310018 China The School of Computer Science and Electronic Engineering University of Surrey GuildfordGU2 7XH United Kingdom Zhejiang University Hangzhou310063 China

Unmanned Aerial Vehicles (UAVs) rely on satellite systems for stable positioning. However, due to limited satellite coverage or communication disruptions, UAVs may lose signals from satellite-based positioning systems. In such situations, vision-based techniques can serve as an alternative, ensuring the self-positioning capability of UAVs. However, most of the existing datasets are developed for the geo-localization tasks of the objects identified by UAVs, rather than the self-positioning task of UAVs. Furthermore, the current UAV datasets use discrete sampling on synthetic data, such as Google Maps, thereby neglecting the crucial aspects of dense sampling and the uncertainties commonly experienced in real-world scenarios. To address these issues, this paper presents a new dataset, DenseUAV, which is the first publicly available dataset designed for the UAV self-positioning task. DenseUAV adopts dense sampling on UAV images obtained in low-altitude urban settings. In total, over 27K UAV-view and satellite-view images of 14 university campuses are collected and annotated, establishing a new benchmark. In terms of model development, we first verify the superiority of Transformers over CNNs in this task. Then, we incorporate metric learning into representation learning to enhance the discriminative capacity of the model and to lessen the modality discrepancy. Besides, to facilitate joint learning from both perspectives, we propose a mutually supervised learning approach. Last, we enhance the Recall@K metric and introduce a new measurement, SDM@K, to evaluate the performance of a trained model from both the retrieval and localization perspectives simultaneously. As a result, the proposed baseline method achieves a remarkable Recall@1 score of 83.05% and an SDM@1 score of 86.24% on DenseUAV. The dataset and code will be made publicly available on https://***/Dmmm1997/DenseUAV. © 2022, CC BY.

关键词： Satellites

来源：评论

学校读者我要写书评

暂无评论

Inferring Circadian Gene Regulatory Relationships from Gene Expression Data with a Hybrid Framework

SSRN

引用

SSRN 2022年

作者： Hu, Shuwen Jing, Yi Li, Tao Wang, You-Gan Liu, Zhenyu Gao, Jing Tian, Yu-Chu School of Computer Science Queensland University of Technology Brisbane4001 Australia Inner Mongolia Autonomous Region Key Laboratory of Big Data Research and Application for Agricultural and Animal Husbandry Hohhot010018 China School of Life Sciences Inner Mongolia Agricultural University Hohhot010018 China Institute for Learning Sciences & Teacher Education Brisbane Australia Australian Catholic University Brisbane4000 Australia School of Computer and Information Engineering Inner Mongolia Agriculture University Hohhot010018 China

The central biological clock controls countless aspects of mammalian physiology, such as sleep, metabolism, and immune system regulation. Gene regulatory relationships play an important role in understanding the mechanisms that underlie a variety of biological cellular processes. In principle, it is feasible to infer circadian gene regulatory relationships from time-series gene expression data. But the correlation-based inference of the relationships between genes does not give sufficient information about causation. Moreover, gene expression data usually have low-sampling observations and high dimensions, leading to difficulties in gene data analysis. In this paper, a new hybrid framework was proposed to infer circadian gene regulatory relationships from gene expression data of rats. For practical clustering computation of the high-dimensional problem, the framework hybridizes the fuzzy C-means clustering algorithm and dynamic time warping (DTW) distance. After a cluster of genes is identified for the target gene, Wilcoxon signed-rank test was conducted to obtain the number of significant genes in the specific cluster. Finally, a dynamic vector autoregressive method was applied to the selected significant gene expression profiles to identify a directed causal regulatory relationship from partial correlation. Circadian gene regulatory relationships are inferred from the gene expression data of rats based on the Aanat target gene. The results show that genes Pde10a, Atp7b, Prok2, Per1, Rhobtb3 and Dclk1 stand out, which have been known to be essential for the regulation of circadian behaviour. The potential relationships between genes Tspan15, Eprs, Eml5 and Fsbp with a circadian rhythm need further experimental research. © 2022, The Authors. All rights reserved.

关键词： Rats

来源：评论

学校读者我要写书评

暂无评论

ECC-NeRF: Anti-Aliasing Neural Radiance Fields with Elliptic Cone-Casting for Diverse Camera Models

引用

IEEE Transactions on Circuits and Systems for Video Technology 2025年 F页

作者： Qin, Haidong Yang, Tao Zhou, Xiaoshi Li, Dongdong Dai, Yanran Li, Jing Northwestern Polytechnical University School of Computer Science SAIIP National Engineering Laboratory for Integrated Aero-Space-Ground-Ocean Big Data Application Technology Xi'an710129 China Xidian University School of Telecommunications Engineering Xi'an710071 China

Anti-aliasing is a crucial research topic in computer graphics, which can significantly enhance the rendering quality of neural radiance fields (NeRF). Recent studies have introduced effective anti-aliasing NeRF methods, utilizing cone-casting to replace ray-casting and modeling the 3D observation area of pixels as circular cones. The cone-casting strategy has successfully reduced blurring and aliasing in novel view rendering. However, we have observed that the light cones are not standard circular cones because the camera projection model distorts them into elliptic cones of diverse sizes and shapes. This finding motivates us to model pixel light cones as anisotropic elliptic cones and propose an elliptic cone-casting-based anti-aliasing NeRF method called "ECC-NeRF». Specifically, we first derive the elliptic cone models for common pinhole, fisheye, and panoramic cameras based on their camera projection models. Then, we integrate the proposed elliptic cone-casting into two representative cone-casting-based anti-aliasing NeRF methods: Mip-NeRF and Zip-NeRF. Our experimental evaluations on multiple datasets demonstrate that our method can achieve more accurate multi-scale anisotropic representation and better novel view rendering quality with negligible additional computation cost. © 1991-2012 IEEE.

关键词： Anti-aliasing

来源：评论

学校读者我要写书评

暂无评论

Sliding-Time-Window and Event-Trigger Based Data Collection Strategy for Non-intrusive Load Monitoring

Sliding-Time-Window and Event-Trigger Based Data Collection ...

引用

IEEE Conference on Energy Internet and Energy System Integration (EI2)

作者： Jian Yang Wenjia Hua Mengfei Xu Zhiyou Ouyang Han Zhang School of Computer NUPT-JSTZ Carbon Peaking & Carbon Neutrality Application Technology R&D Institute Nanjing University of Posts and Telecommunications Nanjing China Institute of Advanced Technology for Carbon Neutrality Nanjing University of Posts and Telecommunications Nanjing China School of Cyber Science and Engineering Southeast University and ZTE Corporation Nanjing China

Non-intrusive load monitoring (NILM) technology can effectively identify the type and operation state of the electrical equipments in the load, so as to help the better analyze the power consumption state and power consumption of users, and achieve better energy saving and emission reduction and load optimization control. However, the large amount of data collected by non-invasive load monitoring terminals, especially high-frequency data, brings challenges to data acquisition, transmission and training of non-intrusive load monitoring models. Hence, in this paper, a sliding-time-window and event trigger based non-intrusive load monitoring data acquisition strategy is proposed, which can select power equipment of state change events through the event trigger mechanism, and fix the length of collect data samples and ensure not lose equipment state change process of context information by using sliding-time-window. Experiments by using widely used non-intrusive load monitoring algorithm evaluation toolkit that named NLIMTK, and widely used public dataset which named REDD, show that the proposed data collection strategy can not only make non-intrusive load monitoring data collected by the size of the largest decreased by 97.07%, and the prediction performance has improved significantly, which fully demonstrate the effectiveness of the proposed method.

关键词： Training Load monitoring Power demand Heuristic algorithms Data acquisition System integration Data collection

来源：评论

学校读者我要写书评

暂无评论

More than Encoder: Introducing Transformer Decoder to Upsample

More than Encoder: Introducing Transformer Decoder to Upsamp...

引用

IEEE International Conference on Bioinformatics and Biomedicine (BIBM)

作者： Yijiang Li Wentian Cai Ying Gao Chengming Li Xiping Hu School of Computer Science and Engineering South China University of Technology Guangzhou China Guangdong Provincial Key Laboratory of Artificial Intelligence in Medical Image Analysis and Application Guangdong Provincial People’s Hospital Guangdong Academy of Medical Sciences Guangzhou China School of Intelligent Systems Engineering Sun Yat-sen University Shenzhen China School of Medical Technology Beijing Institute of Technology Beijing China

ISBN: (数字)9781665468190

ISBN: (纸本)9781665468206

Medical image segmentation methods downsample images for feature extraction and then upsample them to restore resolution for pixel-level predictions. In such schema, upsample technique is vital in restoring information for better performance. However, existing upsample techniques leverage little information from downsampling paths. The local and detailed feature from the shallower layer such as boundary and tissue texture is crucial in segmentation, especially medical image segmentation. To this end, we propose a novel upsample approach for medical image segmentation, Window Attention Upsample (WAU), which upsamples features conditioned on local and detailed features from downsampling path in local windows by introducing attention decoders of Transformer. WAU could serve as a general upsample method and be incorporated into any segmentation model that possesses lateral connections. We first propose the Attention Upsample which consists of Attention Decoder (AD) and bilinear upsample. AD leverages pixel-level attention to model longrange dependency and global information for a better upsample. Bilinear upsample is introduced as the residual connection to complement the upsampled features. Moreover, considering the extensive memory and computation cost of pixel-level attention, we further design a window attention scheme to restrict attention computation in local windows instead of the global range. We evaluate our method (WAU) on classic UNet structure with lateral connections and achieve state-of-the-art performance on Medical Segmentation Decathlon (MSD) Brain and Automatic Cardiac Diagnosis Challenge (ACDC) datasets. We also validate the effectiveness of our method on multiple classic architectures and achieve consistent improvement.

关键词： Image segmentation Visualization Image resolution Shape Semantics computer architecture Transformers

来源：评论

学校读者我要写书评

暂无评论

Tissue Segmentation of Thick-Slice Fetal Brain MR Scans with Guidance from High-Quality Isotropic Volumes

arXiv

引用

arXiv 2023年

作者： Huang, Shijie Zhang, Xukun Cui, Zhiming Zhang, He Chen, Geng Shen, Dinggang School of Biomedical Engineering ShanghaiTech University Shanghai China Academy for Eng. & Tech. Fudan University Shanghai China Department of Radiology Obstetrics and Gynecology Hospital of Fudan University Shanghai China National Engineering Laboratory for Integrated Aero-Space-Ground-Ocean Big Data Application Technology School of Computer Science and Engineering Northwestern Polytechnical University Xi’an China School of Biomedical Engineering ShanghaiTech University Shanghai201210 China Shanghai United Imaging Intelligence company Ltd. Shanghai200030 China Shanghai Clinical Research Trial Center Shanghai201210 China

Accurate tissue segmentation of thick-slice fetal brain magnetic resonance (MR) scans is crucial for both reconstruction of isotropic brain MR volumes and the quantification of fetal brain development. However, this task is challenging due to the use of thick-slice scans in clinically-acquired fetal brain data. To address this issue, we propose to leverage high-quality isotropic fetal brain MR volumes (and also their corresponding annotations) as guidance for segmentation of thick-slice scans. Due to existence of significant domain gap between high-quality isotropic volume (i.e., source data) and thick-slice scans (i.e., target data), we employ a domain adaptation technique to achieve the associated knowledge transfer (from high-quality volumes to thick-slice scans). Specifically, we first register the available high-quality isotropic fetal brain MR volumes across different gestational weeks to construct longitudinally-complete source data. To capture domain-invariant information, we then perform Fourier decomposition to extract image content and style codes. Finally, we propose a novel Cycle-Consistent Domain Adaptation Network (C2DA-Net) to efficiently transfer the knowledge learned from high-quality isotropic volumes for accurate tissue segmentation of thick-slice scans. Our C2DA-Net can fully utilize a small set of annotated isotropic volumes to guide tissue segmentation on unannotated thick-slice scans. Extensive experiments on a large-scale dataset of 372 clinically acquired thick-slice MR scans demonstrate that our C2DA-Net achieves much better performance than cutting-edge methods quantitatively and qualitatively. Copyright © 2023, The Authors. All rights reserved.

关键词： Magnetic resonance

来源：评论

学校读者我要写书评

暂无评论

Splicing Image Forgery Detection by Deploying Deep Learning Model

Splicing Image Forgery Detection by Deploying Deep Learning ...

引用

Automation, Computing and Renewable Systems (ICACRS), International Conference on

作者： N Krishnamoorthy C. Amuthadevi M. K. Geedtha Poli Lokeshwara Reddy Anitha Rani K S R. Gopinathan Department of Masters in Computer Application Ramapuram Campus SRM Institute of Science and Technology Chennai Tamil Nadu India Department of Computational Intelligence SRM Institute of Science and Technology Kattankulathur Tamil Nadu India Department of Information Technology Vel Tech Multi Tech Dr. Rangarajan Dr. Sakunthala Engineering College Chennai Tamil Nadu India Department of Electronics and Communication Engineering Anurag University Hyderabad Telangana India School of CSA REVA University Bangalore Karnataka India Department of Mechatronics Engineering Sri Krishna College of Engineering and Technology Coimbatore Tamil Nadu India

ISBN: (纸本)9781665460859

The use of digital images has become more important in numerous industries, including journalism, medical diagnosis, law enforcement, and forensics. Images could be easily modified due to the availability of photo-altering tools and software, which can distort the original information. The availability of low-cost and simple image-editing software has also drastically decreased the time and money needed to engage in image tampering. This destroys the purity of the image, making it vulnerable to misuse by anyone. Copy-move, splicing, and enhancement falsification are prominent methods used to create fake images. Using a combination of two or more images, a new image is created called splicing, which may be shared across many online channels and used to spread information and impact target audiences. The discovery could have beneficial as well as negative results. Consequently, it is important to provide a method that can identify the splicing forgery in an image. The study used a Deep Learning (DL) model named MobileNet to distinguish between authentic and manipulated images. Both the training and testing of the DL model make use of the CASIA data. Resizing and enhancing methods are used to process the collected data. Accuracy, F1 score, and recall are employed to test the reliability of the created DL model.

关键词： Deep learning Training Renewable energy sources Splicing Computational modeling Forgery Data models

来源：评论

学校读者我要写书评

暂无评论

Study on the Influence of Interaction Style on Learners

Study on the Influence of Interaction Style on Learners

引用

IEEE International Conference on Teaching, Assessment and Learning for engineering (TALE)

作者： Yi Chen Yidan Wang Jin Zhou Jing Gao Ge Gao Wei Zhang School of Computer Science Central China Normal University Wuhan P.R.China School of Computer Science Wuhan University Wuhan P.R.China National Engineering Laboratory of Educational Big Data Application Technology Central China Normal University Wuhan P.R.China

ISBN: (纸本)9781665436885

High-quality mobile learning resources are an important support for mobile learning. Interactivity is an important feature of mobile learning, which directly affects the effect of the interaction between learners and learning resources, and is a key indicator for evaluating the quality of learning resources. This article uses brain-computer interface technology to explore the changes in learning attention and learning effectiveness of mobile learners in different presentation forms and interactive methods of learning resources, including plain text, video and interactive methods represented by dubbing. Research shows that: First, different forms of knowledge presentation and interaction have a significant impact on attention. Second, in the interactive mode represented by dubbing, the learner's attention is higher and the learning effect is the best. Therefore, when designing mobile learning resources in the future, the inter activity of resources should be considered to maintain learners' higher learning attention and achieve better mobile learning effects.

关键词： Learning systems Education Psychology Media Streaming media Mobile learning Brain-computer interfaces

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：