检索结果-内蒙古大学图书馆

International Symposium on Advanced Networks and Telecommunication Systems (ANTS)

作者： Ishtiaque Ahmed Toke Mahir Mohammad Faisal Rahman Md Fahad Monir Tarem Ahmed Khondkar Ayaz Rabbani Department of Computer Science and Engineering Independent University Bangladesh Department of Environmental Science and Management Independent University Bangladesh

Current smart greenhouse systems face challenges with regard to energy efficiency, cost-effectiveness, and the design of user-friendly interfaces. Wide-scale adoption of these systems requires simple and accessible solutions. These issues highlight the need for more straightforward simple and accessible solutions. Our study introduces an automated, user-friendly, affordable and power-optimized greenhouse management system using open-source technologies to tackle current efficiency and accessibility challenges. The system, powered by a solar-charged battery, uses a Node-RED designed interface, Raspberry Pi 3B for data parsing and storage, and an Arduino 2560 for micro-processing. With a custom automation algorithm and a remote control feature, the system provides real-time data visualization and enhanced usability. Particularly suited to Bangladesh's agricultural needs, it addresses local conditions such as climate and rural electricity scarcity.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Predicting Intracardiac Thrombus Formation in the Left Atrial Appendage Using Machine Learning and CT Images

Predicting Intracardiac Thrombus Formation in the Left Atria...

引用

Annual International Conference of the ieee Engineering in Medicine and Biology Society (EMBC)

作者： Shota Nako Chanjin Seo Shota Tsutsui Tadashi Yamamoto Kiyotaka Iwasaki Takaaki Ohkawauchi Jun Ohya Dept. of Modern Mechanical Engineering Waseda University Tokyo Japan Waseda University Tokyo Japan Humanities and Science Department Nihon University Tokyo Japan

This paper proposes a comprehensive method for estimating thrombus formation factors in the left atrial appendage (LAA). First, using 3D CT (Computer Tomography) image data as input, classification of thrombus presence/absence is learned using 3D ResNet. Besides, 3D Grad-CAM is applied to the prediction results to visualize regions of interest in thrombus formation. Second, features are extracted based on the visualization of regions of interest. Using the extracted features and numerical data obtained from the hospital as input, a regression analysis is performed to predict the presence/absence of thrombus using LightGBM. visualization of regions of interest using 3D ResNet and 3D Grad-CAM shows that the right inferior pulmonary vein and the LAA were particularly correlated with thrombus formation. Estimation of important factors for thrombus formation using LightGBM shows that the LAA ostium area has the greatest influence on thrombus *** Relevance—This paper shows the factors that contribute to thrombus formation in the LAA from the viewpoint of three-dimensional structure. In addition, the features considered important in thrombus formation were identified by comparing a variety of features.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Improvement of semantic segmentation algorithm based on ResNeSt network

Improvement of semantic segmentation algorithm based on ResN...

引用

Control, Electronics and Computer Technology (ICCECT), ieee International Conference on

作者： Jia Mian Wan Yaping Yan Wei Xie Jianbin Zhong Hua Wang Yue School of Computer University of South China Hengyang China School of Electronic Science National University of Defense Technology Changsha China School of Data Science and Technology North University of China Taiyuan China

This paper proposes an improved semantic segmentation algorithm based on the split attention network resnest. The original icnet semantic segmentation model has some defects in the segmentation of foreground objects. In view of these problems, we optimized the icnet model with resnest network and CBAM convolution attention module. On the one hand, we replace the backbone network in the original model with resnest network, which enhances the feature extraction capability of the model. On the other hand, we introduce the CBAM convolution attention module to improve the segmentation ability of the model to the target in the foreground region. Without affecting the inference rate of the original model, the training effect of the improved idea proposed in this paper on the cityscapes open data set is improved separately from 71.0% to 71.3% and 71.9%. Respectively compared with the MIoU of the baseline model, the MIoU of the icnet model combined with the resnest backbone network and the CBAM convolution attention module on the cityscapes open data set is increased from 71.0% to 72.1%, and the segmentation effect of the foreground area target is significantly optimized in the visualization images, which has certain positive significance for the application of the semantic segmentation algorithm in the field of unmanned driving.

关键词：

来源：评论

学校读者我要写书评

暂无评论

IceDCST: A Dilated Convolutional Segmentation Transformer for Ice Crystals

IceDCST: A Dilated Convolutional Segmentation Transformer fo...

引用

Information Technology,Big data and Artificial Intelligence (ICIBA), International Conference on ieee International Conference on

作者： Liting Liu Gang Zhao Department of Electronic Engineering and Information Science University of Science and Technology of China Hefei China

Ice crystals are one of the key factors limiting the quality and efficiency of preserved biological materials. However, the current quantitative analysis of ice crystals is often performed by the subjective judgment of the experimental operator, which leads to subjective and error-prone experimental results. Therefore, in this paper, IceDCST is proposed for the semantic segmentation of ice crystals images for objective and accurate quantitative analysis of the ice crystal growth *** designed Crossed Patch Merging to enable multi-scale features to be obtained while maintaining local information. Also, we designed Learable Location Information Module (LLI) to reduce the computational effort and introduce learnable location information. In addition, we design the Dilated Convolutional Block (DCB) to increase the perception of contextual spatial information. On the homemade ice crystals dataset, IceDCST performs best with 94.51% accuracy and 87.76 mIoU, which is 2.1% and 1.89 better than the accuracy and mIoU of the second-performing network, respectively. In this paper, the visualization results are analyzed and verified by quantitative analysis of ice crystals growth using IceDCST. The experimental results show that the quantitative analysis of ice crystals using IceDCST is feasible.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Human Activity Classification Using Recurrence Plot and Residual Network

Human Activity Classification Using Recurrence Plot and Resi...

引用

ieee Conference on Systems, Process & Control (ICSPC)

作者： Ching Hong Lew Kian Ming Lim Chin Poo Lee Jit Yan Lim Faculty of Information Science & Technology Multimedia University Melaka Malaysia Faculty of Information Science & Technology Multimedia University Malacca Malaysia

Human Activity Classification (HAC) is a challenging time series classification task, aiming to discern specific behaviors or movements of individuals based on sensor data. This paper proposed an innovative method for HAC that leverages the power of Residual Network and the unique insights provided by Recurrence Plots (RP). Recurrence Plots are a visualization technique that transforms time series data into graphical representations, highlighting temporal patterns and dependencies. In this proposed method, Recurrence Plots play a crucial role by capturing intricate temporal relationships in the sensor data, enabling more precise activity recognition. Furthermore, Residual Network (ResNet), a prominent deep learning architecture, is employed to create a robust and efficient classification model. ResNet is able to mitigate the vanishing gradient problem through residual connections, which is particularly advantageous in the context of time series classification. The experimental results on MotionSense, UCI-HAR, and USC-HAD datasets demonstrate the superiority of our proposed approach. When compared to five existing methods and a self-constructed transfer learning model, the proposed method consistently outperforms others, achieving the highest average accuracy of 84.01%. Notably, it reaches accuracy rates of 92.31%, 84.72%, and 75.00% for MotionSense, UCI-HAR, and USC-HAD datasets, respectively.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Hierarchical Multimodality Graph Reasoning for Remote Sensing Visual Question Answering

引用

ieee TRANSACTIONS ON GEOscience AND REMOTE SENSING 2024年 62卷

作者： Zhang, Han Wang, Keming Zhang, Laixian Wang, Bingshu Li, Xuelong Northwestern Polytech Univ Sch Artificial Intelligence Opt & Elect iOPEN Xian 710072 Peoples R China Space Engn Univ Key Lab Intelligent Space TTC&O Beijing 101416 Peoples R China Northwestern Polytech Univ Sch Software Xian 710129 Peoples R China Shenzhen Univ Natl Engn Lab Big Data Syst Comp Technol Shenzhen 518060 Peoples R China China Telecom Corp Inst Artificial Intelligence TeleAI Beijing 100033 Peoples R China

Remote sensing visual question answering (RSVQA) targets answering the questions about RS images in natural language form. RSVQA in real-world applications is always challenging, which may contain wide-field visual information and complicated queries. The current methods in RSVQA overlook the semantic hierarchy of visual and linguistic information and ignore the complex relations of multimodal instances. Thus, they severely suffer from vital deficiencies in comprehensively representing and associating the vision-language semantics. In this research, we design an innovative end-to-end model, named Hierarchical Multimodality Graph Reasoning (HMGR) network, which hierarchically learns multigranular vision-language joint representations, and interactively parses the heterogeneous multimodal relationships. Specifically, we design a hierarchical vision-language encoder (HVLE), which could simultaneously represent multiscale vision features and multilevel language features. Based on the representations, the vision-language semantic graphs are built, and the parallel multimodal graph relation reasoning is posed, which could explore the complex interaction patterns and implicit semantic relations of both intramodality and intermodality instances. Moreover, we raise a distinctive vision-question (VQ) feature fusion module for the collaboration of information at different semantic levels. Extensive experiments on three public large-scale datasets (RSVQA-LR, RSVQA-HRv1, and RSVQA-HRv2) demonstrate that our work is superior to the state-of-the-art results toward a mass of vision and query types.

关键词： visualization Semantics Cognition Remote sensing Question answering (information retrieval) Roads Feature extraction Attention mechanisms Sensors Convolution Hierarchical learning parallel multimodality graph reasoning remote sensing visual question answering (RSVQA)

来源：评论

学校读者我要写书评

暂无评论

Web-Based BIM Platform for Building Digital Twin

Web-Based BIM Platform for Building Digital Twin

引用

Digital Twins and Parallel Intelligence (DTPI), ieee International Conference on

作者： Satoru Furuta Jin Nakazato Manabu Tsukada Graduate School of Information Science and Technology The University of Tokyo Tokyo Japan

Building digital twin (BDT) is utilized throughout the lifecycle of a building. It serves for efficient operations during the design and construction phases, and during the operational and maintenance phases, it’s used for asset management and as field maps for robots. Building Information Modeling (BIM), which contains both semantics and geometry data of building elements, holds promise as a data source for BDT. We have extracted four key technical challenges of the digital twin, particularly vital during the operational and maintenance phases: software, visualization, update, and real-scene reconstruction. Due to the exclusive and static nature of BIM, these challenges also pose significant issues in the context of BDT. To address these challenges, we designed and implemented a Web-based BIM platform. The implemented application shows not only geometry data but also semantics data, enables easy overlay with the latest indoor conditions, and provides updating functionality. The developed system is essential for the continuous operation of BDT in dynamic indoor environments. We evaluated the application through a questionnaire survey.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A HPC-Based data Analytics Platform Architecture for data-Driven Animal Phenotype Detection

A HPC-Based Data Analytics Platform Architecture for Data-Dr...

引用

International Conference on Cloud Technologies and Applications (CloudTech)

作者： Manu Agarwal Enkhzol Dovdon Luis Roma Barge Yanja Dajsuren Jakob de Vlieg Mathematics & Computer Science Department Eindhoven University of Technology Eindhoven The Netherlands

Large-scale automatic detection of individual animal phenotypes is a long-standing desire in animal breeding. Further-more, the recent advances in AI, coupled with the availability of technologies such as RFIDs, Wireless Sensor Networks, and Camera Systems, have made it possible to automate the process of animal phenotype detection. An automated phenotype detection process has several phases, comprising of data collection, data fusion, data storage, data analysis/AI model development, and data serving. To facilitate the automated phenotype detection process, we developed a data analytics platform architecture. The designed architecture proposes various approaches vis-a-vis several phases of the automated phenotype detection process, along with implementing these approaches on Dutch nationalHPC infrastructure. Also, how data analysis results are servedto the various end users is very important for the success of adata analytics platform. To this end, we propose a scalable data visualization pipeline that enables animal behavior scientists to deduce quick summaries about animal behavior on a farm.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Training Robust Spiking Neural Networks with Viewpoint Transform and Spatiotemporal Stretching

Training Robust Spiking Neural Networks with Viewpoint Trans...

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Haibo Shen Juyu Xiao Yihao Luo Xiang Cao Liangqi Zhang Tianjiang Wang School of Huazhong University of Science and Technology Yichang Testing Technique Research Institute Changsha University

Neuromorphic vision sensors (event cameras) simulate biological visual perception systems and have the advantages of high temporal resolution, less data redundancy, low power consumption, and large dynamic range. Since both events and spikes are modeled from neural signals, event cameras are inherently suitable for spiking neural networks (SNNs), which are considered promising models for artificial intelligence (AI) and theoretical neuroscience. However, the unconventional visual signals of these cameras pose a great challenge to the robustness of spiking neural networks. In this paper, we propose a novel data augmentation method, View-Point Transform and SpatioTemporal Stretching (VPT-STS). It improves the robustness of SNNs by transforming the rotation centers and angles in the spatiotemporal domain to generate samples from different viewpoints. Furthermore, we introduce the spatiotemporal stretching to avoid potential information loss in viewpoint transformation. Extensive experiments on prevailing neuromorphic datasets demonstrate that VPT-STS is broadly effective on multi-event representations and significantly outperforms pure spatial geometric transformations. Notably, the SNNs model with VPT-STS achieves a state-of-the-art accuracy of 84.4% on the DVS-CIFAR10 dataset.

关键词： visualization Neuromorphics Biological system modeling Transforms Cameras Robustness Spatiotemporal phenomena

来源：评论

学校读者我要写书评

暂无评论

Accelerating Precision Research and Resolution Through Computational Archival science Pedagogy

Accelerating Precision Research and Resolution Through Compu...

引用

ieee International Conference on Big data

作者： Sarah A. Buchanan Jennifer L. Wachtel Jennifer A. Stevenson University of Missouri Columbia USA University of Maryland and National Archives and Records Administration ◊ Washington D.C. USA Defense Threat Reduction Agency Fort Belvoir USA

Use of archival collections is accelerated by the presence of finding aids, which communicate the arrangement and description of collection contents. To arrive at the optimal arrangement of a collection, archivists rely on some item-level processing or knowledge gained by exploring and manipulating digital reproductions of the contents. In this paper we consider archival student and instructor perspectives from hands-on course experiences directly with two distinct collections: one pertaining to the development, 2017 transfer and launch, and ongoing maintenance of the International Research Portal for Records Related to Nazi-Era Cultural Property (IRP2), and one a selection of unclassified catalog entries about digitized nuclear science reports. Visualizing is a data practice that permits the discovery of key content patterns, identification of computational models to be carried out to aid further analysis, and query-resolution for subject experts with precise - and historically significant - research questions. While archival data visualizations have previously been implemented as an extension of descriptive work including finding aid element counts, here we connect visualization to the work of archival outreach and access. We study how visualizations generated by groups of students working with textual and numerical dataset portions can ultimately accelerate time-sensitive uses of collections.

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：