检索结果-内蒙古大学图书馆

2024 International conference on computer Application and information Security, ICCAIS 2024

作者： Fan, Juntao Li, Qizhong Yang, Gaopan Li, Ning Li, Xiangyang Zhang, Wenbo Li, Yafeng Baoji Cigarette Factory Shaanxi Tobacco Industry Co. Ltd. Shaanxi Baoji721000 China School of Computer Baoji University of Arts and Science Shaanxi Baoji721000 China

ISBN: (纸本)9781510689312

Image segmentation is one of the most important tasks in the field of image processing. Although deep learning-based methods have achieved great success in image segmentation tasks, real-world images often contain mixed noise and lack segmentation labels. For such challenging problems, the segmentation accuracy of existing deep learning-based methods is significantly affected. We introduce the concept of depth K-median for the first time and propose an unsupervised image segmentation method based on depth K-median. This method integrates learning representations and clustering into a unified model while designing a novel unsupervised image segmentation algorithm. This algorithm alternately updates the parameters of the deep network and the clustering parameters. In an unsupervised setting, the new method not only effectively addresses the interference of mixed noise in image segmentation but also better learns the essential features of the data. To validate the effectiveness of the proposed method, extensive experiments on the general VOC 2012 and BSDS300 datasets, comparing existing deep learning-based methods and traditional machine learning methods, demonstrate the effectiveness and advantages of the new method. © The Authors.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

Insights into the High-Quality Development of Agriculture under the Wave of Digital Economy 20th

Insights into the High-Quality Development of Agriculture un...

引用

20th International conference on Metaverse, METAVERSE 2024, Held as Part of the Services conference Federation, SCF 2024

作者： ChaoYu, Liang XiangQiong, Wu Hui, Ou Shengping, Liang College of Information Engineering Capital Normal University Beijing100048 China College of Mathematics and Statistics Hunan Normal University Changsha410081 China China Forestry Group Corporation Beijing100026 China

ISBN: (纸本)9783031769764

Based on panel data from 30 provinces, autonomous regions, and municipalities in China from 2012 to 2021, this study conducted a quantitative analysis of the high-quality development of agriculture amidst the wave of the digital economy. The research revealed that the digital economy plays a significant role in promoting the high-quality development of agriculture. It not only directly enhances the high-quality development of agriculture in various provinces, autonomous regions, and municipalities but also indirectly provides a material foundation for such development by elevating the level of agricultural talent cultivation. However, the impact of the digital economy on the high-quality development of agriculture exhibits regional heterogeneity, with notable effects in the northern region and less significant impacts in the southern region. Consequently, this study’s analysis points out that strengthening strategies to empower agriculture with the digital economy, optimizing talent attraction and cultivation mechanisms, and enhancing differentiated policy support are crucial strategies for promoting the high-quality development of agriculture. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Agricultural economics

来源：评论

学校读者我要写书评

暂无评论

Threshold Matters in WSSS: Manipulating the Activation for the Robust and Accurate Segmentation Model Against Thresholds

Threshold Matters in WSSS: Manipulating the Activation for t...

引用

IEEE/CVF conference on computer Vision and Pattern Recognition (CVPR)

作者： Lee, Minhyun Kim, Dongseob Shim, Hyunjung Yonsei Univ Seoul South Korea

ISBN: (数字)9781665469463

ISBN: (纸本)9781665469463

Weakly-supervised semantic segmentation (WSSS) has recently gained much attention for its promise to train segmentation models only with image-level labels. Existing WSSS methods commonly argue that the sparse coverage of CAM incurs the performance bottleneck of WSSS. This paper provides analytical and empirical evidence that the actual bottleneck may not be sparse coverage but a global thresholding scheme applied after CAM. Then, we show that this issue can be mitigated by satisfying two conditions;1) reducing the imbalance in the foreground activation and 2) increasing the gap between the foreground and the background activation. Based on these findings, we propose a novel activation manipulation network with a per-pixel classification loss and a label conditioning module. Perpixel classification naturally induces two-level activation in activation maps, which can penalize the most discriminative parts, promote the less discriminative parts, and deactivate the background regions. Label conditioning imposes that the output label of pseudo-masks should be any of true image-level labels;it penalizes the wrong activation assigned to non-target classes. Based on extensive analysis and evaluations, we demonstrate that each component helps produce accurate pseudo-masks, achieving the robustness against the choice of the global threshold. Finally, our model achieves state-of-the-art records on both PAS-CAL VOC 2012 and MS COCO 2014 datasets.

关键词： Image segmentation computer vision Image resolution Codes Semantics Robustness Pattern recognition

来源：评论

学校读者我要写书评

暂无评论

A Knowledge Graph-Based Deep Learning Framework for Efficient Content Similarity Search of Sustainable Development Goals Data

引用

Data Intelligence 2023年第3期5卷 663-684页

作者： Irene Kilanioti George A.Papadopoulos School of Electrical and Computer Engineering National Technical University of Athens9 Heroon Polytechneiou St.Zografou Campus Athens15780Greece Department of Computer Science University of Cyprus1 University AvenueAglantziaCY-2109NicosiaCyprus

Sustainable development denotes the enhancement ofliving standards in the present without compromising future generations'*** Development Goals(SDGs)quantify the accomplishment of sustainable development and pave the way for a world worth living in for future *** can contribute to the achievement of the SDGs by guiding the actions of practitioners based on the analysis of SDG data,as intended by this *** propose a framework of algorithms based on dimensionality reduction methods with the use of Hilbert Space Filling Curves(HSFCs)in order to semantically cluster new uncategorised SDG data and novel indicators,and efficiently place them in the environment of a distributed knowledge graph ***,a framework of algorithms for insertion of new indicators and projection on the HSFC curve based on their transformer-based similarity assessment,for retrieval of indicators and loadbalancing along with an approach for data classification of entrant-indicators is ***,a thorough case study in a distributed knowledge graph environment experimentally evaluates our *** results are presented and discussed in light of theory along with the actual impact that can have for practitioners analysing SDG data,including intergovernmental organizations,government agencies and social welfare *** approach empowers SDG knowledge graphs for causal analysis,inference,and manifold interpretations of the societal implications of SDG-related actions,as data are accessed in reduced retrieval *** facilitates quicker measurement of influence of users and communities on specific goals and serves for faster distributed knowledge matching,as semantic cohesion of data is preserved.

来源：评论

学校读者我要写书评

暂无评论

Realization of tree and grass recognition based on AlexNet 8

Realization of tree and grass recognition based on AlexNet

引用

8th IEEE International conference on Cloud Computing and Intelligence Systems, CCIS 2022

作者： Yun-Zhou, Wang Min-Ling, Zhu Lei, Chen Peng, Zhao Hao-Nan, Liu Bo-Lang, Xu Beijing Information Science and Technology University Computer School Beijing100101 China Capital Normal University College English Department Beijing100048 China

ISBN: (纸本)9781665477352

With the development of artificial intelligence technology, China is playing an increasingly important role in the field of object neural network recognition. In the past, people's understanding of neural network recognition was limited to the template matching model, which was simple and clear, but the model emphasized that the image must be completely consistent with the template in the brain to be recognized. However, AI recognition should not only recognize images consistent with the template, but also recognize images inconsistent with the template. The accuracy of the final training results reached 99.15%, exceeding the existing expectations. And AlexNet was born in 2012. On the model, AlexNet contains several relatively new technical points, and has successfully applied ReLU, Dropout, LRN and other Tricks in CNN for the first time. At the same time, AlexNet also uses the GPU for computing acceleration. © 2022 IEEE.

关键词： Template matching

来源：评论

学校读者我要写书评

暂无评论

Measurement of the scale and development trend of digital economy core industries in China's provinces 11

Measurement of the scale and development trend of digital ec...

引用

11th International conference on information technology and Quantitative Management, ITQM 2024

作者： Zhang, Tianyu Li, Na School of Economics and Management University of Chinese Academy of Sciences No.80 Zhongguancun East Road Beijing100190 China Research Center on Fictitious Economy and Data Science Chinese Academy of Sciences No.80 Zhongguancun East Road Beijing100190 China

In order to measure the scale and development trend of China's inter-provincial digital economy, this paper calculates the scale of digital economy core industries and the added value of four different categories of core industries of 30 provinces in China from 2007 to 2017 based on the division standard of digital economy industries of the national Bureau of Statistics. In order to measure the results more accurately and reflect the characteristics of each province's industrial structure, this paper uses the input-output tables of more than 100 departments of each province in 2007, 2012 and 2017 and the data of the Economic Census Yearbook to split the digital and non-digital parts of the industry. The results show that the scale of digital economy industry in the eastern region of China is the largest and the development speed is fast. The scale of digital economy in the central region is relatively small, but the development speed is the fastest. The western and northeastern regions have the smallest digital economies and the slowest growth rates. In addition, among the four major industries, the digital economy application industry and digital product manufacturing industry account for the largest proportion, and the digital product service industry accounts for the smallest proportion. The research results of this paper can provide reference for establishing a national unified digital economy core industry measurement system and improving digital industry related data collection. © 2024 The Authors.

关键词： Digital econom Digital industrialization Industrial digitization IO table Regional difference

来源：评论

学校读者我要写书评

暂无评论

Spatio-Temporal Clustering of Forest Fire Hotspots in the Wallacea Region in 2012 – 2022 Using Chronnet 11

Spatio-Temporal Clustering of Forest Fire Hotspots in the Wa...

引用

11th International conference on computer, Control, Informatics and its Applications, IC3INA 2024

作者： Ismanto, Rido Dwi Vatresia, Arie Wahyuni, Intan Nuni Utama, Ferzha Putra Gusmanto, Rizki Anees, Mohd Talha Research Center for Computing National Research and Innovation Agency Bogor Indonesia Informatics-Engineering Faculty University of Bengkulu Bengkulu Indonesia Information System-Engineering Faculty University of Bengkulu Bengkulu Indonesia Department Geology Faculty of Science Universiti Sains Malaysia Kuala Lumpur Malaysia

ISBN: (纸本)9798331542313

Data from the Ministry of Environment and Forestry of Indonesia show that millions of hectares were burned by fires in Indonesia with 2015 being the greatest incidence in the last of ten years. Therefore, a research is required to examine the distribution pattern of hotspots for forest fires. This study used Chronnet to display the clustering of forest fire hotspot data in the Wallacea Region (North Maluku) between 2012 and 2022. The NASA FIRMS provided the MODIS data used in the study. Several findings have been made from the implemented Chronnet method. The degree of connectivity map shows a medium degree which suggests that the North Maluku frequently experiences forest and land fires of a moderate intensity. The strength map shows a complex pattern of fire activity, with high-intensity flames creating major concerns in some, particularly in West Halmahera Regency and low to moderate fire events occurring in most other places. The results from the Chronnet cluster show that most fires occurred in production forests, indicating the need for stronger fire prevention and regulations. While protected forests had fewer incidents, there are still several clusters that require attention. Additionally, the frequent fires in non-forest areas emphasize the need for better land-use planning and coordination. Finally, although conservation forests remained unaffected, it is crucial for the government to prioritize maintaining this record. © 2024 IEEE.

关键词： Premixed flames

来源：评论

学校读者我要写书评

暂无评论

The Storage and Sharing of Big Data in Air Traffic Management System

The Storage and Sharing of Big Data in Air Traffic Managemen...

引用

International conference on Big Data Intelligence and Computing, DataCom 2022

作者： Zhang, Yuan Wang, Qing Li, Jiahao Wu, Zhijun College of Safety Science and Engineering Civil Aviation University of China Tianjin China College of Electronic Information and Automation Civil Aviation University of China Tianjin China

ISBN: (纸本)9789819922321

With the continuous development of aviation business and informatization, the storage and sharing mechanism based on big data will play a more important role in the next generation Air Traffic Management (ATM) system. On the basis of analyzing the traditional aviation network, this paper proposes a new network structure for the storage and sharing of aviation big data to solve the transmission bottleneck, information security and other problems existing in the current network, and effectively ensure the safe and efficient operation of air traffic transportation. The structure is information-centric, with publish and subscribe as the main transmission paradigm, and includes multiple levels such as security, policy, data content and application. The main research contents of this paper are as follows. (i) Analyze the development trend of the new generation of ATM and the problems existing in the current aviation network, and in this context, propose the network structure for aviation big data. (ii) Design the policy layer of the new network structure, including content naming mode, routing mode and caching mode. (iii) Design the security layer of the new network structure, including data security, privacy protection and guarantee service, analyze the application of the new aviation network in specific aviation services. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Big data

来源：评论

学校读者我要写书评

暂无评论

A comparative study of Stock Price Prediction Based on BP and LSTM Neural Network 3

A comparative study of Stock Price Prediction Based on BP an...

引用

3rd International conference on Artificial Intelligence and computer Engineering, ICAICE 2022

作者： Huang, Shujia Wang, Ben Hao, Lingbo Si, Zebin School of Information Science and Technology Hangzhou Normal University Hangzhou China

ISBN: (纸本)9781510663473

In recent years, stock price prediction has become a research hotspot. The price of the stock market is unstable, which often rises or falls sharply due to the national policies, which makes it difficult for investors to achieve stable returns in the stock market. With the rapid rise of artificial intelligence, computers have become flexible in dealing with mathematical problems. Therefore, the extraordinary computing power of computers has been used to analyze and predict the trend of the stock market. More and more computer professionals began to enter the financial market and use neural network to study the trend of the stock market. This paper uses BP neural network and LSTM neural network to learn and predict the stock data of Shanghai Composite Index from January 2012 to June 2022. LSTM is a kind of RNN, but it is superior to other neural networks. It can effectively deal with data forgetting and gradient explosion problems, and bring reliability to the prediction results of the model. The two models are evaluated by analyzing MAE, MSE and the time required for model training. The results show that LSTM model can not only learn longer time span than BP model, but also better than BP model in MAE and MSE indexes, which provides some reference and guidance for the prediction of medium and long-term stocks. © 2023 SPIE.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

Deep Reinforcement Learning based Dynamic Channel Bonding for Wi-Fi Networks 21

Deep Reinforcement Learning based Dynamic Channel Bonding fo...

引用

21st IEEE International Symposium on Parallel and Distributed Processing with Applications, 13th IEEE International conference on Big Data and Cloud Computing, 16th IEEE International conference on Social Computing and Networking and 13th International conference on Sustainable Computing and Communications, ISPA/BDCloud/SocialCom/SustainCom 2023

作者： Chen, Hao Liu, Peng You, Lizhao Guo, Ziyang Luo, Jiajun Sun, Xinghua Wang, Taotao Fu, Liqun Xiamen University Department of Information and Communication Engineering China Shenzhen Research Institute of Xiamen University Shenzhen China Huawei Technologies Co. Ltd. Wireless Technology Lab 2012 Laboratories China Sun Yat-sen University School of Electronics and Communication Engineering China Shenzhen University College of Electronics and Information Engineering Shenzhen China

ISBN: (纸本)9798350329223

This paper presents a decentralized Deep Reinforcement Learning (DRL)-based dynamic channel bonding (DCB) algorithm (i.e., drlDCB) for Wi-Fi networks. Most existing RL-based channel bonding algorithms either are centralized (i.e., requiring a central controller and global information for training), or can easily fall into local optimum (e.g., over-using channels for demand) as the number of stations increases. Moreover, they may converge slowly due to the use of measured throughput or delay over a duration as reward. In contrast, drlDCB is a purely distributed algorithm where each access point gets trained using local information, and addresses the local optimum and the slow convergence problems through the following new designs: 1) a new piecewise throughput model to avoid occupying too many channels if demand is satisfied, and 2) a new reward design using estimated throughput. The use of estimated throughput allows per-frame adaptation and fast convergence speed, while the accurate estimation generalizes the application scenarios, allowing an offline-trained model to be used in a dynamic environment. Simulation results in the static scenario with fixed loads and the dynamic scenario with random loads show that drlDCB outperforms state-of-the-art DCB algorithms in terms of network throughput, delay and convergence speed. © 2023 IEEE.

关键词： Reinforcement 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：