检索结果-内蒙古大学图书馆

您好，读者！请登录

内蒙古大学图书馆

首页
概况
党建
资源
服务
科研支持
- 论文收录引用证明
- 科技查新
知识产权
档案馆
帮助

咨询与建议

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

您的常用邮箱：*

您的手机号码：*

问题描述：

当前已输入0个字，您还可以输入200个字

全部搜索
期刊论文
图书
学位论文
标准
纸本馆藏
外文资源发现
数据库导航
超星发现

高级检索

时间限定

出版年份：

文献类型

图书期刊文献学位论文多媒体

馆藏选择

电子馆藏纸本馆藏

核心期刊

全部期刊 SCI 收录期刊 SSCI 收录期刊 EI 收录期刊 CSCD 收录期刊 CSSCI 收录期刊

语言

中文英文

文献类型

期刊文献图书学位论文标准纸本馆藏

帮助

文字说明：

T=题名（书名、题名），A=作者（责任者），K=主题词，P=出版物名称，PU=出版社名称，O=机构（作者单位、学位授予单位、专利申请人），L=中图分类号，C=学科分类号，U=全部字段，Y=年（出版发行年、学位年度、标准发布年）

检索规则说明：

AND代表“并且”；OR代表“或者”；NOT代表“不包含”；(注意必须大写,运算符两边需空一格)

检索范例：

范例一：(K=图书馆学 OR K=情报学) AND A=范并思 AND Y=1982-2016
范例二：P=计算机应用与软件 AND (U=C++ OR U=Basic) NOT K=Visual AND Y=2011-2016

分类表

所选分类

>> <<

限定检索结果

文献类型

5,062 篇 会议
45 篇 期刊文献
5 册 图书

馆藏范围

5,112 篇 电子文献
0 种 纸本馆藏

日期分布

学科分类号

4,129 篇 工学
- 3,187 篇 计算机科学与技术...
- 1,609 篇 电气工程
- 1,562 篇 信息与通信工程
- 1,288 篇 软件工程
- 564 篇 控制科学与工程
- 344 篇 动力工程及工程热...
- 262 篇 仪器科学与技术
- 248 篇 电子科学与技术（可...
- 201 篇 机械工程
- 116 篇 石油与天然气工程
- 116 篇 网络空间安全
- 111 篇 测绘科学与技术
- 109 篇 安全科学与工程
- 91 篇 交通运输工程
- 77 篇 生物工程
- 76 篇 环境科学与工程（可...
- 75 篇 生物医学工程（可授...
- 74 篇 材料科学与工程（可...
- 64 篇 化学工程与技术
834 篇 理学
- 448 篇 数学
- 204 篇 物理学
- 123 篇 化学
- 123 篇 统计学（可授理学、...
- 102 篇 系统科学
- 99 篇 生物学
581 篇 管理学
- 438 篇 管理科学与工程(可...
- 209 篇 图书情报与档案管...
- 169 篇 工商管理
113 篇 医学
- 82 篇 临床医学
- 65 篇 基础医学(可授医学...
63 篇 经济学
52 篇 法学
40 篇 农学
25 篇 军事学
14 篇 教育学
6 篇 文学
3 篇 艺术学

主题

620 篇 wireless sensor ...
166 篇 sensor nodes
140 篇 sensor systems
137 篇 distributed comp...
123 篇 wireless sensor ...
95 篇 sensor networks
93 篇 distributed powe...
92 篇 internet of thin...
88 篇 energy efficienc...
85 篇 monitoring
81 篇 cloud computing
81 篇 distributed comp...
72 篇 sensors
72 篇 computer network...
68 篇 protocols
65 篇 routing
63 篇 security
59 篇 wireless communi...
59 篇 real-time system...
58 篇 computer archite...

机构

12 篇 maharishi school...
12 篇 jiangsu high tec...
12 篇 ieee
10 篇 college of elect...
9 篇 chitkara univers...
9 篇 chitkara univers...
8 篇 karnataka bangal...
8 篇 henan univ inst ...
8 篇 payame noor univ...
8 篇 wuhan univ sch c...
7 篇 beijing univ pos...
7 篇 tu wien distribu...
6 篇 dalian univ tech...
6 篇 college of techn...
6 篇 shanghai jiao to...
6 篇 centre for inter...
6 篇 univ pittsburgh ...
6 篇 indian sch mines...
6 篇 centre for cyber...
6 篇 chinese acad sci...

作者

14 篇 han zhijie
10 篇 dustdar schahram
10 篇 china
10 篇 zhang wei
9 篇 barolli leonard
8 篇 wang lei
8 篇 liu xue
7 篇 liu yang
6 篇 stankovic john a...
6 篇 xhafa fatos
6 篇 zakaria a.
6 篇 lin chi
6 篇 li wei
6 篇 abdelzaher tarek
6 篇 mukherjee nandin...
6 篇 li hui
5 篇 sarigiannidis pa...
5 篇 yang yang
5 篇 shakaff a.y.m.
5 篇 prasad ramjee

语言

5,016 篇 英文
89 篇 其他
7 篇 中文

检索条件"任意字段=3rd IEEE International Conference on Distributed Computing in Sensor Systems"

共 5112 条记录，以下是11-20 订阅

全选清除本页清除全部题录导出标记到"检索档案"

详细简洁

排序：

A Context-Aware Framework for sensor Placement Using PSO and Voronoi in Dynamic Scenarios 3

A Context-Aware Framework for Sensor Placement Using PSO and...

引用

3rd international conference on Intelligent systems, Advanced computing, and Communication, ISACC 2025

作者： Manna, Prithwish Majumder, Sudipta Singh, N. Hemarjit Bose, Rajesh DUIET Dibrugarh University Dept. of CSE Assam Dibrugarh India DUIET Dibrugarh University Dept. of ECE Assam Dibrugarh India JIS University Dept. of CSE West Bengal India

ISBN: (纸本)9798331523893

When we deploy sensors, we face severe problems to achieve resilience and robust coverage. It happens more in stochastic and dynamic environments. In this paper, we propose a hybrid framework which is a combination of Particle Swarm Optimization (PSO) and Voronoi Diagram method with context aware adaptation to fix the challenges. The proposed framework balances initial distribution of sensors and iterative optimization to enhance coverage, maintain k-coverage (k=3) and adapt to evolving scenarios even under critical situations like sensor failures. Experimental results highlight the performance of the proposed framework across the key scenarios: comparing coverage with or without context aware optimization quantifying the impact of random node failures, tracking cumulative recovery attempts and evaluating final coverage percentage. The results reveal that this hybrid framework maximizes the coverage and provides resilience to sensor failures. It also reallocates resources dynamically to maintain optimal deployment. This framework is suitable for getting more efficiency and minimal iterations of recovery while enhancing the robustness of the network. It also enlightens the adaptability in sensor performance and becomes very popular in modern applications. © 2025 ieee.

关键词： Adaptive Deployment and Fault Tolerant Networks Context Aware Adaptation Particle Swarm Optimization Voronoi Diagram

来源：评论

学校读者我要写书评

暂无评论

Efficient Signal Processing: Harnessing distributed Arithmetic for High-Speed FFT Operations 3

Efficient Signal Processing: Harnessing Distributed Arithmet...

引用

3rd international conference on Intelligent and Innovative Technologies in computing, Electrical and Electronics, IITCEE 2025

作者： Bharathi, M. Mohanarangam, Krithikaa Shirur, Yasha Jyothi M. Andhra Pradesh Tirupati India Symbiosis Institute of Technology Pune India Bnm Institute of Technology Electronics and Communication Engineering Bangalore India

ISBN: (纸本)9798331515911

The present article introduces a novel approach to significantly enhance the computational efficiency of modern Digital Signal Processing (DSP) processors. The algorithm enhances the efficiency of Sum of Products (SOP) operations, which are essential for real-time signal processing applications, by utilizing distributed Arithmetic (DA) techniques for the Fast Fourier Transform (FFT) and its inverse (Inverse FFT) (IDFT) computations. The present study undertakes an exhaustive quantitative analysis to assess the computational delays of the proposed 2, 4, and 8-point Fast Fourier Transform (FFT) and Inverse Discrete Fourier Transform (IDFT) across various bit representations (4, 8, and 16-bit). For example, when using 4-bit representations, the computational delays for a 2-point Fast Fourier Transform (FFT) are measured to be 13.318ns, 30.629ns for a 4-point FFT, and 27.779ns for an 8-point FFT. Similarly, the delays for the 2-point Fast Fourier Transform (FFT) in an 8-bit representation are 13.727ns, the 4-point FFT in 22.276ns, and the 8-point FFT in 30.592ns. The study further includes the addition of appropriate delays for IDFT operations, which enhances the understanding of the effectiveness of the approach suggested in various scenarios. Furthermore, the research presents and evaluates three distinct categories of transforms, namely 2-point, 4-point, and 8-point FFT and IDFT structures. These structures utilize different quantities of Look-Up Tables (LUTs) for computational purposes. The research showcases the exceptional computing efficiency of the suggested design through detailed quantitative analysis, emphasizing its potential advantages in improving DSP implementations in practical situations. © 2025 ieee.

关键词： Inverse problems

来源：评论

学校读者我要写书评

暂无评论

Flex sensor Enabled Glove based Manual Wheelchair for Elderly and Speech-Impaired Assistance 3

Flex Sensor Enabled Glove based Manual Wheelchair for Elderl...

引用

3rd international conference on Intelligent systems, Advanced computing, and Communication, ISACC 2025

作者： Tripathi, Abhishek Gopi Reddy, Kasu V S Amarnath, K. Bavesh, C. Vardhan, Bestha Singh, Sumitra Kalasalingam Academy of Research and Education Department of Computer Science and Engineering Virudhunagar India Central Electronics Engineering Research Institute Semiconductor Devices Design Group Pilani India

ISBN: (纸本)9798331523893

This paper presents a cost-effective manual wheelchair system designed to enhance mobility and communication for deaf-mute, disabled, and elderly individuals. The system employs a gesture-based mechanism using a glove with embedded flex sensors to detect finger movements, translating them into actionable commands. These commands are processed by a microcontroller to provide real-time feedback through three channels: visual messages displayed on an LCD, wireless communication with a caretaker's device, and audio feedback via a speaker. Testing demonstrates reliable gesture detection and response generation, highlighting the system's effectiveness in meeting the diverse needs of users. The design emphasizes affordability, ease of use, and adaptability, offering a practical and inclusive solution for individuals with varying levels of mobility and communication challenges. © 2025 ieee.

关键词： Accessibility Communication Gesture sensors Wheelchair

来源：评论

学校读者我要写书评

暂无评论

Recent Advances in Human Activity Recognition Techniques: A Comprehensive Review 3

Recent Advances in Human Activity Recognition Techniques: A ...

引用

3rd international conference on Intelligent systems, Advanced computing, and Communication, ISACC 2025

作者： Saha, Sukanya Banik, Debapriya ICFAI University Tripura Department of Computer Science and Engineering Agartala India

ISBN: (纸本)9798331523893

Human Activity Recognition (HAR) interprets data from healthcare, security surveillance, human-computer interaction (HCI), and sensors to recognize and categorize human actions. The rapid development of artificial intelligence (AI) and the widespread use of sensors have drawn great interest and demonstrated the immense application potential of HAR systems based on non-invasive ambient sensors. Various methods for performing HAR have been reported by researchers. In this survey, it has been noticed that data can be collected in vision and sensor-based formats. The goal of video-based HAR is to automatically identify actions in video clips. Smart wearable sensors are the most popular kind of sensor-based HAR. Surveys have identified HAR tendencies across many contexts and applications. The researchers classified HAR techniques into three categories: online versus offline, multimodal versus unimodal, and feature-based versus learning-based approaches. This survey has focused on the recognition of different activities like sports, yoga, different dance poses, human interaction, and human actions. This survey also concentrated on single-view and multiview datasets. This study compares and contrasts the most recent HAR techniques with a discussion of widely used datasets. Various criteria, including dataset size and assessment metrics, were used to categorize and analyze the chosen studies. This survey also focused on different dataset availability. © 2025 ieee.

关键词： Computer Vision Human Activity Recognition RGB Camera Security & Surveillance sensor

来源：评论

学校读者我要写书评

暂无评论

Intelligent Image Captioning with InceptionV3, LSTM, and PySpark Integration 3

Intelligent Image Captioning with InceptionV3, LSTM, and PyS...

引用

3rd international conference on Intelligent and Innovative Technologies in computing, Electrical and Electronics, IITCEE 2025

作者： Kurian, Thanu Mathew, Jimsha K. Syam Dev, R.S. Anand, P.K. Vignesh, A. Rakshith, B.S. New Horizon College of Engineering Dept of Artificial Intelligence and Machine Learning Bengaluru India

ISBN: (纸本)9798331515911

Image captioning is a challenging task in artificial intelligence that involves generating descriptive captions for images automatically. In this project, we propose a novel approach leveraging advanced technologies such as PySpark, LSTM, and InceptionV3 to develop an effective image captioning system. We harness the power of PySpark, a distributed computing framework, to efficiently process large-scale image datasets and extract high-level features from images using the InceptionV3 convolutional neural network (CNN) model. These features capture the semantic information present in the images and serve as input to the caption generation model. The caption generation model utilizes LSTM neural network as the decoder component. LSTM is well-suited for sequential data processing and is capable of generating coherent and contextually relevant captions based on the extracted image features. The InceptionV3 model acts as the encoder, extracting meaningful visual features from input images, while the LSTM decoder generates captions by decoding these features into natural language descriptions. This multimodal approach enables the model to understand and describe the content of images accurately. Through experimentation and evaluation on diverse image datasets, our proposed system demonstrates promising results in generating accurate and human-like captions for a wide range of images. By integrating PySpark, LSTM, and InceptionV3, our image captioning system achieves state-of-the-art performance, highlighting the effectiveness of leveraging advanced technologies in solving complex AI tasks. © 2025 ieee.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

GCN-Integrated Multi-Agent Reinforcement Learning (MARL) with Modified Firefly Algorithm for Cross-Layer Optimization in Cognitive sensor Networks 3

GCN-Integrated Multi-Agent Reinforcement Learning (MARL) wit...

引用

3rd international conference on Integrated Circuits and Communication systems, ICICACS 2025

作者： Jakranpally, Karthik Babu, Nagabathula Ramesh Babu, Doddavarapu Sreenu Rakesh, K. Barve, Amit Shelke, Atmaram Valiant It Services 4000 Stanwick ave SacramentoCA95835 United States J.B. Institute of Engineering & Technology Department of Electronics and Communication Engineering Hyderabad India Bharatratna Indira Gandhi College of Engineering Department of Electronics and Communication Engineering Solapur India Malla Reddy Engineering College for Women Department of Electronics and Communication Engineering Hyderabad India Parul University Parul Institute of Technology Faculty of Engineering and Technology Department of Computer Science and Engineering P.O.Limda Gujarat India Pune India

ISBN: (纸本)9798331508456

CSN s are integrating cognitive abilities, making it challenging to achieve optimized cross-layer performance with limited resources. Non-coordinated techniques typically demonstrate lower proficiency in dynamic spectrum management and network synchronization. We introduce Graph Convolutional Networks (GCN) and Multi-Agent Reinforcement Learning (MARL) in our architectural solution, with the help of the specially designed Firefly Algorithm. The framework implements a distributed learning process, utilizing GCN for local topology perception and MARL for cooperative spectrum and optimal routing decisions. The results of experimental analysis show large gains compared to traditional methods. The proposed model provided a 94.2% recognition rate in spectrum hole identification, an 89.7% enhancement in network throughput, and a 78.5% diminution in end-to-end delay. The energy efficiency surpasses baseline techniques by 67.3%, while the raw packet delivery reliability remains at 92.8%. Integrating GCN with MARL and the improved Firefly Algorithm can significantly improve cross-layer optimization, offering a promising solution for the next generation of sensor networks. © 2025 ieee.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Predictive Analytics in Mental Health: Machine Learning Models for Major Depressive Disorder Detection using sensor Data 3

Predictive Analytics in Mental Health: Machine Learning Mode...

引用

3rd international conference on Electronics and Renewable systems, ICEARS 2025

作者： Padmakala, S. Gopukumar, S.T. Saveetha University Saveetha School of Engineering Department of Computer Science and Engineering Tamil Nadu Chennai602 105 India Saveetha Medical College and Hospital Tamil Nadu Chennai602 105 India

ISBN: (纸本)9798331509675

The recent integration of Machine Learning (ML) and sensor technologies in healthcare, particularly for diagnosing Major Depressive Disorder (MDD), has paved the way for advanced predictive analytics in mental health. This study aimed to develop ML models to detect MDD using sensor-derived data, including accelerometer measurements, temperature, pulse rates, and SpO2 levels. The proposed methodology involved collecting and preprocessing a comprehensive dataset, imputing missing values, and splitting the data into training and test sets. Feature scaling and PCA were applied to enhance model efficiency. The performance of various ML models such as Random Forest, SVM, Gradient Boosting, XGBoost, LightGBM, and Logistic Regression was evaluated based on accuracy, precision, recall, and F1-score. Random Forest and XGBoost models performed exceptionally well, with accuracy rates nearing 97%. Ensemble methods, in particular, demonstrated superior predictive power. Logistic Regression showed significantly lower performance, highlighting the complexity of MDD beyond linear classification methods. Future directions include developing real-time analysis models, personalizing predictive algorithms, integrating ML into clinical practice, and conducting longitudinal studies to validate and refine these tools. Ethical and privacy considerations, as well as expanding predictive factors, are critical for advancing this research field. The study concluded that ML models are potent tools for early detection of MDD, which can significantly impact patient outcomes. The potential for these models to be integrated into clinical settings for personalized patient care could revolutionize mental health diagnostics and treatment strategies. © 2025 ieee.

关键词： Patient treatment

来源：评论

学校读者我要写书评

暂无评论

A Flexible Hybrid Resource Allocation Model for Enhanced Efficiency using AI in Wireless sensor Networks 3

A Flexible Hybrid Resource Allocation Model for Enhanced Eff...

引用

3rd international conference on Electronics and Renewable systems, ICEARS 2025

作者： Endla, Purushotham Mohandas, R. Saikia, Achinta Lakshmi, C. Karthikeyan, G. Moharekar, Tejashree Tejpal Sr University School of Sciences and Humanities Department of Physics Telangana Warangal506371 India Chennai Institute of Technology Department of Ece Chennai India Mariani College Department of Statistics Mariani Assam Jorhat785634 India Saveetha Engineering College Chennai India Karpagam College of Engineering Department of Artificial Intelligence & Data Science Coimbatore India Shivaji University Yashwantrao Chavan School of Rural Development Maharashtra Kolhapur India

ISBN: (纸本)9798331509675

Wireless sensor Networks (WSNs) have become a cornerstone technology in various fields, including environmental monitoring, industrial automation, and healthcare. This paper presents a novel hybrid resource allocation model designed to enhance efficiency in Wireless sensor Networks (WSNs) using advanced Artificial Intelligence (AI) techniques. The proposed model integrates static and dynamic resource allocation strategies with machine learning and reinforcement learning to optimize key network resources such as energy, bandwidth, and processing power. Comprehensive evaluations demonstrate that the proposed model outperforms nine existing models across multiple metrics, including energy efficiency, network lifetime, throughput, delay, latency, and accuracy of resource allocation. The proposed model achieved an energy efficiency of 0.98 J/bit, extended network lifetime to 620 days, and maintained a high throughput of 620 Kbps. Notably, the model demonstrated the lowest delay (90 ms) and latency (180 ms) while achieving the highest resource allocation accuracy of 97.8%. These results highlight the model's capability to dynamically adjust resource allocation in response to real-time network conditions, ensuring optimal performance and resource utilization. The proposed model's success in efficiently managing WSN resources marks a significant advancement in the field, offering a robust solution for various WSN applications requiring high efficiency and reliability. © 2025 ieee.

关键词： Artificial Intelligence Energy Efficiency Machine Learning Network Lifetime Reinforcement Learning Resource Allocation Wireless sensor Networks

来源：评论

学校读者我要写书评

暂无评论

Patient Health Internet of Things Wearable Device sensor with Machine Learning for Tracking System 3

Patient Health Internet of Things Wearable Device Sensor wit...

引用

3rd international conference on Electronics and Renewable systems, ICEARS 2025

作者： Nancy, R. Jonathan, R. Moorthy, Jagadeesan Latha, R. Thangavelu, Raja Srm Institute of Science and Technology Srm College of Pharmacy Kattankulathur Chengalpattu India Srm Institute of Science and Technology Srm College of Pharmacy Department of Pharmacy Practice Kattankulathur Chengalpattu603203 India Department of Commerce Nagapattinam India University of Technology and Applied Sciences College of Engineering and Technology Shinas Oman

ISBN: (纸本)9798331509675

Monitoring of patients and updates on their health status are important, especially in pandemic situations. Hospitalization in India is very costly for locals with average incomes because medical equipment is very costly and the population is also increasing, thus reducing the number of hospital beds. Moreover, hospitals in rural areas are mostly far from the locations where patients live. To overcome such challenges, this paper suggests an efficient and economical IoT-based system with sensors and a mobile application to monitor patients' health information from their location continuously. Doctors can also access real-time updates of patients' health through the mobile application. Utilizing improved machine learning processes for better accuracy and real-time response, the current research brings forth a new wearable IoT-integrated device targeted for stuttering monitoring and correction. The gadget combines different hardware and software elements, such as sensors, microcontrollers, and communication protocols, to accurately record the patient's health status. © 2025 ieee.

关键词： Electronic health record

来源：评论

学校读者我要写书评

暂无评论

Design and Development of Smart Monitoring System for Reducing Early Chicken Mortality 3

Design and Development of Smart Monitoring System for Reduci...

引用

3rd international conference on Intelligent systems, Advanced computing, and Communication, ISACC 2025

作者： Oeishee, Moslema Hoque Ahammad Muyeed, Munyem Mehedi, Rejoan Nayan Molla, Md Alvi, Mishkat Brac University Electrical and Electronic Engineering Department Dhaka Bangladesh United International University Computer Science and Engineering Department Dhaka Bangladesh

ISBN: (纸本)9798331523893

This paper presents an IoT-based autonomous farming method for lowering the mortality rate of 7-10-day-old chickens by preventing the flow of infection and continuously monitoring and controlling some very pivotal environmental parameters. In this system, an ESP32 microcontroller is used as the central unit which manages and controls all the sensors, ensuring healthier environmental conditions within the farm. The system uses several sensors like the temperature and humidity sensor DHT11. To create a user-friendly interface, real-time data is displayed via a 16x2 LCD monitor presenting the consumers with instant status of the crucial parameters. Moreover, this system will integrate two specialized robotic arms, identifying inactive or dead chickens and separating them to a designated place from the hatching area. This is proven to be significantly efficient in minimizing infection rates and ensuring a healthier surrounding for the remaining chickens. The incorporation of a buck converter allows high-power sources to step down so that low-power devices can run smoothly. In the implementation phase, we bought 250 chickens which from 222 chickens survived yielding a favorable cost benefit ratio of 0.48 indicating the economic validity of the system. The results of this study show that advanced sensor systems and robotic technology can be used in poultry farming to improve farm productivity reducing mortality rates. © 2025 ieee.

关键词： Autonomous Farming Early Chicken deaths Infection prevention Microcontroller Robotic arm

来源：评论

学校读者我要写书评

暂无评论

没有更多数据了...

全选清除本页清除全部题录导出标记到“检索档案”

共500页 << < 1 2 3 4 5 6 7 8 9 10 > >>

检索报告对象比较合并检索0

隐藏清空

合并搜索

回到顶部

执行限定条件

内容：

评分：

请选择保存的检索档案：

请选择收藏分类：

订阅名称：

通借通还

温馨提示：

图书名称：

借书校区：

取书校区：

手机号码：

邮箱地址：

一卡通帐号：

电话和邮箱必须正确填写，我们会与您联系确认。

联系人：

所在院系：

联系邮箱：

联系电话：

内蒙古自治区呼和浩特市赛罕区大学西街235号邮编: 010021

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：