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

6,783 篇 会议
621 册 图书
128 篇 期刊文献

馆藏范围

7,531 篇 电子文献
2 种 纸本馆藏

日期分布

学科分类号

3,894 篇 工学
- 3,086 篇 计算机科学与技术...
- 1,397 篇 软件工程
- 792 篇 电气工程
- 683 篇 信息与通信工程
- 509 篇 控制科学与工程
- 299 篇 生物工程
- 280 篇 生物医学工程（可授...
- 272 篇 机械工程
- 192 篇 光学工程
- 162 篇 电子科学与技术（可...
- 144 篇 仪器科学与技术
- 140 篇 化学工程与技术
- 59 篇 安全科学与工程
- 45 篇 材料科学与工程（可...
- 43 篇 网络空间安全
1,218 篇 理学
- 670 篇 数学
- 355 篇 物理学
- 334 篇 生物学
- 197 篇 统计学（可授理学、...
- 147 篇 化学
- 105 篇 系统科学
551 篇 管理学
- 314 篇 管理科学与工程(可...
- 253 篇 图书情报与档案管...
- 59 篇 工商管理
274 篇 医学
- 230 篇 临床医学
- 109 篇 基础医学(可授医学...
- 75 篇 药学(可授医学、理...
64 篇 法学
- 60 篇 社会学
51 篇 艺术学
- 49 篇 设计学（可授艺术学...
42 篇 农学
- 39 篇 作物学
20 篇 教育学
19 篇 文学
18 篇 经济学
5 篇 军事学

主题

3,142 篇 pattern recognit...
534 篇 feature extracti...
421 篇 artificial intel...
385 篇 neural networks
294 篇 data mining
275 篇 machine learning
232 篇 face recognition
223 篇 pattern matching
214 篇 support vector m...
203 篇 image segmentati...
194 篇 pattern classifi...
187 篇 computer imaging...
183 篇 computer science
182 篇 local binary pat...
169 篇 computational mo...
160 篇 computer vision
148 篇 humans
145 篇 testing
142 篇 training
141 篇 accuracy

机构

40 篇 school of comput...
38 篇 national laborat...
25 篇 institute of ima...
24 篇 center for resea...
24 篇 computer vision ...
22 篇 institute of ima...
22 篇 university of ch...
21 篇 department of co...
20 篇 indian statistic...
19 篇 national laborat...
19 篇 institute of com...
16 篇 tongji universit...
16 篇 chinese academy ...
15 篇 institute for pa...
14 篇 national center ...
14 篇 chinese acad sci...
14 篇 częstochowa univ...
13 篇 agh university o...
13 篇 school of electr...
12 篇 indian stat inst...

作者

35 篇 yang jie
31 篇 de-shuang huang
25 篇 tieniu tan
23 篇 umapada pal
19 篇 ryszard tadeusie...
19 篇 liang wang
18 篇 ajith abraham
17 篇 marcin korytkows...
17 篇 lu hanqing
17 篇 jacek m. zurada
17 篇 leszek rutkowski
16 篇 jie yang
16 篇 zhenan sun
16 篇 shamsuddin siti ...
16 篇 pal umapada
15 篇 wang da-han
15 篇 chew lim tan
15 篇 rafał scherer
14 篇 xu changsheng
14 篇 lotfi a. zadeh

语言

7,399 篇 英文
74 篇 中文
67 篇 其他

检索条件"任意字段=International Conference of Soft Computing and Pattern Recognition"

共 7533 条记录，以下是91-100 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：

Cross-Modality Image Quality Prediction for Time-Resolved CT from Breathing Signals

Cross-Modality Image Quality Prediction for Time-Resolved C...

引用

Workshop on Longitudinal Disease Tracking and Modeling with Medical Images and Data, LDTM 2024, 5th international Workshop on Multiscale Multimodal Medical Imaging, MMMI 2024, 1st Workshop on Machine Learning for Multimodal/-sensor Healthcare Data, ML4MHD2024 and Workshop on Multimodal Learning and Fusion Across Scales for Clinical Decision Support, ML-CDS 2024 held in conjunction with the 27th international conference on Medical Image computing and Computer Assisted Intervention, MICCAI 2024

作者： Schwarz, Annette Dickmann, Jannis Hofmann, Christian Szkitsak, Juliane Bert, Christoph Maier, Andreas Arias-Vergara, Tomás Pattern Recognition Lab Friedrich-Alexander-Universität Erlangen-Nürnberg Erlangen Germany Siemens Healthineers AG Forchheim Germany Department of Radiation Oncology Universitätsklinikum Erlangen Friedrich-Alexander-Universität Erlangen-Nürnberg Erlangen Germany Erlangen Germany

ISBN: (纸本)9783031845246

Four-dimensional computed tomography (4DCT) is a time-resolved, multi-modal imaging method that captures respiratory signals synchronised with the CT scan in order to track the movement of the lung. It is routinely used in radiation therapy treatment planning for lung or liver cancer patients. However, image artifacts during 4DCT scans caused by irregular patient breathing negatively impact treatment outcomes. This work proposes a method to automatically detect patients at high risk for severe image artifacts even before the scan is conducted based on an pre-scan analysis of their breathing. This can help to take proactive measures to improve image quality, such as changing the scan mode or providing in-depth patient coaching. A deep neural network is trained to predict the image quality score of 28 lung CT phantom scans, each rated by ten clinical experts. Different pretrained networks are investigated for feature generation and combined with two linear output heads to predict the average expert image quality score of unseen scans. We were able to predict the quality of a 4DCT with a mean absolute error of 8% using only the one-dimensional breathing signal as input. This accuracy is comparable to the rating consistency of our clinical experts, which were rating the images directly. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Radiotherapy

来源：评论

学校读者我要写书评

暂无评论

An Efficient Approach for Improving the Performance of Autonomous Vehicle Using Advanced Computer Vision 15th

An Efficient Approach for Improving the Performance of Auton...

引用

15th international conference on soft computing and pattern recognition, SoCPaR 2023 and 14th World Congress on Nature and Biologically Inspired computing, NaBIC 2023

作者： Saini, Vaibhav Jain, Ayushi Anurag Kantipudi, M. V. V. Prasad Department of Electronics and Telecommunication Symbiosis International (Deemed University) Symbiosis Institute of Technology Pune India Department of Computer Science Institute of Technology Roorkee Roorkee India

ISBN: (纸本)9783031810855

The advancement of Autonomous Vehicle technology and the development of the automotive sector both takes the benefit with the use of computer vision. In this research article, we have provided a thoroughly organized, high-level analysis of computer vision, which is widely employed in the automotive industry. By using machine learning with all the features and operating procedures necessary for employing the same technology, we have proposed a concept of sophisticated computer vision for autonomous vehicles. software and hardware aspects of it using a method based on machine learning has been discussed that will produce reliable results. Technological advancements have been highlighted with predicted reliable results, and difficulties that can be overcome when putting our suggested approach into practice. Finally, actual-world examples of cutting-edge computer vision with applications like self-driving cars, assembly automation, quality control, and more have been presented. The automotive industry may greatly profit from this strategy of superior computer vision with machine learning, data optimization, and cutting-edge tools & technology. For future prospective, this level of methodology can be used for other industries as well like health and agriculture etc. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Automotive Industry Autonomous Vehicle Computer Vision Cutting-edge tools & technology Machine Learning Self-driving cars

来源：评论

学校读者我要写书评

暂无评论

Detecting Cyber Bullying in the Social Media Using Deep Learning Algorithms

Detecting Cyber Bullying in the Social Media Using Deep Lear...

引用

15th international conference on soft computing and pattern recognition, SoCPaR 2023 and 14th World Congress on Nature and Biologically Inspired computing, NaBIC 2023

作者： Mridul, Aunik Hasan Ahsan, Nowreen Sultana, Zakia Asif, Asifuzzaman Computer Science and Engineering Daffodil International University Dhaka Bangladesh Department of Data Science Clarkson University New York United States Computer Science and Engineering Lovely Professional University Punjab India

ISBN: (纸本)9783031810824

Cyberbullying is a pressing issue in the digital age, especially in non-English-speaking communities. This study focuses on using deep learning techniques to detect cyberbullying in the Bangla language. The research employs a Bangla dataset that includes both cyberbullying and non-cyberbullying text, which undergoes preprocessing, tokenization, and sequence transformation for input into a LSTM recurrent neural network. The LSTM model incorporates word embedding and dropout techniques to enhance its performance. Evaluation metrics like accuracy and AUC-ROC score are employed to assess the model's effectiveness. The study investigates many methods, such as vernacular-specific preparation and data augmenting, that are specifically designed for the identification of cyberbullying in the Bangla language. The findings demonstrate deep learning's capacity to identify cyberbullying in Bangla, which makes a significant contribution to resolving this problem in non-English settings. The research reveals impressive accuracy rates, with LSTM achieving 99.99% accuracy, and a different approach, Convolutional Neural Network (CNN), yielding 99.022%. Ensemble models, including Gradient Boosting, XGBoost, and Random Forest, are also employed and optimized using hyperparameter tuning. Notably, LSTM stands out as the most reliable classifier for cyberbullying detection, as confirmed by the experimental findings and recent research. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Cyberbullying Deep Learning Dropout Embedding

来源：评论

学校读者我要写书评

暂无评论

Knock Detecting Door Unlock System using IOT

Knock Detecting Door Unlock System using IOT

引用

Intelligent and Innovative Technologies in computing, Electrical and Electronics (IITCEE), international conference

作者： Divyam Arora Anika Kirana Venkat Sudarshan Manjunath G Asuti Dept. of ECE B N M Institute of Technology Bengaluru India

ISBN: (数字)9798331515911

ISBN: (纸本)9798331515928

This paper presents the design and implementation of a Knock Detecting Door Lock System, aimed at enhancing security by eliminating traditional lock and key mechanisms. The proposed system utilizes an ESP32-based setup to detect specific knocking patterns, ensuring that only authorized individuals can unlock the door. The system addresses common security issues associated with key duplication and passcode vulnerabilities by relying on a personalized knocking sequence. Key features include the ability to set, and update knocking patterns, detect discrepancies, and maintain a balance between security and convenience. The hardware components include an ESP32, piezo sensors, and a servo motor, while the software is developed using the Arduino IDE. It can also incorporate two-factor authentication by generating a One-Time Password (OTP) when a valid knocking pattern is recognized. The OTP is sent to the user’s mobile device as an additional verification step before access is granted. By integrating the knock-based mechanism with the OTP feature, this system offers a robust, user-friendly solution that addresses both analog and digital security needs. The system’s innovative approach offers a cost-effective, efficient, and secure solution for modern door locking needs, with potential applications in various settings, including residential and commercial properties.

关键词： Vibrations Accuracy Authentication Sensor systems software pattern recognition Security Intelligent sensors Smart phones Testing

来源：评论

学校读者我要写书评

暂无评论

6th international conference on Smart computing and Informatics, SCI 2024

6th International Conference on Smart Computing and Informat...

引用

6th international conference on Smart computing and Informatics, SCI 2024

ISBN: (纸本)9789819619801

The proceedings contain 46 papers. The special focus in this conference is on Smart computing and Informatics. The topics include: Resilient Domain Authentication Framework for Enhancing Digital Identity Security;traffic Sign Detection with pattern recognition Techniques Using Image Processing;exploring Advanced Techniques in Natural Language Processing and Machine Learning for In-depth Analysis of Insurance Claims;Network Intrusion Detection with SMOTE-ENN and Deep Learning Techniques;leveraging Transfer Learning to Enhance Location Accuracy in Mapping Services: A Case Study of Google Maps;assessment of Enhanced Email Spam Detection System Through Machine Learning Algorithms;machine Learning Methods for Predicting Traffic Congestion Forecasting;hybridization of Computational Intelligence Algorithm for Scheduling of Tasks and Balancing of Load in Cloud Network;MDSV: Mobs Detection by Enhanced Fused Feature Base Deep Neural Network from Surveillance Camera;ioT-Based Solution for Enhanced Tracking of Individuals Living with Dementia;a Novel Task Scheduling Algorithm in Heterogeneous Multi-cloud Environment;Evaluating the Integration and Usage of AI in Higher Education;evaluating the Connectional Benefits of Artificial Intelligence in the Digital Classroom;Influence of AI as an Aspect of Modern Education Era in Present World;Hilbert–Huang Transform Framework-Based Email and SMS Spam Detection;the Advancement and Utilization of Artificial Intelligence and Machine Learning in the Financial Industry and Its Impact on Macro and Microeconomics;analysis on the Cutting-Edge Approach to Assess Artificial Intelligence’s Educational Consequences in Contemporary Studies;data Analytics in Sales and Marketing: A Comprehensive Methodology for Business Analysts;Wireless Energy Transfer for UAV (Drone) Using Machine Learning.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Effective Early Hypothyroid Disease Prediction Using Traditional and Ensemble Machine Learning Algorithms 15th

Effective Early Hypothyroid Disease Prediction Using Traditi...

引用

15th international conference on soft computing and pattern recognition, SoCPaR 2023 and 14th World Congress on Nature and Biologically Inspired computing, NaBIC 2023

作者： Mridul, Aunik Hasan Ahsan, Nowreen Saleh, Md. Abu Asif, Asifuzzaman Afrose, Sonia Computer Science and Engineering Daffodil International University Dhaka Bangladesh Department of Data Science Clarkson University New York United States Computer Science and Engineering Lovely Professional University Punjab India Nutrition and Food Engineering Daffodil International University Dhaka Bangladesh

ISBN: (纸本)9783031810794

In recent years, there has been a noticeable rise in the prevalence of physical ailments, most notably hypothyroidism, a condition that has garnered substantial attention due to its substantial impact on a significant portion of the population. To gain a comprehensive understanding of the disease’s severity, distinguishing between standard and affected diagnostic reports is imperative. In this work, we suggest using algorithmic models to facilitate early identification and increase knowledge of possible health hazards linked to hypothyroidism. Our approach is straightforward and well-suited for the prediction of uncomplicated cases of hypothyroid illness in real-world scenarios. Our dataset, sourced from Kaggle, served as the foundation for our research, which involved the utilization of a variety of classifiers, including RF, LR, GB, KN, ABC, DT, GS, and ensemble techniques. The DT classifier proved to be the most accurate, with an astounding accuracy rate of 98.278%. The findings were quite successful. In addition, the accuracy of the Voting Classifier RDGL was 97.218%, while the Stacking Classifier RDAS demonstrated an accuracy of 97.483%. Our optimization efforts, which included hyperparameter tuning, further enhanced the performance of each classifier. Based on extensive experimentation and a review of contemporary research, our findings unequivocally endorse the Decision Tree (DT) boosting classifier as exceptionally proficient, demonstrating a remarkable accuracy rate of 98.278% in the precise prediction of hypothyroid disease. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Bagging Boosting Ensemble Hypothyroidism

来源：评论

学校读者我要写书评

暂无评论

A Unified Framework for Face Re-Identification Using Entry-Exit Cameras

A Unified Framework for Face Re-Identification Using Entry-E...

引用

computing for Sustainability and Intelligent Future (COMP-SIF), international conference on

作者： Suhas A Shashi Kumar S S Suraj Varma Praveen Kumar B G RNS Institute of Technology Bengaluru Karnataka India Schmid Vision Soft Bengaluru Karnataka India

ISBN: (数字)9798331538538

ISBN: (纸本)9798331538545

Robust re-identification of individuals across multiple cameras has become a fundamental requirement in modern security and surveillance. To address this, we propose an effective system that ensembles detection, tracking, feature extraction, and retrieval mechanisms into a unified pipeline. By leveraging lightweight models, the system minimizes computational overhead and functions effectively on surveillance cameras. The system architecture enables processing and reliable identification, making it suitable for real-time applications. Our approach involves logging entry/exit registries and matching exits with entries through a re-identification pipeline, which allows security personnel to monitor an individual's entry and exit times within a secured premise. We are also releasing an open-source test video dataset with annotations, which is helpful for evaluating the face re-identification approaches for entry-exit cameras. Our approach performed good while considering matching of exit and entry events. The registered persons were identified with much better accuracy considering entry/exit events.

关键词： Accuracy Annotations Face recognition Surveillance Computational modeling Pipelines Streaming media Cameras Real-time systems Security

来源：评论

学校读者我要写书评

暂无评论

Techniques for Influencer Detection, Mapping, and Popularity Prediction in Social Media Networks

Techniques for Influencer Detection, Mapping, and Popularity...

引用

computing for Sustainability and Intelligent Future (COMP-SIF), international conference on

作者： Chandana T M Sreekanth Rallapalli Department of Master of Computer Applications Visveswaraya Technology University Nitte Meenakshi Institute of Technology Bengaluru Karnataka Department of Master of Computer Applications Nitte Meenakshi Institute of Technology Bengaluru Karnataka

ISBN: (数字)9798331538538

ISBN: (纸本)9798331538545

This paper recognizes the fact that in an age where consumption patterns are closely linked with social media platforms the measurement of key influential personalities and their effects on consumers is crucial. This paper offers an extensive analysis of the state-of-the-art methods for influencer identification and influence mapping. Approaches used in this research are based on Machine Learning and Deep Learning Algorithms which include Graph Attention Networks, Gated Recurrent Units, Graph-based Long Short-Term Memory Neural Networks, Recurrent Neural Networks, Stacked Autoencoders, Centrality Measures, and Frameworks such as Stimulus Organism Response. Comparing all of these approaches reveals their role in improving influencer marketing approaches, targeting, and the level of engagement, and profitability. The current study will benefit this area of knowledge by showing the outcomes, performance, and computational strengths of these approaches, paving the way for the enhancement of influencer marketing technology.

关键词： Deep learning Machine learning algorithms Consumer behavior Social networking (online) Profitability Current measurement Organisms pattern recognition Sustainable development Long short term memory

来源：评论

学校读者我要写书评

暂无评论

Optical Observation for Body Angles of Cricket Players’ Motion to Identify the Batting Shots

Optical Observation for Body Angles of Cricket Players’ Mot...

引用

Advanced Research in computing (ICARC), international conference on

作者： Ishan Jayaweerage Ashan Induranga Kaveendra Maduwantha Samantha Kumara Faculty of Computing Sabaragamuwa University of Sri Lanka Belihuloya Sri Lanka Faculty of Technology Sabaragamuwa University of Sri Lanka Belihuloya Sri Lanka

ISBN: (数字)9798331530983

ISBN: (纸本)9798331530990

Cricket, one of the world’s most popular sports, has developed considerably during the past few decades. The competition and complexity of this sport increased largely during the past few years, which brought the attention of the researchers to develop technical solutions to improve the quality of the sport by improving both the on-field decision-making and practicing methods. There are several kinds of high-end equipment available in the market that can be used to improve these skills. Even though such equipment can be used at the national level, it is very difficult to use them at school and club levels due to their higher cost. This research focused on enhancing the batting skills of cricket batsmen using a low-cost setup that uses webcams to capture the footage and the MediaPipe library, which was developed by Google to analyze motion data. This research analyzed two cricket shots, the front foot defence, and the pull shot, using four web cameras. The received footage was analyzed using MediaPipe to identify the variations in the angles of the player’s left elbow and left shoulder while playing the shot. These pattern identifications can help successfully understand the angle variation of the forward foot defence and the pull shots by generating an average angle value graph using four web cameras. These graphs show the special patterns for each batting shot. This methodology and the results are valuable for developing a successful player analyzing system by considering the other important cricket shots at a very low cost with affordable 2D web cameras.

关键词： Integrated optics Costs Webcams Ultraviolet sources Lighting Machine learning Optical imaging pattern recognition Sports Foot

来源：评论

学校读者我要写书评

暂无评论

Optimizing Human Motion Posture Detection and Humanized computing for Smart Health Applications

Optimizing Human Motion Posture Detection and Humanized Comp...

引用

Digital Analysis and Processing, Intelligent Computation (DAPIC), international conference on

作者： Weiwei Wang Jingshan Zhang School of Sports Medicine and Rehabilitation North Sichuan Medical College Nanchong Sichuan China Department of Sports Humanities and Social Sciences Sichuan Sports College Chengdu Sichuan China

ISBN: (数字)9798331536169

ISBN: (纸本)9798331536176

Existing human motion detection technologies have problems such as insufficient accuracy and poor real-time performance, which makes it difficult to meet the needs of health management. This paper proposes a human motion posture detection and humanized calculation method based on multi-sensor fusion and LSTM. This paper first deploys IMU sensors at key parts of the human body, and uses visual sensors to capture the three-dimensional posture information of human motion, and obtains physiological data such as heart rate and number of steps through wearable devices such as smart bracelets. Secondly, the IMU data is denoised and calibrated to eliminate sensor errors. Next, the fused data is trained using a long short-term memory network (LSTM). The rationality of the motion posture is then analyzed, and the user's exercise intensity and health status are evaluated based on the motion data and physiological data. In terms of posture recognition accuracy, the average value of this method is 92.61 %. In terms of abnormal posture detection rate, the detection rate of some data points of this method is close to 100%. At the same time, the maximum recognition time of this method is only 789 milliseconds, and the minimum recognition time is as low as 503 milliseconds, which is better than the traditional method. The optimized human motion posture detection and humanized calculation method proposed in this paper realizes high-precision detection of human motion and intelligent health management through multi-source data fusion, LSTM construction, human kinematics analysis and personalized health assessment.

关键词： Training Accuracy soft sensors Data integration Real-time systems Physiology Data models Long short term memory Intelligent sensors Wearable sensors

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 << < 6 7 8 9 10 11 12 13 14 15 > >>

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：