检索结果-内蒙古大学图书馆

27th International Conference on Pattern recognition, ICPr 2024

作者： Jena, Sushovan Saini, Vishwas Shaw, Ujjwal Jain, Pavitra raihal, Abhay Singh Banerjee, Anoushka Joshi, Sharad Ganesh, Ananth Bhavsar, Arnav School of Computing and Electrical Engineering Indian Institute of Technology Mandi Himachal Pradesh Mandi175005 India R&D Center Hitachi India Pvt. Ltd. Bengaluru560055 India

ISBN: (纸本)9783031781681

Unsupervised anomaly detection encompasses diverse applications in industrial settings where a high-throughput and precision is imperative. Early works were centered around one-class-one-model paradigm, which poses significant challenges in large-scale production environments. Knowledge-distillation based multi-class anomaly detection promises a low latency with a reasonably good performance but with a significant drop as compared to one-class version. We propose a dCAM (distributed Convolutional Attention Module) which improves the distillation process between teacher and student networks when there is a high variance among multiple classes or objects. Integrated multi-scale feature matching strategy to utilise a mixture of multi-level knowledge from the feature pyramid of the two networks, intuitively helping in detecting anomalies of varying sizes which is also an inherent problem in the multi-class scenario. Briefly, our dCAM module consists of Convolutional Attention blocks distributed across the feature maps of the student network, which essentially learns to masks the irrelevant information during student learning alleviating the "cross-class interference" problem. This process is accompanied by minimizing the relative entropy using KL-divergence in Spatial dimension and a Channel-wise Cosine Similarity between the same feature maps of teacher and student. The losses contributes to achieve scale-invariance and capture non-linear relationships. We also highlight that the dCAM module would only be used during training and not during inference as we only need the learned feature maps and losses for anomaly scoring and hence, gaining a performance gain of 3.92% than the multi-class baseline with a preserved latency. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

A vision system based on CNN-LSTM for robotic citrus sorting Author links open overlay panel

引用

Information Processing in Agriculture 2024年第1期11卷 14-25页

作者： Yonghua Yu Xiaosong An Jiahao Lin Shanjun Li Yaohui Chen College of Engineering Huazhong Agricultural UniversityWuhan 430070China Key Laboratory of Agricultural Equipment in Mid-Lower Yangtze River Ministry of Agriculture and Rural AffairsWuhan 430070China National R&D Center for Citrus Preservation Wuhan 430070China Citrus Mechanization Research Base Ministry of Agriculture and Rural AffairsWuhan 430070China

Compared with manual sorting of citrus fruit,vision-based sorting solutions can help achieve higher accuracy and *** this study,we present a vision system based on CNN-LSTM,which can cooperate with robotic grippers for real-time sorting and is readily applicable to various citrus processing plants.A CNN-based detector was adopted to detect the defective oranges in view and temporarily classify them into corresponding types,and an LSTM-based predictor was used to predict the position of the oranges in a future frame based on image sequential *** fusion of CNN and LSTM networks enabled the system to track defective ones during rotation and identify their true types,and their future path was also predicted which is vital for predictive control of visually guided robotic *** detection accuracy of 94.1%was obtained based on experimental results,and the error for path prediction was within 4.33 pixels 40 frames *** average time to process a frame was between 28 and 62 frames per second,which also satisfied real-time *** results proved the potential of the proposed system for automated citrus sorting with good precision and efficiency,and it can be readily extended to other fruit crops featuring high versatility.

关键词： deep learning Long short-term memory Vision system Online citrus sorting Path prediction

来源：评论

学校读者我要写书评

暂无评论

Transformative Approaches to Customer Sentiment Analysis and Customer Feedback Scoring in CrM Platforms 8

Transformative Approaches to Customer Sentiment Analysis and...

引用

8th International Artificial Intelligence and data Processing Symposium, IdAP 2024

作者： Cevik, rabia Celik, Ahmet Erkan Akbulut, Akhan R&d Center of Next4biz Sahrayicedit Mah Pakpen Plaza No:40/4 Istanbul34734 Turkey Istanbul Kültür University Department of Computer Engineering Istanbul34536 Turkey

ISBN: (纸本)9798331531492

This study introduces an innovative system designed to predict customer satisfaction scores through the integration of sentiment analysis of customer feedback alongside all related factors from a Customer relationship Management (CrM) system. The system implements the latest transformer models like BErT and roBErTa then assess customer sentiment using an ensemble learning voting mechanism for accurate sentiment classification, and adaptive customer satisfaction rating. The model generates baseline scores dynamically, based on factors like customer loyalty, and frequency of interactions with the firm, thus enhancing accuracy and relevance when assessing satisfaction. The system is also developed to utilize Turkish data optimizing usage in market shares for firms serving that user group. Empirical results indicate that the ensemble learning approach significantly improves the accuracy of sentiment analysis and the reliability of satisfaction quantification. This resource provides additional contribution to the CrM literature by providing a credible and scalable mechanism to assess customer satisfaction to potentially be implemented in practice across industries. Future work will focus on extending the system's scalability and enhancing its predictive capabilities across diverse sectors. © 2024 IEEE.

关键词： Customer satisfaction

来源：评论

学校读者我要写书评

暂无评论

A design and control method for Pole Changing Induction Machine for Traction Application

A design and control method for Pole Changing Induction Mach...

引用

2024 IEEE Energy Conversion Congress and Exposition, ECCE 2024

作者： Akatsu, Kan Kato, Naoya Masayoshi, Tsurube Mori, Tomonari Watahiki, Masanori Yokohama National University Division of Intelligent Systems Engineering Department of Faculty of Engineering Kanagawa Japan Yokohama National University Department of Electrical Engineering Kanagawa Japan Nidec Product Technology R&d Center Nidec Corporation Kanagawa Japan

ISBN: (纸本)9798350376067

This paper presents a design method of a Pole Changing Induction Machine (PCIM) with double 3 phase inverter. The winding method and a design of the machine geometry to maximize the advantage of the pole changing IM are shown. Also, a driving method of PCIM which can reduce the torque shock under the pole changing is described. Some simulation and experimental results show the effectiveness of PCIM and torque shock reduction under the high torque conditions. © 2024 IEEE.

关键词： Machine windings

来源：评论

学校读者我要写书评

暂无评论

A Prediction Method for Bike-sharing Time Series Based on K-means Clustering and Gaussian Process regression

A Prediction Method for Bike-sharing Time Series Based on K-...

引用

Computer engineering and Intelligent Communications (ISCEIC), International Symposium on

作者： Xuehao Liu Wenxue Liu Chaosan Yang Can Xu Wenjing Zhu Jinhai Li Communications and Information Engineering R&D Center IMECAS Beijing China

ISBN: (数字)9798331518677

ISBN: (纸本)9798331518684

real-time usage prediction for shared bikes has important significance in intelligent transportation and travel management. In this paper, a time series prediction method for shared bikes based on K-means clustering and Gaussian process regression (GPr) is proposed. The method transforms the shared bikes time series into multiple subsequences through sliding window method to ensure that the time information is not lost in the GPr prediction process. By applying K-means clustering to select inducing points, data sparsity is achieved, and the selected points are used in the training of GPr model to achieve real-time usage prediction. Experimental results indicate that the proposed method effectively captures short-term trends and periodic characteristics in the time series, and significantly reduce the computational complexity compared with the traditional GPr model, while improving the training efficiency and generalization ability of the model.

关键词： Training Measurement Adaptation models Accuracy Computational modeling Time series analysis Gaussian processes Predictive models real-time systems Computational complexity

来源：评论

学校读者我要写书评

暂无评论

6G Polymorphic Smart Network Architecture and Key Technologies 5

6G Polymorphic Smart Network Architecture and Key Technologi...

引用

5th International Conference on Frontiers Technology of Information and Computer, ICFTIC 2023

作者： Liao, Xingxing You, Wei Lv, Jingwen Wang, Min Zhao, Ying He, Gang Purple Mountain Laboratories Nanjing China National Digital Switching Engineering & Technological R&d Center Zhengzhou China Southeast University Nanjing China

ISBN: (纸本)9798350309034

In response to the problems of high architecture complexity, only support the IP, difficult against unknown threats, deficient intelligent in 5G network, a new polymorphic smart network architecture is proposed, including network system architecture and network deployment architecture. New network functions, such as polymorphic routing and addressing, endogenous security, and endogenous intelligent, are designed into the architecture. the polymorphic smart network selects corresponding network models for each terminal. SdN plans a network path for the network model that is associated with security and AI services. Finally, this article provides a simple validation of key technologies. © 2023 IEEE.

关键词： 6G architecture endogenous intelligent endogenous security polymorphic SdN

来源：评论

学校读者我要写书评

暂无评论

An Integrated Governance Technology Based on Magnetic Steering System for Cementing Plug in Open-Hole Section of Underground Gas Storage: Theoretical Analysis and Field Test

An Integrated Governance Technology Based on Magnetic Steeri...

引用

2024 SPE Conference at Oman Petroleum and Energy Show, OPES 2024

作者： Che, Yang Yuan, Guang Jie He, Ai Guo Qiao, Lei Liu, Tian En Fang, Ming Xing CNPC Engineering Technology R&D Company Limited Institute of Unconventional Oil and Gas Engineering Beijing China National Engineering Research Center for Oil & Gas Drilling and Completion Technology Beijing China China University of Petroleum Beijing China

ISBN: (纸本)9781959025252

The paper describes a comprehensive control technology based on a magnetic steering system for plugging the cementing plug in open holes without clear beacons in underground gas storage (UGS). Utilizing depleted oil and gas reservoirs for natural gas storage is an important approach towards achieving carbon neutrality. Plugging old wellbores that affect sealing performance is prioritized according to relevant standards. The challenge with open holes with cement plugs is that there are no clear beacons in the wellbore, and the extent of compaction is unknown. Accurate positioning of the cement plug is crucial and the current technical methods, such as electromagnetic and acoustic waves, have limitations. Magnetic steering technology, which is mature and widely used in drilling complex structure wells, requires beacons like pipes and metal pipe strings. On the other hand, acoustic positioning technology is still experimental and suffers from large distance and azimuth errors due to the attenuation characteristics of acoustic waves in heterogeneous formations. These existing technologies are unable to directly position open holes with cement plugs. The paper proposes a comprehensive control technology based on the magnetic steering system. It starts by measuring the positional relationship between the drilling wellbore and a magnetic reference wellbore using the magnetic steering system. The skew data is then used with the ISCWSA model to determine the location of the plugged wellbore and the magnetic reference wellbore. By calculating the relative positional relationship between the drilling wellbore and the plugged wellbore, the cement plug reentry can be achieved within the error ellipse range below the sidetracking point. After reentry, specific work such as crushing, element logging, caliper logging, and full hydrocarbon gas detection is performed to verify if the wellbore is still in the original position and if the caprock is properly squeezed and plugged. The c

关键词：

来源：评论

学校读者我要写书评

暂无评论

Sensing in Terahertz Communications for the Sixth-Generation Mobile Technology 39

Sensing in Terahertz Communications for the Sixth-Generation...

引用

39th International Conference on Information Networking, ICOIN 2025

作者： Muthukkumaran, Mmohaankumar S.-O. Alias, Mohamad Yusoff Bin Bujang, Suhandi Marzuki, Azah Syafiah Mohd Multimedia University Faculty of Engineering Selangor Cyberjaya Malaysia TM R&D Sdn Bhd Center of Excellent for Intelligent Network Cyberjaya Malaysia

ISBN: (纸本)9798331506940

This paper investigates the use of compressed sensing methods in waveform design to enhance transmission efficiency while minimizing free-space path loss (FSPL). It introduces a classification framework tailored for communication and sensing services, considering critical parameters such as data rates, operational frequencies, and latency requirements. Additionally, the research evaluates the potential of integrating Terahertz (THz) sensing into sixth generation (6 G) networks, utilizing the distinctive attributes of THz waves for applications like imaging, spectroscopy, and environmental sensing. The project demonstrated that waveforms developed with compressed sensing techniques improved both data acquisition and reconstruction. Simulation results reveal substantial challenges in THz communication for 6G networks, including significant path losses, atmospheric absorption, and scattering at higher frequencies and over longer distances. Addressing these challenges necessitates advanced technologies - such as high-gain antennas, power amplifiers, and robust error-correcting codes - to ensure reliable communication. These findings are essential for the design and optimization of 6G networks, providing guidance for the development of technologies and infrastructure that can meet the unique demands of THz communication, thereby supporting the ultra-high-speed, low-latency applications envisioned for future networks. © 2025 IEEE.

关键词： Terahertz spectroscopy

来源：评论

学校读者我要写书评

暂无评论

Enhancing rAG Pipeline Performance with Translation-Based Embedding Strategies for Non-English documents

Enhancing RAG Pipeline Performance with Translation-Based Em...

引用

2024 Innovations in Intelligent Systems and Applications Conference, ASYU 2024

作者： Iscan, Can Ozara, Muhammet Furkan Akbulut, Akhan R&D Center of Next4Biz Sahrayicedit Mah Pakpen Plaza No:40/4 Istanbul34734 Turkey Department of Computer Engineering Istanbul Kültür University Istanbul34536 Turkey

ISBN: (纸本)9798350379433

despite the advances in multilingual embedding models, they often underperform compared to embedding documents in English, which limits the effectiveness of retrieval-Augmented Generation (rAG) systems in non-English question-answering contexts. This disparity poses a significant barrier for non-English speakers to access high-quality AI-driven retrieval systems. To address this challenge, this paper proposes a novel approach that leverages English embeddings enhanced by translation. Our approach involves translating texts into English and then embedding them while preserving the original information as metadata. By utilizing robust English and multilingual embedding models, we achieve superior results in rAG systems. We present a comprehensive approach for constructing a question-answering system using GPT-4o technology, which leverages translation models to enhance English embeddings. We conducted actual studies to compare the suggested methodology with conventional methods using various embedding models, employing the rAGAS framework. Initial findings indicate that our method greatly improves the retrieval effectiveness of rAG systems in non-English environments, attaining higher context precision and recall metrics. This adaptable solution incorporates English embeddings into multilingual applications, providing a versatile and effective method to improve non-English scenarios by using the capabilities of translation models. © 2024 IEEE.

关键词： Question answering

来源：评论

学校读者我要写书评

暂无评论

Overview of Coastal Protection Structures in the Mekong river delta 4th

Overview of Coastal Protection Structures in the Mekong Riv...

引用

Proceedings of the 4th International Conference on Sustainability in Civil engineering - ICSCE 2022

作者： An, Le Vinh Thanh, Nguyen Viet Van Hai, Pham Lai, Trinh dinh Faculty of Civil Engineering University of Transport and Communications Hanoi Viet Nam Viet Trung R&D Center University of Transport and Communications Hanoi Viet Nam Faculty of Hydraulic Engineering Hanoi University of Civil Engineering Hanoi Viet Nam

ISBN: (纸本)9789819923441

TheMekong river delta has faced many impacts of climate change, sea level rise,and extreme climate. Besides,the exploitation and use of water resources and sand and gravel lead to seriousriverbank and coastal erosion. In recent years, many innovative technologies of coastal protection have been applied to protect coastal erosion. Inthis paper, based on the results of two field investigations from the coast ofTien Giang province to Ca Mau province, a review of coastal protection alsopoints out some advantages and disadvantages. The results indicated that the applied coastal protection technologies need to be further improved to limit the process of coastal erosion in the Mekong river delta. © 2024, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Sea level

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：