检索结果-内蒙古大学图书馆

3rd international conference on Machine Vision and Augmented Intelligence, MAI 2023

作者： Althaf, Patan Vinta, Surendra Reddy School of Computer Science and Engineering VIT-AP University Andhra Pradesh Amaravati India

ISBN: (纸本)9789819743582

This research aims to integrate TrOCR, an advanced Optical Character Recognition (OCR) technology, with the Langchain framework for Document question answering on image-based queries. The college has many number of students and they face many problems when it comes to the preparation for exams. There may be many question papers they have to prepare manually by searching the answers in books. Our model uses TrOCR to extract text from input images and the large text data is split using Langchain. Vector indexing is to process the text. This integration helps in information retrieval and question answering. The VectorstoreIndexCreator facilitates text vectorization and indexing, and HuggingFaceEmbeddings for semantic understanding. This research aims for information retrieval, document analysis, and natural language processing, offering a connection between image-based text and advanced language technology where users can have easy handling of the model and provides comprehensive answers for given complex questions. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Question answering

来源：评论

学校读者我要写书评

暂无评论

Optimized FLOPs-Aware Knowledge Distillation for TinyML Applications in Agriculture

Optimized FLOPs-Aware Knowledge Distillation for TinyML Appl...

引用

2025 international conference on Advancement in Data science, E-learning and information System, ICADEIS 2025

作者： Alba, Arjay Villaverde, Jocelyn Lacara, Amor Domingo, Jovita Aguirre, Dennis College of Engineering Bulacan State University School of Graduate Studies School of Electrical Electronics and Computer Engineering Mapua University Manila Philippines School of Graduate Studies School of Electrical Electronics and Computer Engineering Mapua University Manila Philippines College of Engineering Bulacan State University Malolos Philippines

ISBN: (纸本)9798331513320

This study introduces a novel FLOPs-Aware Knowledge Distillation (FAKD) framework tailored for TinyML applications in agriculture. By integrating Floating Point Operations (FLOPs) as a regularization term into the distillation loss, the framework ensures computational efficiency while preserving accuracy. The teacher-student architecture employs ResNet-50 and MobileNetV2, respectively, with the latter optimized for edge deployment. The methodology includes advanced techniques such as pruning and INT4 quantization, enabling the model to fit stringent hardware constraints, such as the ESP32's 1MB PSRAM, while maintaining high test accuracy. Experimental results on the PlantVillage dataset reveal a significant accuracy improvement from 92.77% to 96.55% post-FAKD, followed by a tradeoff to 90.15% after optimization for deployment. Furthermore, the study compares the performance of the optimized model across different devices, demonstrating a balance between accuracy, inference time, and power consumption. The results validate the framework's potential for deploying efficient and accurate models on low-power devices in resource-constrained environments, making it suitable for smart agriculture applications. This work underscores the importance of balancing model performance and computational efficiency, paving the way for sustainable TinyML solutions in agriculture. © 2025 IEEE.

关键词： Grain (agricultural product)

来源：评论

学校读者我要写书评

暂无评论

Event information Delivery Method Using Location information for Route Replanning in V2I Local Environment 27

Event Information Delivery Method Using Location Information...

引用

27th international conference on advanced Communications Technology, ICACT 2025

作者： Hong, MinHwa Choi, WonSeok Choi, SeongGon The Depart of Electrical and Computer Engineering Chungbuk National University Chungdae-ro Seowon-gu Cheongju-city Chungcheongbuk-do28644 Korea Republic of

ISBN: (纸本)9791188428137

We propose a new method that transmits a frame containing event location information in V2I LAN environments upon the occurrence of an event. In traditional navigation systems, external servers connected via the internet analyze traffic conditions and calculate estimated travel times based on real-time traffic information, GPS data, and user information. The results are then transmitted to the in-vehicle navigation system. However, these systems have limitations in immediately reflecting delays caused by real-time accidents in route settings. We propose an event information delivery method that quickly triggers a change to a new route by transmitting event information (such as accidents or blockages) within the local area, without communicating with servers in external networks. In addition, we describe a packet format in which location information is included in the header to reduce the delay and algorithms applied to the proposed method for transmitting event information from the AMU and replanning route based on event information in the AMU. To evaluate the performance of this new system, we calculate the end-to-end delay in the worst case and assess the effectiveness of the proposed system based on this. Copyright 2025 Global IT research Institute (GIRI). All rights reserved.

关键词： Travel time

来源：评论

学校读者我要写书评

暂无评论

The Applications and research of NLP and Deep Learning Technologies in Public Security Data 24

The Applications and Research of NLP and Deep Learning Techn...

引用

8th international conference on Electronic information Technology and computer engineering, EITCE 2024

作者： Zhu, Yupeng Zhu, Yuhao College of Information Engineering Hainan Vocational University of Science and Technology Hainan Haikou China The Smart City Research Institute of China Electronics Technology Group Corporation Guangdong Shenzhen China

ISBN: (纸本)9798400710094

With the deepening reform and construction of the police system, it is imperative to fully utilize technologies such as artificial intelligence, big data, blockchain, cloud computing, and the Internet of Things to rapidly enhance the intelligence level of the new generation command center. Natural language processing and deep learning technologies can achieve recognition, understanding, and correlation analysis of public security textual information, which are core steps in building intelligence. Due to the diversity of information in the public security field, the accuracy of these technologies still requires further research. In the analysis of public security incidents, the application of NLP technology faces challenges in optimizing technical models and algorithms, which greatly enhances the efficiency and accuracy of public safety management by analyzing, understanding, predicting, and responding to various security challenges in an intelligent manner. With the advent of the big data era, the public security sector has accumulated massive amounts of police incident data, which contain rich information but also bring difficulties in processing and analysis. Traditional NLP models often struggle to cope with the complex and ever-changing police incident texts, leading to compromised accuracy and reliability of the analysis results. This paper will explore the key technological applications of NLP and deep learning in the field of public safety and provide a review of their research progress. © 2024 Copyright held by the owner/author(s).

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Revolutionizing Energy Storage Management with advanced Battery Algorithms 3

Revolutionizing Energy Storage Management with Advanced Batt...

引用

3rd IEEE international conference on Automation and Computation, AUTOCOM 2025

作者： Bansal, Shubhansh Hussain, Irshed Kalra, Hitesh Manjunath, H.R. Vijaya Krishna, V.V.S. Yadav, Rakesh Kumar Chitkara University Centre for Interdisciplinary Research for Business and Technology Centre of Research Impact and Outcome Punjab Rajpura140417 India Department of Computer Science and Information Technology Odisha Bhubaneswar India Chitkara University Chitkara Centre for Research and Development Himachal Pradesh Baddi174103 India Karnataka Bangalore India Presidency University Department of Electronics and Communication Engineering Karnataka Bangalore India Maharishi School of Engineering & Technology Maharishi University of Information Technology Uttar Pradesh India

ISBN: (纸本)9798331542375

The growing demand for renewable electricity resources has improved the need for green and reliable electricity storage structures. Batteries have emerged as one of the number one solution for storing renewable electricity. However, the overall performance and lifespan of those batteries are limited due to factors including temperature fluctuations, charging and discharging costs, and aging. To address those demanding situations, there's a need for advanced battery management algorithms that can optimize battery performance while ensuring safety and prolonging its lifespan. This assignment aims to revolutionize electricity garage management with the aid of growing superior battery algorithms, which could, as it should be, eliminate battery degradation and optimize performance based on actual-time information. These algorithms will contain machine getting-to-know strategies to evolve to convert environmental situations and usage styles, thereby maximizing battery performance and storage capacity. Additionally, those algorithms permit the mixing of different types of batteries into an unmarried device, considering a more flexible and green energy storage answer. The implementation of these advanced battery algorithms is predicted to have a considerable impact on the renewable power region, making it extra dependable, price-powerful, and sustainable. It can boost the adoption of renewable power sources and make contributions to the worldwide intention of lowering carbon emissions. © 2025 IEEE.

关键词： Battery storage

来源：评论

学校读者我要写书评

暂无评论

Theoretical study of a dual-wavelength laser amplifier based on intracavity pumping technology 4

Theoretical study of a dual-wavelength laser amplifier based...

引用

4th international conference on advanced Manufacturing Technology and Electronic information, AMTEI 2024

作者： Xu, Jian Jin, Long Jin, Yushi Wang, Zijian Dong, Yuan Jilin Key Laboratory of Solid Laser Technology and Application School of Science Changchun University of Science and Technology Changchun130022 China

ISBN: (数字)9781510688247

ISBN: (纸本)9781510688230

We developed a novel theoretical model of a dual-wavelength laser amplifier based on intra-cavity pumping technology. In this model, we derived analytical expressions for the output power of the dual-wavelength laser amplifier. Simulations were conducted to explore the output characteristics of the dual-wavelength amplifier based on intra-cavity pumping technology. The effects of amplifier crystal length and beam radius on the laser's output performance were investigated, providing a solid foundation for further research. © 2025 SPIE.

关键词： Solid state lasers

来源：评论

学校读者我要写书评

暂无评论

Path Planning Technology for Unmanned Aerial Vehicle Swarm Based on Improved Jump Point Algorithm

引用

international Journal of advanced computer science and Applications 2025年第4期16卷 250-260页

作者： Zhang, Haizhou Xu, Shengnan School of Information Engineering Henan Vocational University of Science and Technology Henan Zhoukou466000 China

Multi-unmanned aerial vehicle path planning encounters challenges with effective obstacle avoidance and collaborative operation. The study proposes a swarm planning technique for unmanned aerial vehicles, based on an improved jump point algorithm. It introduces a geometric collision detection strategy to optimize path search and employs the dynamic window method to constrain the flight range. Additionally, the study presents conflict avoidance strategies for multi-unmanned aerial vehicle path planning and establishes collision fields for unmanned aerial vehicles to achieve collaborative path planning. In single unmanned aerial vehicle path planning, the research model exhibits the lowest control errors in the X, Y, and Z axes, with the Y-axis error being 0.05m. In static planning, the model boasts the shortest planning time and length, with 1002ms and 17.85m in multi-obstacle planning, respectively. In multi-unmanned aerial vehicle path planning, the research model effectively avoids local optimal problems in local conflict scenarios and re-plans the route. During testing on a 29m×29m grid map, the research technology successfully avoids obstacles and re-plans routes. However, similar technological obstacles can cause interference and traps in local convergence, preventing re-planning. The research technology demonstrates good application effects in the path planning of unmanned aerial vehicle swarms and will provide technical support for multi-machine collaborative path planning. © (2025), (science and information Organization). All rights reserved.

关键词： Drones

来源：评论

学校读者我要写书评

暂无评论

Photovoltaic Fault Detection in Remote Areas Using Fuzzy-Based Multiple Linear Regression (FMLR)

引用

international Journal of advanced computer science and Applications 2025年第4期16卷 243-249页

作者： Ardianto, Feby Ermatita, Ermatita Sofijan, Armin Doctoral Program in Engineering Science Universitas Sriwijaya Palembang Indonesia Faculty of Computer Science Universitas Sriwijaya Palembang Indonesia Faculty of Engineering Universitas Sriwijaya Palembang Indonesia

This research focused on developing and implementing a fault detection model for photovoltaic (PV) systems in remote areas, utilizing a Fuzzy-Based Multiple Linear Regression (FMLR) approach. The study aimed to address the challenges of monitoring PV systems in locations with limited access to conventional power grids and technical resources. The fault detection system integrated environmental parameters such as solar radiation, temperature, wind speed, and rainfall, alongside PV system parameters like panel voltage, current, battery voltage, and inverter performance. Data collection and preprocessing were conducted over a specified period to identify operational patterns under both normal and faulty conditions, ensuring data accuracy through cleaning, normalization, and categorization. The research was conducted in Pandan Arang Village, Kandis District, Ogan Ilir Regency, South Sumatera, Indonesia, contributing to the improvement of reliability and sustainability of renewable energy sources in isolated communities. The total number of data points for 276 rows with 6 attributes each was 1656 records. The MLR model was developed to predict the output power of the PV system, while fuzzy logic was employed to handle uncertainties in the data, offering a more flexible and adaptive decision-making process. The system applied fuzzy rules to determine the charging status (P3), categorizing it into Optimal Charging, Adjusted Charging, Charging Delay, or Fault Alert. The model was tested with real-time data, and its performance was validated through comparison with manual inspections. The results showed that the FMLR-based fault detection system effectively identified faults and optimized the performance of the PV system, making it suitable for remote areas in South Sumatera. © (2025), (science and information Organization). All Rights Reserved.

关键词： Multiple linear regression

来源：评论

学校读者我要写书评

暂无评论

Improving Automated Audio Captioning with LLM Decoder and BEATs Audio Encoder 24

Improving Automated Audio Captioning with LLM Decoder and BE...

引用

8th international conference on computer science and Artificial Intelligence, CSAI 2024

作者： Liu, Weizhuo Gao, Zhe The College of Information Mechanical and Electrical Engineering Shanghai Normal University ShangHai China

ISBN: (纸本)9798400718182

Automated Audio Captioning (AAC) focuses on producing detailed descriptions for a wide range of sounds, including those from nature and human activities. Recently, this field has garnered significant research attention, with the latest advancements in AAC systems largely driven by the use of sequence-to-sequence (seq2seq) frameworks. These frameworks are increasingly powered by powerful models like Transformers, which have become the backbone of state-of-the-art AAC approaches. We propose an advanced Automated Audio Captioning (AAC) system that leverages the LLaMA3.1 decoder for robust language generation and the BEATs encoder for fine-grained acoustic feature extraction. These components are connected via a Q-Former intermediate layer, ensuring seamless audio-to-text alignment. To overcome the challenge of limited training data, we employ a sophisticated data augmentation strategy using Qwen, enhancing the model’s generalization and resilience. A multi-step training process with low-rank adaptation (LoRA) further optimizes performance. Our method obtains a 28.5 SPIDEr-FL score, outperforming the participates of DCASE 2023 Task 6A. Our code will be stored on GitHub. © 2024 Copyright held by the owner/author(s).

关键词： Signal encoding

来源：评论

学校读者我要写书评

暂无评论

DataTransfer PDU-based Rollback Mechanism for Securing OCPP 1.6 Against Spoofing Attacks 27

DataTransfer PDU-based Rollback Mechanism for Securing OCPP ...

引用

27th international conference on advanced Communications Technology, ICACT 2025

作者： Kim, Dong Uk Kim, Keon Oh Kang, Sun-Moo Hong, Choong Seon Department of Computer Science and Engineering Kyung Hee University 446-701 Korea Republic of

ISBN: (纸本)9791188428137

The Open Charge Point Protocol is a widely adopted communication standard that facilitates interoperability between Charge Points and Central Systems in the electric vehicle charging ecosystem. This paper explores the security implications of OCPP 1.6, with a particular focus on the risks posed by spoofed or unauthorized messages. We propose a rollback mechanism utilizing the DataTransfer PDU to mitigate these risks, allowing the Central System to trigger reversion of unauthorized changes at the CP level. The proposed solution ensures secure, efficient management of configurations and mitigates potential threats such as replay attacks, spoofed requests, and unauthorized modifications. This paper contributes to the growing body of knowledge on securing EV charging infrastructure by highlighting underexplored vulnerabilities in OCPP 1.6 and providing practical solutions to enhance the security and resilience of the protocol. Copyright 2025 Global IT research Institute (GIRI). All rights reserved.

关键词： State of charge

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：