检索结果-内蒙古大学图书馆

5th International Conference on Image Processing and Capsule Networks, ICIPCN 2024

作者： Bharadwaj, Rakhi Borse, Rushikesh Ahire, Purvesh Aldhar, Prathamesh Hoshing, Atharva Vishwakarma Institute of Technology Department of Computer Engineering Pune India

ISBN: (纸本)9798350367171

In today's digital landscape, the demand for highly realistic 3D representations of human faces has surged, driven by the rapid expansion of virtual reality, augmented reality, and face recognition technologies. Despite significant strides in 3D facial geometry reconstruction from 2D images, achieving accurate and detailed translations of facial features and expressions remains a formidable challenge, critical for creating truly immersive digital experiences. In response, our research presents a groundbreaking approach to detailed reconstruction, specifically designed to convert single 2D images into intricately detailed and lifelike 3D models. Our methodology combines state-of-the-art techniques, such as UV displacement mapping and detail decoding, to meticulously capture even the subtlest nuances within 2D images, enhancing aspects like facial textures, sharp edges, and nuanced lighting effects. Furthermore, we employ advanced detail rendering techniques to further enrich the fidelity of the resulting 3D models, effectively narrowing the gap between digital representations and their real-world counterparts. Through the utilization of specialized loss functions, we meticulously optimize the reconstruction process, ensuring the faithful preservation of fine details while maintaining overall coherence and fidelity. This meticulous refinement promises to deliver immersive digital experiences that not only capture users' attention but also deeply engage them, offering a significant advancement in the realm of 3D facial reconstruction. © 2024 IEEE.

关键词： 3D reconstruction

来源：评论

学校读者我要写书评

暂无评论

Refined YOLOv8 for Enhanced Small Object Detection in PCB Defect

Refined YOLOv8 for Enhanced Small Object Detection in PCB De...

引用

2024 IEEE International Conference on Intelligent Signal Processing and Effective Communication Technologies, INSPECT 2024

作者： Raj, Abhishek Ahlawat, Priyanka National Institute of Technology Department of Computer Engineering Kurukshetra India

ISBN: (纸本)9798350379525

In the electronics industry, the timely and accurate detection of defects in printed circuit boards (PCBs) is crucial for maintaining high standards of product quality and efficiency. Tra-ditional methods often struggle with identifying small and subtle defects, leading to the need for more advanced approaches. This paper presents an enhanced version of the YOLOv8 model tai-lored for PCB defect detection. The proposed model incorporates several key modifications: the use of He initialization and batch normalization to improve network stability and convergence, and a learnable parameter for better feature fusion. The loss function was adapted to emphasize small object detection, which is critical for identifying minute defects. Extensive experiments on a comprehensive PCB defect dataset demonstrate that the modified YOLOv8 model significantly outperforms the standard version, offering a practical solution for real-time, high-precision defect detection in industrial applications. Experimental results demonstrate the model's superior performance, achieving a precision of 95.95%, a recall of 97.74%, and a mean Average Precision (mAP50) of 98.32%. These improvements make the enhanced YOLOv8 model highly effective for real-time PCB defect detection in industrial settings, offering a robust solution for quality assurance processes. © 2024 IEEE.

关键词： Electronics industry

来源：评论

学校读者我要写书评

暂无评论

Advancing Software-Defined Vehicles: An End-to-End Framework with Digital Twin Based Attestation for OTA Updates 17

Advancing Software-Defined Vehicles: An End-to-End Framework...

引用

17th International Conference on COMmunication Systems and NETworkS, COMSNETS 2025

作者： Agrawal, Krish Rohan, Jha Luthra, Nishkarsh Sekhar, Pilla Venkata Sharma, Hrishesh Banda, Gourinath Indian Institute of Technology Department of Computer Science and Engineering Indore India

ISBN: (纸本)9798331531195

Software-defined vehicles (SDVs) are an emerging paradigm in the automotive industry where vehicles' functionality, performance, and safety can be enhanced and updated through software, even after production. Unlike traditional vehicles, which rely primarily on physical components, SDVs leverage advanced connectivity, real-time data analytics, and cloud integration to adapt to changing regulations, driver preferences, and environmental conditions. This shift enables vehicles to evolve continuously, responding to new technological advances and customer needs. In this paper, we propose an end-to-end framework to demonstrate and realize the full benefits of SDVs. We incorporate the concept of digital twin (of the vehicle) driven software update authorization to download or update an application on the vehicle. In our framework, attestation refers to the verification of each software update's compatibility and functionality with the vehicle's current ECUs before deployment. Any impending/requested update is first verified for its compatibility with the vehicle's architecture (as per the twin) on the cloud. The idea is via the SDVs' digital twin - since each and every ECU is virtualized - the simulation and testing of the actual hardware setup with the new application software can be done without direct impact on the physical vehicle. Once the application is installed across all relevant virtualized ECUs, the framework confirms compatibility, ensuring smooth deployment and functionality in the real-world vehicle. Post this successful attestation, the installation will be done on the real vehicle corresponding to that digital twin. Through this proposed framework, we intend to ensure the safety of the updation, while the new updates contribute to the functionality and performance improvements. © 2025 IEEE.

关键词： Automotive industry

来源：评论

学校读者我要写书评

暂无评论

Dynamic Emotion-Adaptive Attention Mechanism (DEAAM) 5th

Dynamic Emotion-Adaptive Attention Mechanism (DEAAM)

引用

5th EAI International Conference on Cognitive Computing and Cyber Physical Systems, IC4S 2024

作者： Garugu, Suresh Kumar Madhuri, P. Shyamala Atchana, Venu Gopal Penmetsa, Gopala Krishnam Raju Bandaru, Venkata Naga Rani Sastry Eemani, V.R.J. Department of Computer Science and Engineering Vishnu Institute of Technology Bhimavaram India

ISBN: (纸本)9783031770807

The Dynamic Emotion-Adaptive Attention Mechanism (DEAAM) offers a groundbreaking framework for analyzing emotions in real-time video streams, leveraging a state-of-the-art convolution neural network (CNN) for rapid and accurate emotion detection. By integrating an adaptive attention mechanism, DEAAM dynamically adjusts video sampling rates and focuses on areas of emotional intensity, optimizing computational resources for the most expressive moments. This system intricately combines facial landmark detection with saliency mapping to pinpoint critical expressive features, enhancing the depth and precision of the analysis. Furthermore, DEAAM incorporates temporal emotional coherence tracking, utilizing advanced recurrent neural networks to capture the evolution of emotional states, adding a rich contextual layer to the emotion recognition process. This comprehensive approach not only increases the accuracy of emotion analysis but also ensures efficiency by focusing on key emotional expressions. Ideal for applications in virtual communication, mental health assessment, and any domain requiring nuanced emotion recognition, DEAAM sets a new standard in empathetic technology, offering a sophisticated, contextually aware system for extracting real-time emotional insights from video feeds. © ICST institute for computer Sciences, Social informatics and Telecommunications engineering 2025.

关键词： Recurrent neural networks

来源：评论

学校读者我要写书评

暂无评论

Efficient video summarization through MobileNetSSD: a robust deep learning-based framework for efficient video summarization focused on objects of interest

引用

Multimedia Tools and Applications 2024年 1-26页

作者： Yarrarapu, Manasa Leelavathy, Narkedamilly Haritha, Dasari Department of Computer Science and Engineering Jawaharlal Nehru Technological University Andhra Pradesh Kakinada533003 India Department of Computer Science and Engineering Godavari Institute of Engineering and Technology Andhra Pradesh Rajahmundry533296 India Department of Computer Science and Engineering University College of Engineering Jawaharlal Nehru Technological University Andhra Pradesh Kakinada533003 India

Now-a-days, the generation of videos has increased dramatically due to the quick growth of multimedia and the internet. The need for effective ways to store, manage, and index the massive numbers of videos has become imperative due to this expansion. As a result, a method needs to be proposed that collects only the necessary data from the original recording. In computer vision, Video summarization is a significant task, and its primary goal is to give a quick summary of the video by removing irrelevant information and capturing key frames from the video. Many approaches have developed over the last several decades, using the most recent deep neural network architectures that represent the current state-of-the-art. Our method involves extracting vital key frames from the input video using the MobileNetSSD model, which is well-known for its efficient recognition and localization of objects of interest. These highlighted frames are essential in creating a detailed video summary. Furthermore, a method of temporal analysis is applied to guarantee that the summary accurately reflects the relevant events in the order in which they occurred, contributing to a coherent and meaningful representation of the information. We evaluated the proposed approach on TV Sum and SUM me video datasets, comparing the results against cutting-edge video summarization techniques. Our approach works effectively to produce clear and meaningful video summaries. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2024.

关键词： Network architecture

来源：评论

学校读者我要写书评

暂无评论

Deep Convolutional Neural Network Implementation for Detecting Generative Adversarial Network Generated Deepfake Videos

Deep Convolutional Neural Network Implementation for Detecti...

引用

International Conference on Systems, Control and Automation, ICSCA 2023

作者： Gautam, Abhishek Singh, Awadhesh Kumar Department of Computer Engineering National Institute of Technology Haryana Kurukshetra India

ISBN: (纸本)9789819773831

Deep learning algorithms have become so powerful as computing power has increased that it has become much easier to create indistinguishable human-synthesized videos, dubbed DeepFake. It’s easy to imagine scenarios in which these realistic Deepfakes are used to commit political pranks, staged terrorist assaults, revenge pornography, and blackmail. This challenge describes a revolutionary deep learning-based technique for detecting genuine AI-generated videos from forgeries. Our method can automatically recognise surrogate and recurring deepfakes. We’re using artificial intelligence to try to counteract it. Our solution use a Deep Convolutional Neural Network to extract framelevel data and then train those features. A recurrent neural network with long-short-term memory used to determine if a picture has been edited. A huge set of balanced blended datasets was created by combining the numerous datasets available in order to recreate real- time events and enhance model performance using real-time data. A simple and dependable way for illustrating how a system might produce competitive performance. An interactive Graphical User Interface (GUI) is also developed for easy access. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Enhancing DNA Sequence Analysis for Disease Diagnosis and Prediction Using Parallelised Hash-Based Approach 2

Enhancing DNA Sequence Analysis for Disease Diagnosis and Pr...

引用

2nd International Conference on Computational Intelligence, Communication technology and Networking, CICTN 2025

作者： Sreeram, Anusha Bhukya, Raju Department of Computer Science and Engineering National Institute of Technology Warangal India

ISBN: (纸本)9798331530389

DNA sequencing is a critical tool in genetics, helping to identify disease-causing mutations, predict disease risks, screen for genetic diseases, and monitor disease progression. By determining the order of nucleotides that make up DNA, DNA sequencing allows for the identification of genetic information in specific DNA fragments, entire genomes, or complex microbiomes. It has transformed disease diagnosis and treatment and has applications in biotechnology, forensic science, and evolutionary biology. Comparing DNA sequences helps researchers identify genetic disorders, cancers, and antibody repertoires, leading to tailored treatments for individual patients. However, the diagnosis of diseases using DNA requires an efficient algorithm due to the large number of nucleotides in DNA sequences. The hash-based approach, an optimal string pattern-matching algorithm, is used to identify relevant patterns in DNA sequences. This approach has shown significant improvement in speed compared to the single pattern matching approach for finding multiple string patterns and a parallelized approach for hash based using GPU is also proposed. © 2025 IEEE.

关键词： DNA sequences

来源：评论

学校读者我要写书评

暂无评论

Optimized CNN Model to Develop a Decision-Making System for Spraying on Cotton Crop 3rd

Optimized CNN Model to Develop a Decision-Making System for ...

引用

3rd International Conference on Intelligent Systems and Applications, ICISA 2024

作者： Sudhir, Supriya Kumar, Kalyani S. Reddy, P. Santosh Department of Computer Science & Engineering BNM Institute of Technology Bangalore India

ISBN: (纸本)9789819798384

Dynamic crop morphology capture aids real-time precision in field spraying decisions. Ultrasonic radar and costly LiDAR can't maximize their high precision. They may slow down decision speed because the current spraying amount control needs to include more information about branches and leaves, which highly relies on the statistical features of regional morphology. Since most agricultural tools in China are medium in size and relatively inexpensive, it makes sense to create a unique identification system that combines deep learning with cheap monocular vision. A crop's growth cycle was a helpful gauge for direct spraying because of the knowledge gained. For this reason, let's examine cotton as an example: This study aims to develop a single-camera acquisition system and propose an algorithm based on the Convolutional Neural Network (CNN) to identify different growth stages of cotton plants. An optimal CNN model structure is chosen among nine possible method structures using an optimization procedure based on the matrix of confusion and recognition efficiency, and its dependability is established by repeatedly switching the training and test sets by the k-fold test concept. This CNN model has an accuracy of 94.27%, an F1-score of 96.39%, a recall of 95.76%, a precision of 95.31%, and a recognition speed of 72.46 ms per picture. Also, the convolution neural network, the model suggested in this study, performs better than those of VGG16 and Alex Net. Finally, spraying deposition tests were conducted over three distinct cotton growth stages to validate the accuracy of the developed identification method and the viability of the sprayed decision-making process based on CNN. The findings of the studies confirm that the ideal spray parameters were applied during distinct growth times correspondingly, leading to an increase of up to 63.24% in the optimum index over the operations that did not discriminate growth periods, yielding an overall increase of 43.29%. © The Author(

关键词： Cotton

来源：评论

学校读者我要写书评

暂无评论

Indic Language Empowerment in E-commerce: Integrating Generative Translations and Vectorized Searches 1

Indic Language Empowerment in E-commerce: Integrating Genera...

引用

1st International Conference on AIML-Applications for engineering and technology, ICAET 2025

作者： Bhosale, Parth Gopale, Kunal Taware, Karan Bewoor, Laxmi Vishwakarma Institute Of Information Technology Department of Computer Engineering Pune India

ISBN: (纸本)9798350355611

E-commerce has revolutionized the retail landscape, offering consumers unparalleled convenience and a vast array of choices from the comfort of their homes. Enabling e-commerce in native languages is crucial for creating an inclusive shopping experience, allowing businesses to effectively reach and engage diverse linguistic communities. India is renowned for its multilingual and cultural legacy and this work is an attempt to reach a wider consumer base and increase market potential in India by providing online shopping for cultural festivals. The proposed work is an attempt to develop native language E-commerce platform for various festivals celebrated in India by using Large Language Models (LLM) to generate contextually rich and relevant contents making the user experience more intuitive and personalized. Finally, we talk about making search features better by using modern technology. By using smart algorithms and models, we help users search for products in their own language, and show them results that are most relevant to what they're looking for. Through careful testing and analysis, we show that our solution works well in improving how people interact with e-commerce websites, reaching more customers, and making more sales. Our research is important because it helps bridge the language gap online, giving businesses new opportunities to grow and making online shopping more accessible for everyone. © 2025 IEEE.

关键词： Marketplaces

来源：评论

学校读者我要写书评

暂无评论

EfficientNetB0-based Automated Diabetic Retinopathy Classification in Fundus Images 3

EfficientNetB0-based Automated Diabetic Retinopathy Classifi...

引用

3rd International Conference on Automation, Computing and Renewable Systems, ICACRS 2024

作者： Majaw, Ezekiel Alaric Sundar, G. Naveen Narmadha, D. Thangavel, Senthil Kumar Venkatesan Ajibesin, Adeyemi Abel Department of Computer Science and Engineering Karunya Institute of Technology and Sciences Coimbatore India Department of Computer Science and Engineering Amrita School of Computing Coimbatore India Department of Information Technology Faculty of Informatics and Design Cape Peninsula University of Technology South Africa

ISBN: (纸本)9798331532420

The most common cause of blindness among diabetics is diabetic retinopathy. This disease must be detected in its early stages as delaying treatment can result in permanent blindness. In today’s modern world with the development of technologies, automated techniques have improved the accuracy and efficiency of detecting and classifying this disease. This study suggests a deep learning model utilizing EfficientNetB0 that focuses on designing a neural network structure that requires fewer parameters and computations compared to traditional architectures. Further to help in increasing the accuracy, the image is being processed with methods like CLAHE which is an image processing algorithm that helps in enhancing the contrast while keeping the noise of the images as low as possible. Also, augmentation is being done on the images and SMOTE is applied to handle class imbalance. We collected the APTOS Blindness Detection dataset consisting of retinal fundus images. The accuracy that is being obtained from the proposed methodology is 98.78% indicating the model's effectiveness in detection and classifying between the various stages of diabetic retinopathy. © 2024 IEEE.

关键词： Contrastive 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：