检索结果-内蒙古大学图书馆

5th International Conference on Data Science, machine Learning and applications

作者： Deepan, S. Torres-Cruz, Fred Placido-Lerma, Ruben L. Udhayakumar, R. Anuradha, S. Kapila, Dhiraj SR Univ Sch Comp Sci & Artificial Intelligence Warangal Telangana India Univ Nacl Altiplanode Puno Unit Stat & Comp Engn Puno Peru Univ Andina Nestor Caceres Velasquez Syst Engn Juliaca Peru VelTech Rangarajan Dr Sagunthala R&D Inst Sci & T Chennai Tamil Nadu India Sri Sairam Engn Coll Dept English Chennai Tamil Nadu India Lovely Profess Univ Dept Comp Sci & Engn Phagwara Punjab India

ISBN: (纸本)9789819780334;9789819780310;9789819780303

Deep neural networks have recently seen a significant surge in adoption for different Artificial Intelligence technologies due to the development of powerful computer systems. However, because of the growing security concerns, they are susceptible to dangerous risks. Adversarial instances were initially discovered in the field of computer vision (CV), where systems were deceived by altering their initial inputs. In the field of natural language processing (NLP), additionally they occur. Several approaches are put up to address this gap and handle an extensive variety of NLP applications. We give an organized survey of these works in this *** text is distinct and meaningful in nature, in contrast to the image, which makes the creation of hostile assaults much more challenging. In this study, we present a thorough analysis of adversarial attacks and counterattacks in the textual domain. In order to make the essay self-contained, we examine related important works in computer vision and cover the fundamentals of NLP. We explore unresolved concerns to close the gap between current advancements and increasingly powerful adversarial assaults on NLP DNNs in our survey's conclusion.

关键词： NLP DNN adversarial text attacks adversarial defense textual data

来源：评论

学校读者我要写书评

暂无评论

Wild Animal Detection System on a Farm Using image processing

Wild Animal Detection System on a Farm Using Image Processin...

引用

International Conference on Advances and applications of Artificial Intelligence and machine Learning, ICAAAIML 2023

作者： Barhate, Madhuri Gulhane, Nandini Gunjal, Tanvi Gudi, Swaroop Gulhane, Vedant Wagade, Prasad Gude, Ayush Vishwakarma Institute of Technology Maharashtra Pune411037 India

ISBN: (纸本)9789819795062

Human civilization is based on agriculture, and one of the biggest threats to agricultural productivity is the existence of wild animals on farmlands. Animals are taking over some farmlands, causing significant crop losses, property damage, and livestock risks. As a result, suitable solutions must be developed. A system to support farmers is necessary because the conventional method of constant guard monitoring is impractical. Modern technologies like machine learning, computer vision, and image processing are used by this wild animal detection system to effectively manage conflicts between humans and wildlife. machine learning and computer vision rely on the integration of particular algorithms to guarantee the accuracy and effectiveness of the system. With an emphasis on sustainability, the technology protects property and crops while also aiding in the conservation of wildlife. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Livestock

来源：评论

学校读者我要写书评

暂无评论

TempA-VLP: Temporal-Aware vision-Language Pretraining for Longitudinal Exploration in Chest X-Ray image

TempA-VLP: Temporal-Aware Vision-Language Pretraining for Lo...

引用

2025 IEEE/CVF Winter Conference on applications of Computer vision, WACV 2025

作者： Yang, Zhuoyi Shen, Liyue University of Michigan United States

ISBN: (纸本)9798331510831

Longitudinal medical image processing is a significant task to understand the dynamic changes of disease by taking and comparing image series over time, providing insights into how conditions evolve and enabling more accurate di-agnosis and treatment planning. While recent advance-ments in biomedical vision-Language Pre-training (VLP) have enabled label-efficient representation learning with paired medical images and reports, existing methods pri-marily pair a single image with the corresponding textual report, limiting their ability to capture temporal relation-ships. To address this limitation, it is essential to learn temporal-aware cross-modal representations from sequen-tial medical images and text reports that highlight the tem-poral changes occurring between examinations. Specifi-cally, we introduce TempA- Vlp, a temporal-aware vision language pre-training framework with a cross-exam en-coder to integrate the information from both prior and cur-rent examinations. This approach enables the model to capture dynamic representations that reflect disease pro-gression over time, which allows us to (i) achieve state-of-the-art performance in disease progression classification, (ii) localize dynamic progression regions across consecutive examinations, as demonstrated in our new task, dynamic phrase grounding on the Chest-imagenome Gold dataset, and (iii) highlight progression localized regions, often rele-vant to lesion areas, which in turn improves disease classi-fication tasks on a single image. © 2025 IEEE.

关键词： image coding

来源：评论

学校读者我要写书评

暂无评论

Stochastic Fractal Search: A Decade Comprehensive Review on Its Theory, Variants, and applications

引用

CMES-COMPUTER MODELING IN ENGINEERING & SCIENCES 2025年第3期142卷 2339-2404页

作者： El-Shorbagy, Mohammed A. Bouaouda, Anas Abualigah, Laith Hashim, Fatma A. Prince Sattam Bin Abdulaziz Univ Coll Sci & Humanities Al Kharj Dept Math Al Kharj 11942 Saudi Arabia Hassan II Univ Casablanca Fac Sci & Tech Mohammadia 28806 Morocco Sunway Univ Sch Engn & Technol Petaling Jaya 27500 Selangor Malaysia Chitkara Univ Inst Engn & Technol Ctr Res Impact & Outcome Rajpura 140401 Punjab India Helwan Univ Fac Engn Cairo 11792 Egypt Appl Sci Private Univ Appl Sci Res Ctr Amman 11937 Jordan

With the rapid advancements in technology and science, optimization theory and algorithms have become increasingly important. A wide range of real-world problems is classified as optimization challenges, and meta-heuristic algorithms have shown remarkable effectiveness in solving these challenges across diverse domains, such as machine learning, process control, and engineering design, showcasing their capability to address complex optimization problems. The Stochastic Fractal Search (SFS) algorithm is one of the most popular meta-heuristic optimization methods inspired by the fractal growth patterns of natural materials. Since its introduction by Hamid Salimi in 2015, SFS has garnered significant attention from researchers and has been applied to diverse optimization problems across multiple disciplines. Its popularity can be attributed to several factors, including its simplicity, practical computational efficiency, ease of implementation, rapid convergence, high effectiveness, and ability to address single- and multi-objective optimization problems, often outperforming other established algorithms. This review paper offers a comprehensive and detailed analysis of the SFS algorithm, covering its standard version, modifications, hybridization, and multi-objective implementations. The paper also examines several SFS applications across diverse domains, including power and energy systems, image processing, machine learning, wireless sensor networks, environmental modeling, economics and finance, and numerous engineering challenges. Furthermore, the paper critically evaluates the SFS algorithm's performance, benchmarking its effectiveness against recently published meta-heuristic algorithms. In conclusion, the review highlights key findings and suggests potential directions for future developments and modifications of the SFS algorithm.

关键词： Meta-heuristic algorithms stochastic fractal search evolutionary computation engineering applications swarm intelligence optimization

来源：评论

学校读者我要写书评

暂无评论

Advancing Deep Learning on Edge Devices: Fine-Tuning and Deployment of YOLOv7 Model for Efficient Object Detection in AI based Computer vision applications 3

Advancing Deep Learning on Edge Devices: Fine-Tuning and Dep...

引用

3rd International Conference on Intelligent Data Communication Technologies and Internet of Things, IDCIoT 2025

作者： Shekhar, Sudhanshu Sathwik, T.S. Pritwani, Mayank Mohana Ramakanth Kumar, P. Sreelakshmi, K. RV College of Engineering® Bengaluru India

ISBN: (纸本)9798331527549

This paper investigates the optimization and deployment of YOLOv7 deep learning model on NVIDIA Jetson Nano, an AI-focused edge computing platform for object detection in various computer vision applications. The work leverages TensorRT and quantization techniques for model acceleration for good detection accuracy. Further it examines performance metrics such as speed, accuracy, and resource utilization for image dataset. The model is trained using 80 different classes of objects and demonstrates the use of 6 classes. The average detection accuracy obtained 92.35% and the average processing time is 117.8ms. This work supports AI by demonstrating the feasibility of running deep learning models on edge devices and provides insight into the challenges and opportunities of optimizing AI models for energy-efficient, real-time operations on edge devices for various computer vision applications. © 2025 IEEE.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Raw Camera Data Object Detectors: An Optimisation for Automotive Video processing and Transmission

引用

IEEE ACCESS 2025年 13卷 21695-21706页

作者： Chan, Pak Hung Wei, Chuheng Huggett, Anthony Donzella, Valentina Univ Warwick WMG Coventry CV4 7AL England Onsemi Greenwood House Bracknell RG12 2AA England

Whilst Deep Neural Networks (DNNs) have been developing swiftly, most of the research has been focused on videos based on RGB frames. RGB data has been traditionally optimised for human vision and is a highly re-elaborated and interpolated version of the collected raw data. In fact, the sensor collects the light intensity value per pixel, but an RGB frame contains 3 values, for red, green, and blue colour channels. This conversion to RGB requires computational resource, time, power, and increases by a factor of three the amount of output data. This work investigates DNN based detection using (for training and evaluation) Bayer frames, generated from a benchmarking automotive dataset (i.e. KITTI dataset). A Deep Neural Network (DNN) is deployed in an unmodified form, and also modified to accept only single channel frames, such as Bayer frames. Eleven different re-trained versions of the DNN are produced, and cross-evaluated across different data formats. The results demonstrate that the selected DNN has the same accuracy when evaluating RGB or Bayer data, without significant degradation in the perception (the variation of the Average Precision is <1%). Moreover, the colour filter array position and the colour correction matrix do not seem to contribute significantly to the DNN performance. This work demonstrates that Bayer data can be used for object detection in automotive without significant perception performance loss, allowing for more efficient sensing-perception systems.

关键词： image color analysis Artificial neural networks Accuracy Pedestrians Cameras Automotive engineering machine learning Pipelines Feature extraction Europe Bayer data object detection perception sensors assisted and automated driving intelligent vehicles

来源：评论

学校读者我要写书评

暂无评论

A Micro-Topography Enhancement Method for DEMs: Advancing Geological Hazard Identification

引用

REMOTE SENSING 2025年第5期17卷 920-920页

作者： He, Qiulin Dong, Xiujun Li, Haoliang Deng, Bo Sima, Jingsong Chengdu Univ Technol State Key Lab Geohazard Prevent & Geoenvironm Prot Chengdu 610059 Peoples R China

Geological hazards in densely vegetated mountainous regions are challenging to detect due to terrain concealment and the limitations of traditional visualization methods. This study introduces the LiDAR image highlighting algorithm (LIHA), a novel approach for enhancing micro-topographical features in digital elevation models (DEMs) derived from airborne LiDAR data. By analogizing terrain profiles to non-stationary spectral signals, LIHA applies locally estimated scatterplot smoothing (Loess smoothing), wavelet decomposition, and high-frequency component amplification to emphasize subtle features such as landslide boundaries, cracks, and gullies. The algorithm was validated using the Mengu landslide case study, where edge detection analysis revealed a 20-fold increase in identified micro-topographical features (from 1907 to 37,452) after enhancement. Quantitative evaluation demonstrated LIHA's effectiveness in improving both human interpretation and automated detection accuracy. The results highlight LIHA's potential to advance early geological hazard identification and mitigation, particularly when integrated with machine learning for future applications. This work bridges signal processing and geospatial analysis, offering a reproducible framework for high-precision terrain feature extraction in complex environments.

关键词： digital elevation model LiDAR micro-topographical features vision enhancement wavelet decomposition edge detection geological hazard identification

来源：评论

学校读者我要写书评

暂无评论

vision-Based Autonomous Car Brake and Steering Assistance Using Deep Learning 1

Vision-Based Autonomous Car Brake and Steering Assistance Us...

引用

1st International Conference on AIML-applications for Engineering and Technology, ICAET 2025

作者： Anees, H. Agrawal, Pooja School of Robotics Defence Institute of advanced technology Pune India

ISBN: (纸本)9798350355611

This paper presents a novel approach for enhancing vehicle safety and navigation through an integrated system for lane detection, vehicle alignment, and automatic braking using visual feedback. Our proposed system employs advanced deep learning and computer vision techniques with real-time processing to detect the exact boundaries of lane and ensures precise vehicle movement within the lane. The system continuously analyses lane markings and modifies the vehicle's position to ensure optimal lane adherence by utilizing a combination of machine learning algorithms and camera-based image processing. Additionally, the system incorporates an adaptive braking mechanism that identifies vehicles ahead using visual inputs. Furthermore, the jerks experienced during steering alignment can be greatly reduced by the suggested steering control system. The system's efficiency in various driving conditions is evidenced by its experimental simulation results, which also show improvements in collision avoidance and lane-keeping accuracy. This approach contributes to improved driving convenience and road safety by marking a substantial advancement in autonomous driving technologies. © 2025 IEEE.

关键词： Steering

来源：评论

学校读者我要写书评

暂无评论

Automatic mango leaf disease detection using different transfer learning models

引用

Multimedia Tools and applications 2025年第11期84卷 9185-9218页

作者： Varma, Teena Mate, Prajwal Azeem, Noamaan Abdul Sharma, Sanjeev Singh, Bhupendra Indian Institute of Information Technology Pune India

The cultivation of mangoes contributes significantly to the economy and food security of many tropical and subtropical regions. However, mango trees are susceptible to various leaf diseases that can significantly affect crop yield and quality. Detecting and diagnosing these diseases early is crucial for sustainable mango production. As a subset of machine learning, deep learning has emerged as a powerful tool for image analysis and classification, offering promising solutions for mango leaf disease detection. This work presents a systematic analysis of existing studies, including data collection, pre-processing techniques, network architectures, and performance evaluation metrics. It uses various deep learning pre-trained models like VGG19, InceptionV3, ResNet152V2, DenseNet121, InceptionResNetV2, MobileNetV2, and Xception Models. InceptionV3 model performed best with a 99.87% accuracy rate, compared to other models. A comparison with earlier research is also presented to demonstrate the effectiveness of the research. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2024.

关键词： Computer vision

来源：评论

学校读者我要写书评

暂无评论

Baking Neural Radiance Fields for Real-Time View Synthesis

引用

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND machine INTELLIGENCE 2025年第5期47卷 3310-3321页

作者： Hedman, Peter Srinivasan, Pratul P. Mildenhall, Ben Reiser, Christian Barron, Jonathan T. Debevec, Paul Google Res Mountain View CA 94043 USA Netflix Los Gatos CA 95032 USA

Neural volumetric representations such as Neural Radiance Fields (NeRF) have emerged as a compelling technique for learning to represent 3D scenes from images with the goal of rendering photorealistic images of the scene from unobserved viewpoints. However, NeRF's computational requirements are prohibitive for real-time applications: rendering views from a trained NeRF requires querying a multilayer perceptron (MLP) hundreds of times per ray. We present a method to train a NeRF, then precompute and store (i.e., "bake") it as a novel representation called a Sparse Neural Radiance Grid (SNeRG) that enables real-time rendering on commodity hardware. To achieve this, we introduce 1) a reformulation of NeRF's architecture and 2) a sparse voxel grid representation with learned feature vectors. The resulting scene representation retains NeRF's ability to render fine geometric details and view-dependent appearance, is compact (averaging less than 90 MB per scene), and can be rendered in real-time (higher than 30 frames per second on a laptop GPU). Actual screen captures are shown in our video.

关键词： Rendering (computer graphics) Three-dimensional displays Real-time systems image color analysis Vectors Graphics processing units image reconstruction Computer vision neural rendering real-time rendering view synthesis

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：