检索结果-内蒙古大学图书馆

Evaluating the Impact of Underwater Image Enhancement on Object Detection Performance: A Comprehensive Study

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Awad, Ali Saleem, Ashraf Paheding, Sidike Lucas, Evan Al-Ratrout, Serein Havens, Timothy C. Department of Applied Computing College of Computing Michigan Technological University HoughtonMI49931 United States Department of Computer Science and Engineering Fairfield University FairfieldCT06824 United States Department of Computer Science College of Computing Michigan Technological University HoughtonMI49931 United States Institute of Computing and Cybersystems Michigan Technological University HoughtonMI49931 United States

Underwater imagery often suffers from severe degradation that results in low visual quality and object detection performance. This work aims to evaluate state-of-the-art image enhancement models, investigate their impact on underwater object detection, and explore their potential to improve detection performance. To this end, we selected representative underwater image enhancement models covering major enhancement categories and applied them separately to two recent datasets: 1) the Real-World Underwater Object Detection Dataset (RUOD), and 2) the Challenging Underwater Plant Detection Dataset (CUPDD). Following this, we conducted qualitative and quantitative analyses on the enhanced images and developed a quality index (Q-index) to compare the quality distribution of the original and enhanced images. Subsequently, we compared the performance of several YOLO-NAS detection models that are separately trained and tested on the original and enhanced image sets. Then, we performed a correlation study to examine the relationship between enhancement metrics and detection performance. We also analyzed the inference results from the trained detectors presenting cases where enhancement increased the detection performance as well as cases where enhancement revealed missed objects by human annotators. This study suggests that although enhancement generally deteriorates the detection performance, it can still be harnessed in some cases for increased detection performance and more accurate human annotation. © 2024, CC BY.

关键词： Underwater photography

THE SECOND-ORDER ZERO DIFFERENTIAL SPECTRA OF SOME APN AND OTHER MAPS OVER FINITE FIELDS

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Garg, Kirpa Ul Hasan, Sartaj Riera, Constanza Stanica, Pantelimon Department of Mathematics Indian Institute of Technology Jammu Jammu181221 India Department of Computer Science Electrical Engineering and Mathematical Sciences Western Norway University of Applied Sciences Bergen5020 Norway Applied Mathematics Department Naval Postgraduate School MontereyCA93943 United States

MSC Codes 12E20, 11T06, 94A60The Feistel Boomerang Connectivity Table and the related notion of F-Boomerang uniformity (also known as the second-order zero differential uniformity) has been recently introduced by Boukerrou et al. [3]. These tools shall provide a major impetus in the analysis of the security of the Feistel network-based ciphers. In the same paper, a characterization of almost perfect nonlinear functions (APN) over fields of even characteristic in terms of second-order zero differential uniformity was also given. Here, we find a sufficient condition for an odd or even function over fields of odd characteristic to be an APN function, in terms of second-order zero differential uniformity. Moreover, we compute the second-order zero differential spectra of several APN or other low differential uniform functions, and show that our considered functions also have low second-order zero differential uniformity, though it may vary widely, unlike the case for even characteristic when it is always zero. Copyright © 2023, The Authors. All rights reserved.

关键词：

Structural modification of Naproxen;physicochemical, spectral, medicinal, and pharmacological evaluation

学校读者我要写书评

暂无评论

Informatics in Medicine Unlocked 2025年 53卷

作者： Farque, Md Omor Islam, Rahat Moinul Rahman Joni, Md Ferdous Akter, Mimona Akter, Shilpy Islam, Mohammad Didarul Salim, MD Jubaer Bin Aziz, Ahamed Abdul Kabir, Emranul Uzzaman, Monir Faculty of Science Department of Chemistry University of Chittagong Chittagong 4331 Bangladesh Faculty of Science and Engineering Department of Pharmacy International Islamic University Chittagong Chittagong 4318 Bangladesh Department of Pharmacy Comilla University Comilla 3506 Bangladesh Faculty of Science Department of Applied Chemistry & Chemical Engineering University of Chittagong Chittagong 4331 Bangladesh Department of Drug Design Computer in Chemistry and Medicine Laboratory Dhaka Bangladesh Faculty of Science and Engineering Department of Electrical and Electronic Engineering International Islamic University Chittagong 4318 Bangladesh

Naproxen (Nap), a widely used nonsteroidal anti-inflammatory drug (NSAID), effectively reduces inflammation, pain, and fever by inhibiting cyclooxygenase enzymes (i.e., COX-1 and COX-2). However, its therapeutic use is often limited by significant adverse effects, including gastrointestinal hemorrhage, nephrotoxicity, hepatotoxicity, hematuria, and aphthous ulcers. In this study, we aimed to enhance both the efficacy and safety profile of Nap by making targeted structural modifications to the parent drug. Specifically, selected functional groups (i.e., CH_3, OCH₃, CF₃, OCF₃, NH₂, CH₂NH₂, NHCONH₂ and NHCOCH₃) were introduced into the naphthalene nucleus. The geometry of the modified compounds was optimized via DFT with the B3LYP functional and 6-31+G (d, p) basis set, facilitating physicochemical and spectral analysis. Molecular docking studies were conducted against the human Prostaglandin G/H synthase 2 (5F19) and Mus musculus Prostaglandin-endoperoxide synthase 2 (3NT1), and these candidates were subjected to MD simulation. ADMET and PASS analyses were performed to evaluate the medicinal efficacy and toxicological profiles of the derivatives. Our findings identified several promising candidates with enhanced therapeutic benefits and reduced toxicity compared with the parent Nap. Docking analysis revealed that analogs exhibited stronger binding affinities compared to Nap and selectivity towards COX-2. These candidates demonstrated stability over a 100 ns MD simulation, exhibiting significant hydrogen bonding. Compared with the parent drug, most of these analogs displayed reduced hepatotoxicity, nephrotoxicity, carcinogenicity, and gastrointestinal hemorrhage activity, as supported by pharmacokinetic calculations. This study demonstrated that improved chemical and medicinal properties lead to a reduction in side effects. © 2025 The Authors

关键词： Cyclooxygenase Medicinal and toxicological assessment Naproxen Physicochemical and spectral analysis

ENHANCING PRINTED CIRCUIT BOARD DEFECT DETECTION THROUGH ENSEMBLE LEARNING

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Law, Ka Nam Canaan Yu, Mingshuo Zhang, Lianglei Zhang, Yiyi Xu, Peng Liu, Jun Gao, Jerry Department of Applied Data Science San José State University San José United States Dept. of ISE Virginia Tech Blacksburg United States Dept. of Computer Engineering San José State University San José United States

The quality control of printed circuit boards (PCBs) is paramount in advancing electronic device technology. While numerous machine learning methodologies have been utilized to augment defect detection efficiency and accuracy, previous studies have predominantly focused on optimizing individual models for specific defect types, often overlooking the potential synergies between different approaches. This paper introduces a comprehensive inspection framework leveraging an ensemble learning strategy to address this gap. Initially, we utilize four distinct PCB defect detection models utilizing state-of-the-art methods: EfficientDet, MobileNet SSDv2, Faster RCNN, and YOLOv5. Each method is capable of identifying PCB defects independently. Subsequently, we integrate these models into an ensemble learning framework to enhance detection performance. A comparative analysis reveals that our ensemble learning framework significantly outperforms individual methods, achieving a 95% accuracy in detecting diverse PCB defects. These findings underscore the efficacy of our proposed ensemble learning framework in enhancing PCB quality control processes. © 2024, CC BY.

关键词： Adversarial machine learning

Adaptive computer Vision Solutions for Individuals with Special Needs

学校读者我要写书评

暂无评论

Adaptive Computer Vision Solutions for Individuals with Spec...

Computational Intelligence Approaches and Applications (ICCIAA), International Conference on

作者： Issam Jebreen Shadi Nashwan Ahmad Al-Qerem Ali Mohd Ali Mohammad A Al Khaldy Mohammed Rajab Faculty of Information Technology Middle East University Jordan Computer Science Department Zarka University Jordan College of Engineering Technology Al-Balqa Applied University Jordan Business Intelligence Data Analytics University of Petra Amman Jordan University Headquarter University Of Anbar Iraq

ISBN: (数字)9798331523657

ISBN: (纸本)9798331523664

With the rapid evolution of artificial intelligence (AI) and computer vision, assistive technologies have become increasingly adept at supporting individuals with special needs. This paper provides a comprehensive examination of two critical areas: AI-powered visual assistance systems for the visually impaired and sign language recognition and generation frameworks for the Deaf and hard-of-hearing communities. We review the latest research trends, algorithms, and tools, highlighting emerging deep learning architectures such as convolutional neural networks (CNNs), recurrent neural networks (RNNs), and Transformers in both domains. We then propose a unified framework that leverages edge computing, multimodal learning, and advanced human-computer interaction (HCI) techniques to address challenges such as real-time object recognition, complex gesture capture, data scarcity, and cultural variations in sign languages. We further detail an experimental design and set of results that demonstrate the feasibility and benefits of these approaches, followed by a discussion of ethical considerations, limitations, and future directions. Our goal is to illustrate a path toward more inclusive AI systems that offer robust solutions for accessibility, ultimately bridging communication gaps and fostering independence for individuals with special needs [1]–[11].

关键词： Sign language computer vision Visualization Ethics Visual impairment Assistive technologies Transformers Real-time systems Object recognition Artificial intelligence

First steps towards modeling the interaction of cardiovascular agents and smooth muscle activation in arterial walls

学校读者我要写书评

暂无评论

First steps towards modeling the interaction of cardiovascul...

92nd Annual Meeting of the International Association of applied Mathematics and Mechanics, GAMM 2022

作者： Nurani Ramesh, S. Uhlmann, K. Saßmannshausen, L. Rheinbach, O. Klawonn, A. Heinlein, A. Balzani, D. Civil and Environmental Engineering Ruhr University Bochum Universitätsstraße 150 Bochum44801 Germany University of Cologne Department of Mathematics and Computer Science Cologne Germany Technische Universität Bergakademie Freiberg Department of Mathematics and Informatics Freiberg Germany Delft Institute of Applied Mathematics Delft University of Technology Mekelweg 4 Delft2628 CD Netherlands

Numerical simulation of the response of healthy and pathological arteries to cardiovascular agents can provide valuable information to the physician in the treatment of diseases such as hypertension, atherosclerosis, and the Marfan syndrome. Here, we provide a first step towards a computational framework to model the effects of antihypertensive agents on the mechanical response of arterial walls. A material model is developed by extending an existing formulation for wall tissue to incorporate the effects of calcium-ion channel blockers. The resulting coupled deformation-diffusion problem is then solved using the finite element method. Simulation results with drug activity show that, indeed, an increased lumen diameter due to reduced contraction is obtained. Additionally, a decrease in the rate of arterial contraction is observed, which is also consistent with expected behavior. Finally, we compare results for an implicit or explicit treatment of the the deformation-diffusion coupling, and we observe that both coupling schemes yield comparable results for a wide range of time step sizes. © 2023 The Authors. Proceedings in applied Mathematics & Mechanics published by Wiley-VCH GmbH.

关键词： Diseases

Time-Series Anomaly Detection of Mozi Malware in IoT Devices Using Arima and Local Outlier Factor

学校读者我要写书评

暂无评论

Time-Series Anomaly Detection of Mozi Malware in IoT Devices...

International Symposium on applied Computational Intelligence and Informatics ( SACI)

作者： Tünde Kaufman Jozsef Katona National Media and Infocommunications Authority Budapest Hungary Kandó Kálmán Faculty of Electrical Engineering Institute of Electronics and Communication Systems Obuda University Budapest Hungary Department of Software Development and Application Institute of Computer Engineering University of Dunaujvaros Dunaujvaros Hungary GAMF Faculty of Engineering and Computer Science John Von Neumann University Kecskemet Hungary Department of Applied Quantitative Methods Faculty of Finance and Accountancy Budapest University of Economics and Business Budapest Hungary

ISBN: (数字)9798331515478

ISBN: (纸本)9798331515485

The increasing demand for IoT devices in industries has generated a high level of cybersecurity threats, with botnets including Mozi exploiting poor passwords and unpatched vulnerabilities to target networked infrastructure. Peer-to-peer (P2P) topology of Mozi, employing Distributed Hash Tables (DHTs) for Command-and-Control (C2) communications, makes it resistant to takedown operations and capable of high-scale Distributed Denial-of-Service (DDoS) attack, data exfiltration, and remote command execution. Traditional one-class, machine learning techniques such as OneClass Support Vector Machine (OCSVM) and Local Outlier Factor (LOF) have been widely used for discovering malwarerelated anomalies. However, such techniques require high labelled training datasets, whose availability in dynamically changing environments of malwares proves challenging. In this work, an AutoRegressive Integrated Moving Average (ARIMA)-based mechanism for discovering time-series anomalies in IoT-infected devices with Mozi infection is proposed. In an experimental setup using a Raspberry Pi running Pi-hole, baseline profiles of CPU consumption, RAM usage, and network activity are first established under normal conditions. Next, a system is subjected to a simulation of infection with Mozi malware and monitored for deviation with ARIMA forecasting and Local Outlier Factor (LOF) for discovering anomalies. By comparing predicted values with actual values, effective intrusion detection for malware-related anomalies is attained, providing a real-time, adaptive mechanism for early intrusion detection. In conclusion, with its effectiveness in discovering stealthy behaviour of malwares, employing ARIMA for discovering time-series anomalies in IoTinfected devices proves effective, and its employment in IoT cybersecurity aids in providing real-time mechanism for discovering threats for system admins and security professionals. As future work, a combination of ARIMA with deep neural networks such as Long

关键词： Training Botnet Intrusion detection Malware Real-time systems Peer-to-peer computing Topology Internet of Things Long short term memory Anomaly detection

WeatherQA: Can Multimodal Language Models Reason about Severe Weather?

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Ma, Chengqian Hua, Zhanxiang Anderson-Frey, Alexandra Iyer, Vikram Liu, Xin Qin, Lianhui Department of Applied Mathematics University of Washington United States Department of Atmospheric Sciences University of Washington United States Department of Computer Science and Engineering University of Washington United States University of California San Diego United States Allen Institute for Artificial Intelligence

Severe convective weather events, such as hail, tornadoes, and thunderstorms, often occur quickly yet cause significant damage, costing billions of dollars every year. This highlights the importance of forecasting severe weather threats hours in advance to better prepare meteorologists and residents in at-risk areas. Can modern large foundation models perform such forecasting? Existing weather benchmarks typically focus only on predicting time-series changes in certain weather parameters (e.g., temperature, moisture) with text-only features. In this work, we introduce WeatherQA, the first multimodal dataset designed for machines to reason about complex combinations of weather parameters (a.k.a., ingredients) and predict severe weather in real-world scenarios. The dataset includes over 8,000 (multiimages, text) pairs for diverse severe weather events. Each pair contains rich information crucial for forecasting-the images describe the ingredients capturing environmental instability, surface observations, and radar reflectivity, and the text contains in-depth forecast analyses written by human experts. With WeatherQA, we systematically evaluate state-of-the-art vision language models (VLMs), including GPT4, Claude3, Gemini-1.5, and a fine-tuned Llama3-based VLM, by designing two challenging tasks: (1) multi-choice QA for predicting affected area and (2) classification of the development potential of severe convection. These tasks require deep understanding of domain knowledge (e.g., atmospheric dynamics) and complex reasoning over multimodal data (e.g., interactions between weather parameters). We show a substantial gap between the strongest VLM, GPT4o, and human reasoning. Our comprehensive case study with meteorologists further reveals the weaknesses of the models, suggesting that better training and data integration are necessary to bridge this gap. WeatherQA is accessible through https://***/chengqianma/WeatherQA. © 2024, CC BY.

关键词： Data integration

Improving Solar Power Generation with InceptionV3 Dust Detection on the Solar Panel Energy Systems

学校读者我要写书评

暂无评论

Improving Solar Power Generation with InceptionV3 Dust Detec...

Ubiquitous Computing and Intelligent Information Systems (ICUIS), International Conference on

作者： Milind Shah Mittal Joshi Pratik Patel Niyati Mevada Rashmika Baria Mayur Chauhan Department of Computer Engineering Faculty of Technology & Engineering (FTE) Devang Patel Institute of Advance Technology & Research (DEPSTAR) Charotar University of Science & Technology (CHARUSAT) Changa India Department of Computer Science & Engineering Krishna School of Emerging Technology & Applied Research - Drs. Kiran & Pallavi Patel Global University (KPGU) Vadodara Gujarat India Gujarat Power Engineering & Research Institute (GPERI) A Constitutient College of Gujarat Technological University Mehsana Gujarat India

As the demand for renewable energy increases, solar (PV) innovation has become a matter of concern. Diverse research proposals have been developed to derive the most significant benefit from the sun's rays, but dust gathering on solar panels and air pollution are two significant challenges. The installation of 40GW of residential solar panels and solar capacity connected to the grid is a significant accomplishment for the Indian government. Technology development to manage solar panel-based energy generation has been crucial for increasing reliability and decreasing costs. As a renewable energy source, solar panels produce zero pollution while generating electricity. When dust accumulates on the surface of the solar panel, the intensity of the light that falls on it is decreased. In turn, this reduces the amount of electricity the solar panel generates. Since solar panel cleaning is essential, continual surveillance and assessment are required to optimize these processes. This highlights the significance of using machine learning or deep learning to detect dirt and clean solar panels to enhance performance. In this research, we implemented dust detection of solar panels using the InceptionV3 model and obtained a remarkable 92.34 percent accuracy. The dataset was obtained from Kaggle and consisted of 1,200 images evenly divided between clean and dirty panels. Each image is 100 pixels by 100 pixels in size. The images have been assigned to the Clean and Dusty categories.

关键词：