检索结果-内蒙古大学图书馆

Efficient Saliency Map Detection for Low-Light Images Based on Image Gradient

IEEE Transactions on Circuits and Systems for Video technology 2024年第2期34卷 852-865页

作者： Lin, Chun-Yi Haq, Muhamad Amirul Chen, Jiun-Han Ruan, Shanq-Jang Naroska, Edwin National Taiwan University of Science and Technology Department of Electronic and Computer Engineering Taipei10607 Taiwan Hochschule Niederrhein University of Applied Science Faculty of Electrical Engineering and Computer Science Krefeld47805 Germany

Recently, deep learning has been widely employed across various domains. The Convolution Neural Network (CNN), a popular deep learning algorithm, has been successfully utilized in object recognition tasks, such as face recognition, vehicle recognition, and license plate recognition. However, conventional methods for object recognition may not be appropriate for low-light image recognition due to information loss in the dark regions and unexpected noise that can impair object quality. Therefore, the development of specialized techniques for low-light image enhancement has become a major research focus for object detection. This paper proposed a gradient-based saliency map detection method with an improved ResNet architecture that outperforms previous works in detecting multiple or large objects. Additionally, the proposed method enhances images with the object as the center and emphasizes foreground-background differences. Compared with previous works, this paper achieves1.28× improvements in the parameters and 1.32× faster inference speed than the original ResNet architecture. © 1991-2012 IEEE.

关键词： Neural networks

来源：评论

学校读者我要写书评

暂无评论

Deep Learning-Based Cloud Security: Innovative Attack Detection and Privacy Focused Key Management

引用

IEEE Transactions on computers 2025年第6期74卷 1978-1989页

作者： Ahmad, Shahnawaz Arif, Mohd Mehfuz, Shabana Ahmad, Javed Nazim, Mohd Bennett University School of Computer Science Engineering and Technology Greater Noida201310 India Galgotias University School of Computer Science & Engineering Greater Noida203201 India Jamia Millia Islamia Department of Electrical Engineering New Delhi110025 India Sharda University Department of Computer Science and Engineering Sharda School of Engineering and Technology Greater Noida201310 India Noida Institute of Engineering and Technology Department of Computer Science and Engineering School of Computer Science and IT Greater Noida201306 India

Cloud Computing (CC) is widely adopted in sectors like education, healthcare, and banking due to its scalability and cost-effectiveness. However, its internet-based nature exposes it to cyber threats, necessitating advanced security frameworks. Traditional models suffer from high false positives and limited adaptability. To address these challenges, VECGLSTM, an attack detection model integrating Variable Long Short-Term Memory (VLSTM), capsule networks, and the Enhanced Gannet Optimization Algorithm (EGOA), is introduced. This hybrid approach enhances accuracy, reduces false positives, and dynamically adapts to evolving threats. EGOA is employed for its superior optimization capability, ensuring faster convergence and resilience. Additionally, Chaotic Cryptographic Pelican Tunicate Swarm Optimization (CCPTSO) is proposed for privacy-preserving key management. This model combines chaotic cryptographic techniques with the Pelican Tunicate Swarm Optimization Algorithm (PTSOA), leveraging the pelican algorithm’s exploration strength and the tunicate swarm’s exploitation ability for optimal encryption security. Performance evaluation demonstrates 99.675% accuracy, 99.5175% recall, 99.7075% precision, and 99.615% F1-score, along with reduced training (1.79s), encryption (0.986s), and decryption (1.029s) times. This research significantly enhances CC security by providing a scalable, adaptive framework that effectively counters evolving cyber threats while ensuring efficient key management. © 1968-2012 IEEE.

关键词： Cloud security

来源：评论

学校读者我要写书评

暂无评论

Deep learning and machine learning based highly accurate reflection prediction model for multi layers anti-reflection coatings

引用

Optical and Quantum Electronics 2025年第1期57卷 1-22页

作者： Oktay, Semih Duru, İremnur Bakır, Halit Tabaru, Timuçin Emre Department of Electrical-Electronics Engineering Sivas University of Science and Technology Sivas Turkey Computer Engineering Department Sivas University of Science and Technology Sivas Turkey

This study focuses on creating an accurate reflection prediction model that will guide the design of filters with multilayer Anti-Reflection Coating (ARC) to optimize the thickness parameters using Machine Learning (ML) and Deep Learning (DL) techniques. This model aims to shed light on the design process of a multilayer optical filter, making it more cost-effective by providing faster and more precise production. In creating this model, a dataset containing data obtained from 3000 (1500 Ge–Al2O3, 1500 Ge–SiO2) simulations previously performed on a computer based on the thicknesses of multilayer structural materials was used. The data are generated using Computational Electromagnetic simulation software based on the Finite-Difference Time-Domain method. To understand the mechanism of the proposed model, two different two-layer coating simulations were studied. While Ge was used as the substrate in both coatings, Al2O3 and SiO2 were used as the second layers. The data set consists of the 3–5 µm and 8–12 µm bands typical for the mid-wave infrared (MWIR) and long-wave infrared (LWIR) bands and includes reflectance values for wavelengths ranging between these spectra. In the specified 2-layer data set, the average reflectance was obtained with a minimum of 0.36 at 515 nm Ge and 910 nm SiO2 thicknesses. This value can be increased by adapting the proposed model to more than 2 layers. Six ML algorithms and a DL model, including artificial neural networks and convolutional neural networks, are evaluated to determine the most effective approach for predicting reflectance properties. Furthermore, in the proposed model, a hyperparameter tuning phase is used in the study to compare the efficiency of ML and DL methods to generate dual-band ARC and maximize the prediction accuracy of the DL algorithm. To our knowledge, this is the first time this has been implemented in this field. The results show that ML models, particularly decision tree (MSE: 0.00000069, RMSE: 0.00083), rand

关键词： Antireflection coatings

来源：评论

学校读者我要写书评

暂无评论

Multi-Modal Self-Supervised Learning AlgorithmBased Product Recommendation

引用

IEEE Transactions on Automation science and engineering 2025年 22卷 15371-15380页

作者： Gao, Li Fan, Haiping Chen, Qingkui Yang, Heyu School of Optical-Electrical and Computer Engineering University of Shanghai for Science and Technology Shanghai China

With the rise of e-commerce, personalized recommendation algorithms have received much attention in recent years. Meanwhile, multimodal recommendation algorithms have become the next competitive track of personalized recommendation. The current mainstream recommendation algorithms (e.g., NeuMF, NGCF) rely too much on empirical data and do not effectively utilize multimodal information to model user preferences, while there is the problem of sparsity of user preference data. Secondly, they are still limited to using the same weights to fuse the item features of different modalities, and have the problem of inaccurate recommendation. For this reason, this paper proposes Multi-Modal Self-Supervised Learning Algorithm Based Product Recommendation (MMSLRec). (1) Data augmentation by self-supervised adversarial perturbation generation learning is performed to supplement the missing interaction information of tailed products to solve the common data sparsity problem in product recommendation. (2) A multimodal representation learning module is introduced to capture deep features in image and text modalities using AlexNet and Bert, respectively, in order to avoid the recommendation accuracy degradation problem caused by insufficient data feature learning. (3) Learning the user's perceived strength in different modalities, using the user co-occurrence matrix to fuse the features and make recommendations. Experiments on Amazon-Electronics, Amazon-Clothing, and Amazon-Baby, with different degrees of data sparsity verify the effectiveness of our MMSLRec in solving the data sparsity problem and the multimodal user preference modeling problem, and compared to the state-of-the-art baseline recommendation model on the three datasets, the Recall@20 improved by 9.26%, 3.11% and 2.92%, respectively. © 2004-2012 IEEE.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Improving Machine Translation Formality with Large Language Models

引用

computers, Materials & Continua 2025年第2期82卷 2061-2075页

作者： Murun Yang Fuxue Li School of Computer Science and Engineering Northeastern UniversityShenyang110819China College of Electrical Engineering Yingkou Institute of TechnologyYingkou115014China

Preserving formal style in neural machine translation (NMT) is essential, yet often overlooked as an optimization objective of the training processes. This oversight can lead to translations that, though accurate, lack formality. In this paper, we propose how to improve NMT formality with large language models (LLMs), which combines the style transfer and evaluation capabilities of an LLM and the high-quality translation generation ability of NMT models to improve NMT formality. The proposed method (namely INMTF) encompasses two approaches. The first involves a revision approach using an LLM to revise the NMT-generated translation, ensuring a formal translation style. The second approach employs an LLM as a reward model for scoring translation formality, and then uses reinforcement learning algorithms to fine-tune the NMT model to maximize the reward score, thereby enhancing the formality of the generated translations. Considering the substantial parameter size of LLMs, we also explore methods to reduce the computational cost of INMTF. Experimental results demonstrate that INMTF significantly outperforms baselines in terms of translation formality and translation quality, with an improvement of +9.19 style accuracy points in the German-to-English task and +2.16 COMET score in the Russian-to-English task. Furthermore, our work demonstrates the potential of integrating LLMs within NMT frameworks to bridge the gap between NMT outputs and the formality required in various real-world translation scenarios.

关键词： Neural machine translation formality large language model text style transfer style evaluation reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Generalized halftone classification approach using stochastic analysis

引用

Journal of Ambient Intelligence and Humanized Computing 2023年第11期14卷 14907-14919页

作者： Guo, Jing Ming Seshathiri, Sankarasrinivasan College of Electrical Engineering and Computer Science National Taiwan University of Science and Technology Taipei Taiwan

Halftone classification is a primary requisite for the perfect reconstruction of binary patterns during inverse halftone process. Majority of the halftone classification techniques are either limited to error diffused halftone or to limited categories, and cannot be generalized to all halftone versions. In this scenario, a new classification approach is proposed based on the premise that the stochastic analysis can uniquely characterize the halftone patterns. The proposed scheme exploits inherent association of halftone patterns and stochastic geometry, and utilizes its spatial and spectral parameters for feature vector construction. Extreme learning machine based multi-classifier model is adopted resulting in rapid and accurate classification. A digital halftone database comprises of 96 reference images along with the 21 varieties of halftone and multitone class is developed. From extensive analysis, it has been validated that the proposed scheme can achieve a 100% classification rate on many halftone versions and exhibit superior performance over the exiting classification techniques. © 2018, Springer-Verlag GmbH Germany, part of Springer Nature.

关键词： Classification (of information)

来源：评论

学校读者我要写书评

暂无评论

Highly-efficient(>70%)and Wide-spectral(400-1700 nm)sub-micron-thick InGaAs photodiodes for future high-resolution image sensors

引用

Light(science & Applications) 2024年第12期13卷 3224-3235页

作者： Dae-Myeong Geum Jinha Lim Junho Jang Seungyeop Ahn SeongKwang Kim Joonsup Shim Bong Ho Kim Juhyuk Park Woo Jin Baek Jaeyong Jeong SangHyeon Kim School of Electrical Engineering Korea Advanced Institute of Science and Technology(KAIST)Daejeon 34141Republic of Korea Department of Electrical and Computer Engineering Inha UniversityIncheon 22212Republic of Korea

This paper demonstrates the novel approach of sub-micron-thick InGaAs broadband photodetectors(PDs)designed for high-resolution imaging from the visible to short-wavelength infrared(SWIR)*** approaches encounter challenges such as low resolution and crosstalk issues caused by a thick absorption layer(AL).Therefore,we propose a guided-mode resonance(GMR)structure to enhance the quantum efficiency(QE)of the InGaAs PDs in the SWIR region with only sub-micron-thick *** TiOx/Au-based GMR structure compensates for the reduced AL thickness,achieving a remarkably high QE(>70%)from 400 to 1700 nm with only a 0.98μm AL InGaAs PD(defined as 1μm AL PD).This represents a reduction in thickness by at least 2.5 times compared to previous results while maintaining a high ***,the rapid transit time is highly expected to result in decreased electrical *** effectiveness of the GMR structure is evident in its ability to sustain QE even with a reduced AL thickness,simultaneously enhancing the transit *** breakthrough offers a viable solution for high-resolution and low-noise broadband image sensors.

关键词： thick resolution image

来源：评论

学校读者我要写书评

暂无评论

Rolling velocity and relative motion of particle detector in local granular flow

引用

Chinese Physics B 2022年第11期31卷 388-394页

作者： Ran Li Bao-Lin Liu Gang Zheng Hui Yang School of Health Science and Engineering University of Shanghai for Science and TechnologyShanghai 200093China School of Optical-Electrical and Computer Engineering University of Shanghai for Science and TechnologyShanghai 200093China

The velocity of a particle detector in granular flow can be regarded as the combination of rolling and sliding *** study of the contribution of rolling velocity and sliding velocity provides a new explanation to the relative motion between the detector and the local granular *** this study,a spherical detector using embedded inertial navigation technology is placed in the chute granular flow to study the movement of the detector relative to the granular *** is shown by particle image velocimetry(PIV)that the velocity of chute granular flow conforms to Silbert’s *** the velocity of the detector is greater than that of the granular flow around *** decomposing the velocity into sliding and rolling velocity,it is indicated that the movement of the detector relative to the granular flow is mainly caused by *** rolling detail shown by DEM simulation leads to two potential mechanisms based on the position and drive of the detector.

关键词： local velocity distribution rolling velocity inertial navigation technology relative velocity dependent(RVD)rolling friction

来源：评论

学校读者我要写书评

暂无评论

ESG Investing: A Statistically Valid Approach to Data-Driven Decision Making and the Impact of ESG Factors on Stock Returns and Risk

引用

IEEE Access 2024年 12卷 69434-69444页

作者： Teja, Kamurthi Ravi Liu, Chuan-Ming National Taipei University of Technology International Program of Electrical Engineering and Computer Science Taipei10608 Taiwan National Taipei University of Technology Department of Computer Science and Information Engineering Taipei10608 Taiwan

This study examines the impact of environmental, social, and governance (ESG) factors on economic investment from a statistical perspective, aiming to develop a tested investment strategy that capitalizes on the connection between ESG factors and financial performance. ESG investing: A statistically valid approach to data-driven decision-making (ESGI-SVADDM) investment strategy is based on a rigorous, statistically valid approach that utilizes data, math, statistics, and data science libraries to drive investment decisions, eliminating the need for personal opinions and subjectivity. The process includes establishing an investment thesis, formulating testable hypotheses (HPS), retrieving and refining relevant data, calculating relevant measures, and testing and validating all hypotheses. The study uses S&P 500 stock data and ESG data from Sustainalytics to test the hypotheses. The empirical tests conducted revealed a negative correlation between ESG risk and expected returns, as well as a positive trend in the relationship between ESG risk and the overall risk of stocks. Moreover, the study found that higher ESG risk scores are associated with lower returns for investors, and that adopting a strategy of investing in stocks with low ESG risk and shorting stocks with high ESG risk yields superior returns compared to the market portfolio. © 2013 IEEE.

关键词： Investments

来源：评论

学校读者我要写书评

暂无评论

Data-driven output regulation control for constrained linear systems

引用

science China(Information sciences) 2025年第3期68卷 338-353页

作者： Chaoyu XIA Yi DONG Chaoli WANG Shengyuan XU Shanghai Research Institute for Intelligent Autonomous Systems Tongji University College of Electronic and Information Engineering Shanghai Research Institute for Intelligent Autonomous SystemsTongji University Department of Control Science and Engineering School of Optical-Electrical and Computer EngineeringUniversity of Shanghai for Science and Technology Department of Automation Nanjing University of Science and Technology

This study introduces a data-driven approach for state and output feedback control addressing the constrained output regulation problem in unknown linear discrete-time systems. Our method ensures effective tracking performance while satisfying the state and input constraints, even when system matrices are not available. We first establish a sufficient condition necessary for the existence of a solution pair to the regulator equation and propose a data-based approach to obtain the feedforward and feedback control gains for state feedback control using linear programming. Furthermore, we design a refined Luenberger observer to accurately estimate the system state, while keeping the estimation error within a predefined set. By combining output regulation theory, we develop an output feedback control strategy. The stability of the closed-loop system is rigorously proved to be asymptotically stable by further leveraging the concept of λ-contractive sets.

关键词： output regulation constrained system data-driven.

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：