检索结果-内蒙古大学图书馆

JOURNAL OF APPLIED SCIENCE AND ENGINEERING 2024年第11期27卷 3419-3430页

作者： Liang, Xunyang Yang, Qi Wang, Yilu Wang, Shida Huang, Xingzhuo Shenyang Ligong Univ Sch Mech Engn Shenyang 110159 Peoples R China

In this paper, an adaptive fractional-order optical flow selection algorithm with improved ant colony clustering is proposed to address the issues of texture processing and dynamic noise perturbation in simultaneous localization and mapping algorithms in dynamic scenes with strong static assumption theory. The algorithm combines the characteristics of fractional differentiation and sparse optical flow algorithm, and makes full use of the weak texture gradient of the image. The ant colony algorithm is improved by using the elite sharing mechanism, and the improved ant colony algorithm is combined with the clustering algorithm. The experimental results show that the algorithm not only realizes the adaptive selection of the best order, but also achieves better dynamic disturbance differentiation ability through the clustering of feature selection. While distinguishing dynamic and static information effectively, more details of optical flow with weak gradient feature are preserved. The proposed algorithm holds promise for simultaneous localization and mapping systems.

关键词： dynamic feature sparse optical -flow fractional differential adaptive algorithm elite sharing mechanism cohesive clustering

来源：评论

学校读者我要写书评

暂无评论

optical technologies and the visual picture of the world: iconics and neuroiconics

引用

JOURNAL OF optical TECHNOLOGY 2022年第8期89卷 434-436页

作者： Shelepin, Yu E. Lutsiv, V. R. Korotaev, V. V. Russian Acad Sci Pavlov Inst Physiol St Petersburg Russia St Petersburg State Univ Aerosp Instrumentat St Petersburg Russia ITMO Univ St Petersburg Russia

A definition of neuroiconics is proposed as a branch of science at the intersection of human and animal physiology and iconics that studies neurophysiological processes and algorithms for processing video information ... 详细信息

关键词： algorithms Image processing Neural networks Pattern recognition Thermal imaging visual system

来源：评论

学校读者我要写书评

暂无评论

Hyperspectral Compute-In-Memory Architecture and Key optical Components 6

Hyperspectral Compute-In-Memory Architecture and Key Optical...

引用

2025 Conference on AI and optical Data Sciences

作者： Park, Byoung Jun Latifpour, Mostafa Honari Yamamoto, Yoshihisa Suh, Myoung-Gyun NTT Res Inc Phys & Informat Labs Sunnyvale CA 94085 USA

ISBN: (纸本)9781510684980;9781510684997

Advancements in artificial intelligence ( AI) have driven a s hift toward disaggregated computing infrastructures, integrating optical technologies in data centers to enhance data transfer. As data centers evolve into hybrid opto-electronic systems, we explore the potential of optics for computational tasks such as matrix-vector multiplication (MVM). Various proposals and demonstrations of optical MVM have been reported, with some 3D optical systems demonstrating notable scalability advantages. However, the predominant systems primarily utilize space multiplexing, leaving the frequency dimension underexploited. In this talk, we introduce a hyperspectral compute-in-memory (CIM) architecture that simultaneously utilizes frequency and space dimensions for single-shot matrix-matrix multiplication (MMM). We will overview the system architecture, discuss the design of key components, and address the use of chip-integrated optical elements such as micro-combs and photonic interposers. The architecture offers remarkable parallelism, s calability, programmability, and efficient chip area use, enabling high computational throughput with a compute density greater than Peta operation per second (PetaOPS/mm(2)). Our work shows potential for energy-efficient, three-dimensional opto-electronic computing in future data center applications.

关键词： optical computing compute-in-memory hyperspectral information processing optical parallelism

来源：评论

学校读者我要写书评

暂无评论

Datasets and Benchmarks for Nanophotonic Structure and Parametric Design Simulations 37

Datasets and Benchmarks for Nanophotonic Structure and Param...

引用

37th Conference on Neural information processing systems (NeurIPS)

作者： Kim, Jungtaek Li, Mingxuan Hinder, Oliver Leu, Paul W. Univ Pittsburgh Pittsburgh PA 15261 USA

ISBN: (纸本)9781713899921

Nanophotonic structures have versatile applications including solar cells, anti-reflective coatings, electromagnetic interference shielding, optical filters, and light emitting diodes. To design and understand these nanophotonic structures, electrodynamic simulations are essential. These simulations enable us to model electromagnetic fields over time and calculate optical properties. In this work, we introduce frameworks and benchmarks to evaluate nanophotonic structures in the context of parametric structure design problems. The benchmarks are instrumental in assessing the performance of optimization algorithms and identifying an optimal structure based on target optical properties. Moreover, we explore the impact of varying grid sizes in electrodynamic simulations, shedding light on how evaluation fidelity can be strategically leveraged in enhancing structure designs.

关键词： optical properties

来源：评论

学校读者我要写书评

暂无评论

Spatial and Temporal Evolution of Wetlands in the Yellow River Delta Based on optical Flow Algorithm

引用

JOURNAL OF information processing systems 2025年第1期21卷 102-113页

作者： Liu, Chao Mu, Ruolan Wang, Chuanlong Yuan, Xiuhe Miao, Sheng Qingdao Univ Technol Sch Environm & Municipal Engn Qingdao Peoples R China Qingdao Univ Technol Sch Informat & Control Engn Qingdao Peoples R China

The Yellow River Delta (YRD) wetlands are the largest coastal wetlands in China, and it serves to control soil erosion, nourish the climate and protect biodiversity. At present, due to climate change and human activities, the wetlands of the YRD are facing ecological and environmental problems such as species invasion, vegetation degradation, and biodiversity reduction. In order to study the evolution of wetland landscape types more intuitively, this paper proposed a semantic segmentation network based on the encoder and decoder structure of ResNet-18. Then, the wetland landscapes in the YRD were classified into five types by combining the Landsat series of remote sensing images. In addition, this paper used the optical flow algorithm to visualize the identification results, which can represent the evolution pattern of wetland landscape types in different years. The results of this paper have an important significance for the subsequent development planning and protection in the YRD.

关键词： Identification of Wetland Landscape Types optical Flow Algorithm ResNet Spatial and Temporal Evolution

来源：评论

学校读者我要写书评

暂无评论

Analysis on Different Optimization Methods, Applications, and Categories of optical Fiber Networks: A Review

Analysis on Different Optimization Methods, Applications, an...

引用

2023 International Conference on Intelligent Data Communication Technologies and Internet of Things, IDCIoT 2023

作者： Dubey, Arjun Singh, Harmeet Kaur, Simarpreet Chandigarh University Department of Ece Punjab Mohali India

ISBN: (纸本)9781665474511

OFs (optical Fibers) are broadly used for information broadcast systems due to their extensive information-carrying capability and dielectric environment. The OFs network architectures that use several WLs (wavelengths) per OFs connect thousands of users in central, urban, or large-scale applications with a wide range of transmission speeds and capabilities. This article defines a detailed introduction to optical fiber networks and general phases of communication of OF (optical Fiber) procedure. The taxonomy may be based on optical fibers such as;(i) Structure, (ii) number of modes, (iii) refractive index profile, (iv) dispersion, (v) signal processing capacity (vi) Polarization. After that, it explains the merits and demerits of optical fiber networks. They studied optical wireless communications applications such as SMs (shopping malls), UW (underwater), industries, etc. Several analyses with different optimization methods such as Genetic Algorithm (GA), Ant Colony Optimization (ACO), and Particle-Swarm Optimization (PSO) algorithms. An effective resource-allocation method is required to reduce the issue of EE (energy-efficiency) and improve the presentation of quality-of-services of the ONs (optical Networks). © 2023 IEEE.

关键词： Genetic algorithms

来源：评论

学校读者我要写书评

暂无评论

Research on the design of multi-source and multi-granularity educational information system based on deep learning

Research on the design of multi-source and multi-granularity...

引用

2024 International Conference on algorithms, High Performance Computing, and Artificial Intelligence, AHPCAI 2024

作者： Wang, Li Shandong Institute of Commerce and Technology Shandong Jinan China

ISBN: (纸本)9781510685840

In the surge of the digital era, the role of educational information systems is increasingly prominent, serving as essential tools to enhance teaching quality and management efficiency. This paper addresses the limitations of traditional educational information systems in handling complex, multi-source data, particularly the challenges in data integration and information correlation, proposing an innovative design approach. Through deep learning technologies, notably a multi-source information correlation method based on convolutional neural networks (CNNs), this study aims to develop a system capable of effectively processing multi-source and multi-granular data within the educational domain. The system design encompasses the entire process from data collection and information correlation to system construction, striving to provide the educational sector with an efficient, intelligent information processing platform. Experimental results demonstrate significant advantages of the system in terms of educational information accuracy and processing speed, promising innovative changes for educational management and pedagogical research. © 2024 SPIE.

关键词： Data modeling Data acquisition Data processing Machine learning Neurons Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Color optical computing: visualization, numbers, alphabet 12

Color optical computing: visualization, numbers, alphabet

引用

12th International Scientific and Practical Conference "information Control systems and Technologies", ICST 2024

作者： Timchenko, victor Kreinovich, Vladik Kondratenko, Yuriy Admiral Makarov National University of Shipbuilding av. Geroev of Ukraine 9 Mykolaiv054025 Ukraine University of Texas at El Paso 550 W University El PasoTX79968 United States Petro Mohyla Black Sea National University 10 68th Desantnykiv Str. Mykolaiv054003 Ukraine Institute of Artificial Intelligence Problems of MES and NAS of Ukraine Kyiv01001 Ukraine

The article is devoted to the development of methods and architecture of optical color computing, techniques of transforming color information for textual representation and numerical calculation, including transmission by optical channels, and thus expanding the functionality of a promising approach to creating artificial intelligence components in decision support systems. This approach is based on the representation of fuzzy information as an information quantum of visible light of a certain color and the construction of an architecture of logical operations based on the additive and subtractive transformation of light radiation using color filters. The paper presents the principles of forming fuzzy input and output color information, algorithms for hybrid calculations with the transformation of optical data into numerical calculus, methods for converting text messages into color information and, further, into numerical encoding. Components of the optical architecture are presented that allow the implementation of transformation algorithms with high computation speed and wide possibilities for constructing parallel structures. © 2024 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).

关键词： optical data processing

来源：评论

学校读者我要写书评

暂无评论

Frequency-domain diffusion adaptation with quantized communication 16

Frequency-domain diffusion adaptation with quantized communi...

引用

16th International Conference on Signal processing systems, ICSPS 2024

作者： Peng, Yishu Zhang, Sheng Zhou, Zhengchun Han, Hongyu School of Information Science and Technology Southwest Jiaotong University Chengdu China School of Computer Science Sichuan Normal University Chengdu China

ISBN: (纸本)9781510689251

In energy- and bandwidth-limited networks, frequency-domain diffusion algorithms based on periodic communication are attractive candidates because of their low computational complexity and communication overhead. Nevertheless, these algorithms assume that intermediate estimates are transmitted exactly in each communication round, which is not realistic in practice. To this end, this paper proposes novel frequency-domain diffusion algorithms with quantized communication. Using the randomized quantizer to compress the shared information, we first develop the quantized frequency-domain diffusion least mean square (Q-FDLMS) algorithm. To mitigate the performance degradation incurred by quantization errors, the improved Q-FDLMS (IQ-FDLMS) algorithm is also designed. This enhanced version modifies the combination step and incorporates the moving-average scheme. Additionally, the mean-square stability conditions and steady-state error bound of the IQ-FDLMS algorithm are derived. Experiment results indicate that our proposed algorithms are superior to other competing algorithms. © 2025 SPIE.

关键词： Frequency domain analysis

来源：评论

学校读者我要写书评

暂无评论

A Rule-Based Expert System for Automated Document Editing 1

引用

20th International Conference on Computing and information Technology (IC2IT2)

作者： Varma, Sandeep Shivam, Shivam Roy, Soumya Deep Ray, Biswarup ZS Associates Pune Pune Maharashtra India

ISBN: (数字)9783031585616

ISBN: (纸本)9783031585609;9783031585616

This research paper introduces a Rule-Based Expert System designed for the automated editing of documents in PDF and PPT formats. The system employs a set of predefined rules, extracted from guideline documents using a Large Language Model (LLM), to execute tasks such as redaction of sensitive text/logo detection and annotation of text elements that deviate from prescribed font size guidelines. Following the detection and annotation process, the system further enhances documents by resizing the detected text elements based on the predefined rules. To achieve these editing tasks, the system integrates advanced image processing techniques, leveraging fine-tuned optical Character Recognition (OCR) for accurate text extraction from document images. Furthermore, Natural Language processing (NLP) algorithms are utilized to analyze and interpret textual content. The combination of image processing, OCR, NLP, and rule extraction using LLM ensures a comprehensive approach to document editing, enhancing efficiency and accuracy. The proposed system addresses the need for automated and rule-driven document editing, contributing to advancements in information security and document standardization.

关键词： Rule-based Expert System Image processing optical Character Recognition Document Sanitization Natural Language processing

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：