检索结果-内蒙古大学图书馆

21st International computer Conference on Wavelet Active Media technology and Information Processing, ICCWAMTIP 2024

作者： He, Liu Ao, Ren Jianping, Li College of Computer Science Chongqing University Chongqing401331 China Chongqing Academy of Educational Sciences Chongqing400015 China School of Computer Science and Engineering University of Electronic Science and Technology of China Chengdu611731 China

ISBN: (纸本)9798331519254

This paper studies the Spiking high-dimensional information coding method based on wavelet decomposition technology. Firstly, the wavelet decomposition technique is used to extract the multi-scale features of the input signal to capture the time-frequency characteristics of the signal. Subsequently, the extracted features are mapped to the pulse sequence of the Spiking neural network to achieve efficient coding of highdimensional information. In this paper, the influence of wavelet decomposition parameters on coding performance is analyzed, and the advantages of the proposed method in processing highdimensional data are verified by experiments. The experimental results show that the proposed method can significantly reduce the data dimension and computational complexity while maintaining the accuracy of information, which provides a new effective way for high-dimensional information processing. © 2024 IEEE.

关键词： Network coding

来源：评论

学校读者我要写书评

暂无评论

Design and Implementation of Advanced Driver Assistance Systems

Design and Implementation of Advanced Driver Assistance Syst...

引用

2024 International Telecommunications Conference, ITC-Egypt 2024

作者： Ali, Waleed Abd-Elshafi Saleh, Hatem Abdelhamid Eid, Haneen Salah Fathi, Ahmed Mohamed Mohamed, Abd-El Rahman Amr Higher Institute of Engineering EL- Shorouk Academy Department of Communications and Computer Engineering Cairo Egypt Misr University for Science and Technology Department of Electronics and Communications Engineering Giza Egypt

ISBN: (纸本)9798350351408

Advanced Driver Assistance Systems (ADAS) are a pivotal advancement in automotive technology, enhancing vehicle safety and driving experience. In our project, we have successfully designed and built a robotic prototype car. We have implemented a comprehensive control system for the car's movement, and the following features of ADAS systems: Bump Detection, Precise Lane Tracking, Adaptive Cruise Control, Traffic Sign Recognition, and Blind Spot Detection. Bump detection systems are equipped to detect and alert drivers of speed bumps and other road irregularities, in which we use NVIDIA Jetson Nano, Astra Pro Plus Depth camera as RGB and Depth sensor and use Custom AI Model based on YOLOv5 to detect different Bumps on the road and we use Python, TensorRT, and OpenCV. Lane Departure Warning (LDW) alerts drivers when their vehicle unintentionally drifts out of their lane, we use NVIDIA Jetson Nano as our main microcontroller unit (MCU) responsible for our Machine learning and Astra Pro Plus Depth camera as our RGB sensor, we use Python, Python's libraries such as OpenCV, and NumPy to implement LDW. Adaptive Cruise Control (ACC) adjusts the vehicle's speed automatically to maintain a safe distance from the vehicle ahead, we use NVIDIA Jetson Nano, Astra Pro Plus Depth camera as RGB and Depth sensor and use Custom AI Model based on YOLOv5 to detect different vehicles on the road, we use Python, TensorRT, and OpenCV. Traffic Sign Recognition can interpret and display traffic signs to keep drivers informed about speed limits, no-entry signs, and more, we use NVIDIA Jetson Nano, and Astra Pro Plus Depth camera as our RGB sensor and use Custom AI Model based on YOLOv5 to detect different Traffic Signs on the road, we use Python, TensorRT, and OpenCV. Blind Spot Detection (BSD) technology helps drivers identify vehicles in their blind spots and alerts them to avoid dangerous maneuvers, we use ESP32 as an MCU and Ultrasonic sensor, and use C++ to program the BSD feature. © 2024

关键词： Advanced driver assistance systems

来源：评论

学校读者我要写书评

暂无评论

WiFi for Privacy-Friendly Surveillance 15

WiFi for Privacy-Friendly Surveillance

引用

15th International Conference on Computing Communication and Networking Technologies, ICCCNT 2024

作者： Ghosh, Akhilbaran Chouksey, Arjun Academy Of Technology Department of Computer Science and Engineering Adisaptagram West Bengal Hooghly India Jabalpur Engineering College Department of Electronics & Communications Engineering Madhya Pradesh Jabalpur India

ISBN: (纸本)9798350370249

In the ongoing debate between surveillance necessity and privacy concerns, WiFi sensing emerges as a promising solution for efficient data monitoring while safeguarding personal privacy. Unlike traditional surveillance methods relying on camera recordings, WiFi sensing utilizes Channel State Information (CSI) data to detect specific movements, offering advantages such as reduced data storage and processing time. This paper explores the effectiveness of WiFi CSI data in detecting human interactions within radio waves, highlighting its practical usability and enhanced accuracy compared to Received Signal Strength (RSS) measurements. Leveraging Convolutional Neural Networks (CNNs), a blind machine learning paradigm, this study demonstrates the potential of WiFi sensing for human activity recognition (HAR). While previous HAR research predominantly focused on camera feeds and wearable devices, this study pioneers the application of CNNs to WiFi CSI data analysis, showing promising results for occupancy detection, activity recognition, and gesture identification. Although there is room for further improvement in accuracy, this research lays a solid foundation for future studies in the burgeoning field of WiFi sensing-based surveillance. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

RISC-V Multicore for Miniature DNA Sequencers

RISC-V Multicore for Miniature DNA Sequencers

引用

2023 IEEE Biomedical Circuits and Systems Conference, BioCAS 2023

作者： Beyene, Abel Wu, ZhongPan Hammad, Karim Ghafar-Zadeh, Ebrahim Magierowski, Sebastian York University Electrical Engineering and Computer Science Toronto Canada Technology and Maritime Transport Arab Academy for Science Cairo Egypt

ISBN: (纸本)9798350300260

Hand-sized DNA sequencers have demonstrated excellent scientific utility. They also present exciting opportunities for the expansion of DNA analysis to a diverse range of uses and users. But this potential is currently blunted by mini-sequencers' lack of on-board computing. In this paper we explore the potential of embedded processors to address this gap. We outline the general hardware and software organization of existing systems and identify a deep learning bioinformatics task as an exemplary workload for processors embedded in miniature sequencers. We then describe a candidate multicore RISC-V arrangement and experimentally study the relative performance, power, and cost trade-offs for alternate configurations of this arrangement. The experiments show no saturation in performance scaling up to at least 6 cores with maximum throughput and efficiency achieved with 512KB and 128KB L2 cache sizes respectively. © 2023 IEEE.

关键词： DNA sequencing embedded computing machine learning multicore RISC-V

来源：评论

学校读者我要写书评

暂无评论

High-integrated photonic tensor core utilizing high-dimensional lightwave and microwave multidomain multiplexing

引用

Light(science & Applications) 2025年第1期14卷 227-237页

作者： Xiangyan Meng Nuannuan Shi Guojie Zhang Junshen Li Ye Jin Shiyou Sun Yichen Shen Wei Li Ninghua Zhu Ming Li Key Laboratory of Optoelectronic Materials and Devices Institute of SemiconductorsChinese Academy of SciencesBeijing100083China College of Materials Science and Opto-Electronic Technology University of Chinese Academy of SciencesBeijing100049China School of Electronic Electrical and Communication EngineeringUniversity of Chinese Academy of SciencesBeijing100049China China Academy of Space Technology(Xi’an) Xi’anShaanxi710100China WeChat Pay Lab 33 Shenzhen Tencent Computer System Co.Ltd.Shenzhen518054China Lightelligence Group Hangzhou311121China

The burgeoning volume of parameters in artificial neural network models has posed substantial challenges to conventional tensor computing *** from the available optical multidimensional information entropy,optical intelligent computing is used as an alternative solution to address the emerging challenges of electrical *** limitations,in terms of device size and photonic integration scale,have hindered the performance of optical ***,an ultrahigh computing density optical tensor processing unit(OTPU),which is grounded in an individual microring resonator(MRR),is introduced to respond to these *** the independent tuning of multiwavelength lasers,the operational capabilities of an MRR are orchestrated,culminating in the formation of an optical tensor *** design facilitates the execution of tensor convolution operations via the lightwave and microwave multidomain hybrid multiplexing in terms of the time,wavelength,and frequency of *** experimental results for the MRR-based OTPU show an extraordinary computing density of 34.04 TOPS/***,the achieved accuracy rate in recognizing MNIST handwritten digits was 96.41%.These outcomes signify a significant advancement toward the realization of high-performance optical tensor processing chips.

关键词： tuning resonator tensor

来源：评论

学校读者我要写书评

暂无评论

Fortaleza: The Emergence of a Network Hub

引用

arXiv 2023年第1期3卷 272-279页

作者： Bragion, Eric Akter, Habiba Kumar, Mohit Xu, Minxian Abdelmoniem, Ahmed M. Gill, Sukhpal Singh School of Electronic Engineering and Computer Science Queen Mary University of London United Kingdom Department of Information Technology National Institute of Technology Jalandhar India Shenzhen Institute of Advanced Technology Chinese Academy of Sciences Shenzhen China

Digitalisation, accelerated by the pandemic, has brought the opportunity for companies to expand their businesses beyond their geographic location and has considerably affected networks around the world. Cloud services have a better acceptance nowadays, and it is foreseen that this industry will grow exponentially in the following years. With more distributed networks that need to support customers in different locations, the model of one-single server in big financial centres has become outdated and companies tend to look for alternatives that will meet their needs, and this seems to be the case with Fortaleza, in Brazil. With several submarine cables connections available, the city has stood out as a possible hub to different regions, and this is what this paper explores. Making use of real traffic data through looking glasses, we established a latency classification that ranges from exceptionally low to high and analysed 800 latencies from Roubaix, Fortaleza and Sao Paulo to Miami, Mexico City, Frankfurt, Paris, Milan, Prague, Sao Paulo, Santiago, Buenos Aires and Luanda. We found that non-developed countries have a big dependence on the United States to route Internet traffic. Despite this, Fortaleza proves to be an alternative for serving different regions with relatively low latencies. © 2023, CC0.

关键词： Cloud data centre Latency Network hubs Fortaleza Roubaix Networking

来源：评论

学校读者我要写书评

暂无评论

Balancing communication overhead and accuracy in compression integration: a survey

引用

Journal of Supercomputing 2025年第8期81卷 1-63页

作者： Yang, Aiqiang Liu, Jie Yang, Bo Mo, Zeyao Li, Keqin College of Computer Science and Technology National University of Defense Technology Changsha410000 China Laboratory of Digitizing Software for Frontier Equipment National University of Defense Technology Changsha410073 China National Key Laboratory of Parallel and Distributed Computing National University of Defense Technology Changsha410073 China China Academy of Engineering Physics Beijing China Department of Computer Science State University of New York New PaltzNY12561 United States

In large-scale distributed training, communication compression techniques are widely used to reduce the significant communication overhead caused by the frequent exchange of model parameters or gradients between training nodes. However, these techniques often introduce additional computational complexity and may lead to data loss, thereby affecting model convergence and performance. This review examines key optimization methods in communication compression, including pruning techniques that remove irrelevant weights, quantization techniques that convert floating-point parameters to low-precision representations, and sparsification techniques that transmit only critical gradients. Low-rank approximation techniques, which compress parameters through matrix factorization, are particularly useful for large-scale models. These techniques have also been widely applied in various application scenarios, demonstrating their effectiveness in different environments. Application scenarios include distributed training, federated learning, and edge computing, where bottlenecks are carefully identified and evaluated in common scenarios, providing a basis for further optimization. Future development directions emphasize co-design of hardware and algorithms, dynamic strategies, and cross-layer optimization. This study provides valuable comparisons of key methods and theoretical analysis for efficient communication compression in distributed systems. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2025.

关键词： Edge computing

来源：评论

学校读者我要写书评

暂无评论

Pendant Graphs: Unveiling Regularity, Irregularity, and Support Vertices

引用

IAENG International Journal of Applied Mathematics 2025年第1期55卷 45-51页

作者： Mathew, Suji Elizabeth Kalayathankal, Sunny Joseph Thankachan, Baiju Mathew, Bibin Department of Mathematics Catholicate College Kerala Pathanamthitta689643 India Department of Computer Science and Engineering Rajagiri School of Engineering and Technology Kerala Kochi682039 India Department of Mathematics Manipal Institute of Technology Manipal Academy of Higher Education Karnataka Manipal576104 India Department of Education National Institute of Technology Calicut Kerala 673601 India

This paper investigates Pendant Graphs, which are graphs that contain pendant vertices. It introduces a new perspective on the regularity and irregularity of pendant graphs, and discusses their relationship with Highly Irregular Graphs. Additionally, this work introduces a novel graph based on the concept derived from (m, k) Regular Graphs. The (m, k) regular graphs are closely related to Pendant Graph. The paper also delves into the nature and characteristics of pendant vertices. By exploring the characteristics of pendant vertices and defining the Cyclic Dendrimer Graph, the paper significantly contributes to both Graph Theory and Chemical Graph Theory. © (2025), (International Association of Engineers). All rights reserved.

关键词： Graph theory

来源：评论

学校读者我要写书评

暂无评论

An Enhanced Approach to Intelligent computer-Assisted Localization of Liver Tumor on Computed Tomography Images

引用

SN computer science 2024年第7期5卷 809页

作者： Arakeri, Megha Lakshmana Reddy, Raghavendra Ravishankar, H. Deepa, K.R. Department of Computer Science and Engineering Manipal Institute of Technology Bengaluru Manipal Academy of Higher Education Karnataka Manipal India Seamovations Labs Pvt Ltd Karnataka Bengaluru India School of Computer Science and Engineering REVA University Karnataka Bengaluru India School of Computer Science and Information Technology REVA University Karnataka Bengaluru India School of Electrical and Electronics Engineering REVA University Karnataka Bengaluru India

Computed Tomography (CT) Imaging is frequently used to find liver cancer. CT imaging of the liver generates cross-sectional images of the abdominal region. The task of segmenting a liver tumor on CT images is tedious due to the anatomic complexity of the liver. Therefore, liver tumor detection has been performed manually or semi-automatically which is an exhaustive process. Further, a tumor may be present on only some slices of the CT scan and hence there is a need to automatically identify images that contain tumors from acquired CT images and then locate a region of the tumor on those images for further diagnosis. Automatic detection helps the radiologist to obtain fast, accurate, and effective results. Accordingly, in this paper, an enhanced approach is proposed based on deep learning, efficient fuzzy c-means (EFCM) clustering, and a region-based algorithm for the detection of a liver tumor. A deep learning model built on a convolutional neural network is used to identify whether the CT image contains a tumor. If it contains the tumor, then the proposed EFCM and region-based algorithms are applied to locate the tumor on the CT image. The EFCM algorithm is combined with a region-based technique to automatically determine the initial pixel and threshold values required to segment the liver tumor. The proposed method is applied to several CT abdominal images. The results of the tumor segmentation are validated by comparing them with the tumor regions detected by radiologists. The estimated results of the experiment show that the proposed technique successfully detects the liver tumor on the CT image with less than 2% relative error. © The Author(s) 2024.

关键词： Computed tomography computer-aided detection Healthcare Liver tumor Medical image Segmentation

来源：评论

学校读者我要写书评

暂无评论

Bypassing LLM Safeguards: The In-Context Tense Attack Approach 14th

Bypassing LLM Safeguards: The In-Context Tense Attack Approa...

引用

14th International Conference on computer engineering and Networks, CENet 2024

作者： Wang, Shaohuang Geng, Ruijing Lei, Shuai Lv, Yanfei Zhang, Huaping School of computer science Xinjiang University Urumqi China Space Engineering University Beijing China Academy of Military Science Beijing China Beijing Institute of Technology Beijing China

ISBN: (纸本)9789819642441

In the field of cybersecurity, the emergence of Large Language Models (LLMs) has opened up a new domain of potential risks. Although these models show an impressive level of capability across a range of applications, they are not free from the possibility of creating content that could negatively impact both social and digital safety. This paper examines the implications of In-Context Learning based command attacks, a burgeoning threat to the security and ethical integrity of LLMs. We introduce the In-Context Tense Attack(ITA) framework, a novel approach that employs harmful examples to undermine the integrity of LLMs. Our theoretical analysis elucidates how a constrained set of context examples can significantly influence the security mechanisms of LLMs. Through rigorous experimentation, we have substantiated the potency of ITA in elevating the success rate of jailbreaking prompts. On the PKU-Alignment/SafeRLHF dataset, ITA achieved a remarkable 92.99% increase in Accuracy, a 73.36% improvement in Rouge-L, and a 27.01% enhancement in Bleu-4 scores. Similarly, on the NVIDIA/Aegis-Safety dataset, ITA demonstrated a 72.03% increase in Accuracy, an 80.87% rise in Rouge-L, and a 40.24% boost in Bleu-4 scores. These results underscore the effectiveness of ITA in manipulating LLMs to generate harmful outputs, thereby highlighting the necessity for more robust security measures. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Cyber attacks

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：