检索结果-内蒙古大学图书馆

IEEE International Conference on Electro-Information technology

作者： Mikhail Gromov David Arnold Jafar Saniie Department of Electrical and Computer Engineering Embedded Computing and Signal Processing (ECASP) Research Laboratory Illinois Institute of Technology Chicago IL U.S.A.

computer vision has proven itself capable of accurately detecting and classifying objects within images. This also works in cases where images are used as a way of representing data, without being actual photographs. In cybersecurity, computer vision is rarely used, however it has been used to detect botnets successfully. We applied computer vision to determine how well it would be able to detect and classify a large number of attacks and determined that it would be able to run at a decent rate on a Jetson Nano. This was accomplished by training a convolutional neural network using data publicly available in the IoT-23 database, which contains packet captures of IoT devices with and without different malware infections. The neural network was evaluated on an RTX 3050 and a Jetson Nano to see if it could be used in IoT.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Drowning Recognition for Ocean Surveillance using computer Vision and Drone Control

Drowning Recognition for Ocean Surveillance using Computer V...

引用

IEEE International Conference on Electro-Information technology

作者： Maureen Rakotondraibe Tianyang Fang Jafar Saniie Department of Electrical and Computer Engineering Embedded Computing and Signal Processing (ECASP) Research Laboratory Illinois Institute of Technology Chicago IL U.S.A.

According to WHO's report from 2021, Drowning is the 3rd leading cause of unintentional death worldwide. The use of autonomous drones for drowning recognition can increase the survival rate and help lifeguards and rescuers with their life saving mission. This paper presents a real-time drowning recognition model and algorithm for ocean surveillance that can be implemented on a drone. The presented model has been trained using two different approaches and has 88% accuracy. Compared to the contemporary models of drowning recognition designed for swimming pools, the model presented is better suited for outdoor applications in the ocean.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Heterogeneity Tolerance in IoT Botnet Attack Classification

Heterogeneity Tolerance in IoT Botnet Attack Classification

引用

IEEE International Conference on Electro-Information technology

作者： Samuel Kalenowski David Arnold Mikhail Gromov Jafar Saniie Department of Electrical and Computer Engineering Embedded Computing and Signal Processing (ECASP) Research Laboratory Illinois Institute of Technology Chicago IL U.S.A.

Due to the rapid adoption of Internet of Things (IoT) technologies, many networks are composed of a patchwork of devices designed by different software and hardware developers. In addition to the heterogeneity of IoT networks, the general rush-to-market produced products with poor adherence to core cybersecurity principles. Coupled together, these weaknesses leave organizations vulnerable to attack by botnets, such as Mirai and Gafgyt. Infected devices pose a threat to both internal and external devices as they attempt to add new devices to the collective or to perpetrate targeted attacks within the network or against third parties. Artificial Intelligence (AI) tools for intrusion detection are popular platforms for detecting indicators of botnet infiltration. However, when training AI tools, the heterogeneity of the network hampers detection and classification accuracy due to the differences in device architecture and network layout. To investigate this challenge, we explored the application of a Neural Network (NN) to the N-BaIoT dataset. The NN achieved 94% classification accuracy when trained using data from all devices in the network. Further, we examined the model's transferability by training on a single device and applying it to data from all devices. This resulted in a noticeable decline in classification accuracy. However, when considering cyberattack detection the model retained a very high true positive rate of 99.6%.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Application of Machine Learning and Image Recognition for Driver Attention Monitoring

Application of Machine Learning and Image Recognition for Dr...

引用

IEEE International Conference on Electro-Information technology

作者： Manav Tailor Jahangir Ali Xinrui Yu Won-Jae Yi Jafar Saniie Department of Electrical and Computer Engineering Embedded Computing and Signal Processing (ECASP) Research Laboratory Illinois Institute of Technology Chicago IL U.S.A.

This paper presents a real-time prototype system for monitoring the distraction levels of the driver. Due to the nature of high traffic conditions commonly seen nowadays, accidents are highly likely to occur as drivers cannot always recognize their exhaustion levels themselves. We utilize a single low-cost camera facing the driver connected to a single-board computer; a series of frame captures from the camera are fed to a neural network, and a pattern detection algorithm to predict the driver's distraction level is utilized. All training is conducted under personalized training sets to increase accuracy and to match an individual's driving patterns as accurately as possible. This system is designed to serve as a baseline for further system development, and many vital sub-components can be changed regarding input data type and choices of machine learning algorithms.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Ultrasonic Imaging and Flaw Detection with Optimized Convolutional Transformer Neural Networks

Ultrasonic Imaging and Flaw Detection with Optimized Convolu...

引用

IEEE Symposium (IUS) Ultrasonics

作者： Xin Zhang Jafar Saniie Department of Electrical and Computer Engineering Embedded Computing and Signal Processing (ECASP) Research Laboratory Illinois Institute of Technology Chicago Illinois U.S.A.

Flaw detection using the ultrasonic imaging technique has been widely used for Structural Health Monitoring (SHM). Ultrasonic testing has the advantages of one-sided measurement, high penetration depth, and inspection accuracy. Artificial intelligence (AI) such as Deep Learning (DL) methods can automate the inspection process with high reliability for imaging and SHM. Neural Networks (NNs) and DL methods can detect flaws using ultrasonic images with high accuracy but suffer extensive computational costs for training and deployment. In this study, we propose a lightweight transformer NN, UFDCTNet: Ultrasonic Flaw Detection Convolutional Transformer Neural Network (TNN), optimized with data-efficient Convolutional NN (CNN) for flaw detection using ultrasonic imaging. TNN utilizes the self-attention network architecture that learns global representations and allows high parallelism in computation resulting in reduced training time. CNNs learn local representations with fewer parameters because of inherent spatial inductive bias. This UFDCTNet utilizes the advantages of CNNs to learn spatially local representations resulting in fewer model parameters for fast training. For performance analysis, we trained data-efficient TNN and CNN using ultrasonic images to detect flaws. To examine training results, NNs are trained with the USimgAIST dataset consisting of 7000 experimental B-scan images representing without-flaw and with-flaw cases. A pulsed laser ultrasonic scanning system was used to collect these B-scan images from 17 stainless steel specimen plates with various types of flaws and some plates without any damage.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Evaluation of Homomorphic Encryption for Privacy in Principal Component Analysis

Evaluation of Homomorphic Encryption for Privacy in Principa...

引用

IEEE International Conference on Electro-Information technology

作者： David Arnold Jafar Saniie Department of Electrical and Computer Engineering Embedded Computing and Signal Processing (ECASP) Research Laboratory Illinois Institute of Technology Chicago IL U.S.A.

Principal Component Analysis (PCA) is a versatile Unsupervised Learning (UL) technique for reducing the dimensionality of datasets. As a result, PCA is widely used in consumer and research applications as a preprocessing tool for identifying important features prior to further analysis. In instances where on-site personnel or developers do not have the expertise to apply UL techniques, third party processors are frequently retained. However, the release of client or proprietary data poses a substantial security risk. This risk increases the regulatory and contractual burden on analysts when interacting with sensitive or classified information. Homomorphic Encryption (HE) cryptosystems are a novel family of encryption algorithms that permit approximate addition and multiplication on encrypted data. When applied to UL models, such as PCA, experts may apply their expertise while maintaining data privacy. In order to evaluate the potential application of Homomorphic Encryption, we implemented Principal Component Analysis using the Microsoft SEAL HE libraries. The resulting implementation was applied to the MNIST Handwritten dataset for feature reduction and image reconstruction. Based on our results, HE considerably increased the time required to process the dataset. However, the HE algorithm is still viable for non-real-time applications as it had an average pixel error of near-zero for all image reconstructions.

关键词：

来源：评论

学校读者我要写书评

暂无评论

System-on-Chip Design for Fast Ultrasonic Chirplet Signal Decomposition Algorithm

System-on-Chip Design for Fast Ultrasonic Chirplet Signal De...

引用

IEEE Symposium (IUS) Ultrasonics

作者： Tianyang Fang Austin Fite Mikhail Gromov Jafar Saniie Department of Electrical and Computer Engineering Embedded Computing and Signal Processing (ECASP) Research Laboratory Illinois Institute of Technology Chicago IL U.S.A.

Decomposition and characterization of backscattered ultrasonic multiple interfering echoes is a critical step for analyzing the ultrasonic propagation path, propagation modes, coupling condition of the transducers and detecting defects within the propagation path. Chirplet Signal Decomposition (CSD) is an efficient way of analyzing ultrasonic echoes. However, CSD is computationally expensive and time-consuming for real-time signal processing applications. To address this problem, we present a System-on-Chip (SoC) implementation of the CSD algorithm, with the goal of speed optimization and real-time execution without sacrificing the accuracy of the signal decomposition and reconstruction. The implementation was tested on a Zynq Ultrascale+ series FPGA and achieved an echo estimation within two milliseconds.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Data Compression for Ultrasonic Microstructure Scattering Signals using Unsupervised Neural Networks

Data Compression for Ultrasonic Microstructure Scattering Si...

引用

IEEE Symposium (IUS) Ultrasonics

Ultrasonic Nondestructive Evaluation (NDE) has been extensively used to characterize the microstructure of metallic structures for early exposure of materials integrity. However, industrial NDE requires the processing, storage, and real-time transmission of large volumes of ultrasonic data. Therefore, it is indispensable to compress ultrasonic data with high fidelity. In this study, we explore the development of Unsupervised Learning (UL) based Neural Network (NN) models for massive ultrasonic data compression and an innovative multilayer perceptron residual autoencoder: Ultrasonic Residual Compressive Autoencoder (URCA), is introduced to compress ultrasonic data with high compression performance. This URCA can be fast-trained and utilizes the sparsity penalty with residual connection to optimize compression performance. UL-based NNs allow for memory-efficient training and rapid online augmentation of the model. To examine the results, a high-performance ultrasonic signal acquisition system was assembled to automatically collect ultrasonic data from heat-treated 1,018 steel blocks for microstructure characterization. Compression performance is analyzed based on compression ratio, reconstruction accuracy and model training time. The reconstruction accuracy was measured using the Structural Similarity Index Measure (SSIM) and Peak Signal-to-Noise Ratio (PSNR). By training the URCA NN for a high reconstruction performance of 0.96 SSIM, we obtained 91.25% memory space-saving. For a higher compression performance of 0.80 SSIM, we obtained 96.04% memory space-saving.

关键词：

来源：评论

学校读者我要写书评

暂无评论

IoT-Enabled Smart Bike Helmet with an AI-Driven Collision Avoidance System

IoT-Enabled Smart Bike Helmet with an AI-Driven Collision Av...

引用

IEEE International Conference on Electro-Information technology

作者： Jacob Solus Maureen Rakotondraibe Xinrui Yu Won-Jae Yi Mikhail Gromov Jafar Saniie Department of Electrical and Computer Engineering Embedded Computing and Signal Processing (ECASP) Research Laboratory Illinois Institute of Technology Chicago IL U.S.A.

This paper presents a system design for a smart bike helmet with multiple safety features that are intended to empower bicycle riders to proactively avoid potential sources of danger or injury. A Smart Sensor/Actuator Node (SSAN), driven by an Arduino Uno single-board microcontroller, contains input sensors and actuators to provide riders the ability to send and receive warnings promptly on their helmet. A Vision Node, driven by an NVIDIA Jetson Nano and a cable pin-connected camera, executes AI object detection algorithms for any dangerous objects that are out of sight of the rider and sends alerts to the SSAN as needed. By combining safety features of the SSAN and Vision Node while continuously sending data to an IoT-enabled backend web server, the safety operation of a typical bike ride can be substantially improved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Parallel Read-Write Circuit With Fast Amplitude-Adaptive Matching Scheme to Memristor Crossbar Array

引用

IEEE Transactions on Circuits and Systems I: Regular Papers 2025年

作者： Jiang, Ningye Jiang, Mingxuan Huang, Haoen Yang, Le Huang, Yi Zeng, Zhigang Huazhong University of Science and Technology School of Artificial Intelligence and Automation Key Laboratory of Image Information Processing and Intelligent Control Ministry of Education of China Wuhan430074 China Wuhan Institute of Technology School of Electrical and Information Engineering Wuhan430205 China University of Massachusetts Amherst Department of Electrical and Computer Engineering AmherstMA01003 United States

Memristor crossbar array (MCA) is a computing-in-memory (CIM) module for computational acceleration. However, conventional read-write (R-W) circuits for MCA rely heavily on external components and have shortcoming in long writing times. To address these issues, we propose a parallel R-W circuit with a fast amplitude-adaptive matching (AAM) scheme. The fast AAM scheme is designed to accelerate time of writing memristors in MCA by adaptively adjusting the writing voltage to match an optimal amplitude. The architecture of parallel R-W circuit are outlined with the implementation of R-W processes. The design principles of parallel R-W unit, fast AAM unit, and associated control logic are described for analyzing their functions within the circuit. By integrating the parallel R-W circuit with MCA, a computing block is leveraged in a neural network for classification tasks. The experimental results demonstrate that the neural network with multiple computing blocks maintains high accuracy across various datasets (over 80%). Furthermore, the proposed AAM scheme achieves an average 54% reduction in writing time compared to that one without AAM scheme. Compared to the pulse-width scheme, the parallel R-W circuit exhibits an average 3.05× speedup in adjusting a 5×5 MCA, and the speedup efficiency is higher as the increases of MCA size. © 2025 IEEE.

关键词： Memristors

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：