检索结果-内蒙古大学图书馆

Energy-Efficient Fast Object Detection on Edge Devices for IoT Systems

IEEE Internet of Things Journal 2025年第11期12卷 16681-16694页

作者： Nurul Achmadiah, Mas Ahamad, Afaroj Sun, Chi-Chia Kuo, Wen-Kai National Formosa University SMIM Research Center Department of Electro-Optics Huwei632 Taiwan Yuan Ze University Department of Computer Science and Engineering Taoyuan320 Taiwan National Taipei University Department of Electrical Engineering Taipei237303 Taiwan National Formosa University Department of Electro-Optics Huwei632 Taiwan

This article presents an Internet of Things (IoT) application that utilizes an AI classifier for fast-object detection using the frame difference method. This method, with its shorter duration, is the most efficient and suitable for fast-object detection in IoT systems, which require energy-efficient applications compared to end-to-end methods. We have implemented this technique on three edge devices: 1) AMD AlveoTMU50;2) Jetson Orin Nano;and 3) Hailo- 8TM AI Accelerator, and four models with artificial neural networks and transformer models. We examined various classes, including birds, cars, trains, and airplanes. Using the frame difference method, the MobileNet model consistently has high accuracy, low latency, and is highly energy-efficient. YOLOX consistently shows the lowest accuracy, lowest latency, and lowest efficiency. The experimental results show that the proposed algorithm has improved the average accuracy gain by 28.314%, the average efficiency gain by 3.6 times, and the average latency reduction by 39.305% compared to the end-to-end method. Of all these classes, the faster objects are trains and airplanes. Experiments show that the accuracy percentage for trains and airplanes is lower than other categories. So, in tasks that require fast detection and accurate results, end-to-end methods can be a disaster because they cannot handle fast object detection. To improve computational efficiency, we designed our proposed method as a lightweight detection algorithm. It is well suited for applications in IoT systems, especially those that require fast-moving object detection and higher accuracy. © 2014 IEEE.

关键词： Neural networks

来源：评论

学校读者我要写书评

暂无评论

Fast Readout of Innovative Glow Discharge Detector Focal Plane Arrays for Affordable Millimeter Wave Imaging: Design, Implementation, and Advancements 9

Fast Readout of Innovative Glow Discharge Detector Focal Pla...

引用

Emerging Imaging and Sensing Technologies for Security and Defence IX 2024

作者： Ramachandra Kurup, Arun Rozban, Daniel Azulay, Yehuda Kopeika, Natan S. Yitzhaky, Yitzhak Abramovich, Amir Department of Electrical and Electronic Engineering Ariel University Israel School of Electrical and Computer Engineering Department of Electro-optics and Photonics Engineering Ben-Gurion University of the Negev Beer Sheva Israel

ISBN: (数字)9781510681170

ISBN: (纸本)9781510681163

This study introduces an affordable approach to developing a millimeter-wave (MMW) focal plane array (FPA) using glow discharge detectors (GDDs) with rapid readout capabilities. The system investigates the effect of MMW incidence on GDD discharge current and employs a scanning mechanism with a step motor for sub-pixel imaging. The experimental setup features an 8x8 FPA with 64 GDD detectors, MMW source, optical components, and a data acquisition platform. Leveraging electrical detection and a signal processing algorithm, the system captures low-intensity MMW signals with each GDD value corresponding to a pixel in the image. The Fourier transform (FFT) is used to extract information from the modulated signals. The FPA demonstrated an improved imaging duration, reducing acquisition time by 75% compared to prior studies, with an 8x8 image captured in just 4 seconds. Responsivity was measured at 21.8 V/W, detecting signals as weak as 0.95 nW. These advancements offer significant benefits for time-sensitive applications like security and dynamic object tracking. The study proposes enhancing resolution through larger arrays and integrating image processing with AI techniques to further improve performance. This research provides a cost-effective solution for MMW imaging, addressing existing detector limitations. © 2024 SPIE.

关键词： Glow discharges

来源：评论

学校读者我要写书评

暂无评论

Photonic orbital angular momentum with controllable orientation

引用

National Science Review 2022年第7期9卷 56-62页

作者： Chenhao Wan Jian Chen Andy Chong Qiwen Zhan School of Optical-Electrical and Computer Engineering University of Shanghai for Science and Technology School of Optical and Electronic Information Huazhong University of Science and Technology Department of Electro-Optics and Photonics University of Dayton Department of Physics University of Dayton

Vortices are whirling disturbances,commonly found in nature,ranging from tremendously small scales in Bose-Einstein condensations to cosmologically colossal scales in spiral *** optical vortex,generally associated with a spiral phase,can carry orbital angular momentum(OAM).The optical OAM can either be in the longitudinal direction if the spiral phase twists in the spatial domain or in the transverse direction if the phase rotates in the spatiotemporal *** this article,we demonstrate the intersection of spatiotemporal vortices and spatial vortices in a wave *** a result of this intersection,the wave packet hosts a tilted OAM that provides an additional degree of freedom to the applications that harness the OAM of photons.

关键词： orbital angular momentum spatiotemporal vortex spatial vortex spiral phase

来源：评论

学校读者我要写书评

暂无评论

Hybrid integration of silicon slot photonics with ferroelectric nematic liquid crystal for poling-free Pockels-effect modulation

Hybrid integration of silicon slot photonics with ferroelect...

引用

2024 Conference on Lasers and electro-optics, CLEO 2024

作者： Onural, Deniz Wang, Imbert Chiang, Li-Yuan Raja, Shiva Zhang, Xinchang Singh, Manuj Li, Dingning Dao, Howard Pajk, Spencer Sickler, Jason W. Pecinovsky, Cory Popović, Miloš A. Department of Electrical and Computer Engineering Photonics Center Boston University BostonMA02215 United States Polaris Electro-Optics Inc. BroomfieldCO80020 United States

Recently discovered ferroelectric nematic liquid crystal promises to enable both strong phase shifters and fast Pockels' effect modulation without poling. We demonstrate the first silicon slot-waveguide hybrid int... 详细信息

ISBN: (纸本)9781957171395

关键词： Nematic liquid crystals

来源：评论

学校读者我要写书评

暂无评论

Hybrid integration of silicon slot photonics with ferroelectric nematic liquid crystal for poling-free Pockels-effect modulation

Hybrid integration of silicon slot photonics with ferroelect...

引用

CLEO: Science and Innovations in CLEO 2024, CLEO: S and I 2024 - Part of Conference on Lasers and electro-optics

关键词： Nematic liquid crystals

来源：评论

学校读者我要写书评

暂无评论

Unsupervised Homography Estimation on Multimodal Image Pair via Alternating Optimization 38

Unsupervised Homography Estimation on Multimodal Image Pair ...

引用

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Song, Sanghyeob Lew, Jaihyun Jang, Hyemi Yoon, Sungroh Interdisciplinary Program in Artificial Intelligence Seoul National University Korea Republic of Department of Electrical and Computer Engineering Seoul National University Korea Republic of Samsung Electro-Mechanics Korea Republic of

Estimating the homography between two images is crucial for mid- or high-level vision tasks, such as image stitching and fusion. However, using supervised learning methods is often challenging or costly due to the difficulty of collecting ground-truth data. In response, unsupervised learning approaches have emerged. Most early methods, though, assume that the given image pairs are from the same camera or have minor lighting differences. Consequently, while these methods perform effectively under such conditions, they generally fail when input image pairs come from different domains, referred to as multimodal image pairs. To address these limitations, we propose AltO, an unsupervised learning framework for estimating homography in multimodal image pairs. Our method employs a two-phase alternating optimization framework, similar to Expectation-Maximization (EM), where one phase reduces the geometry gap and the other addresses the modality gap. To handle these gaps, we use Barlow Twins loss for the modality gap and propose an extended version, Geometry Barlow Twins, for the geometry gap. As a result, we demonstrate that our method, AltO, can be trained on multimodal datasets without any ground-truth data. It not only outperforms other unsupervised methods but is also compatible with various architectures of homography estimators. The source code can be found at: https://***/songsang7/AltO. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Sensitivity Enhancement of THz Metamaterial by Reduction of the Fabry-Pérot Oscillations and Decoupling its Resonance from Substrate Losses

Sensitivity Enhancement of THz Metamaterial by Reduction of ...

引用

International Conference on Infrared and Millimeter Waves

作者： Heena Khand Rudrarup Sengupta Gabby Sarusi Department of Photonics and Electro-Optics Engineering School of Electrical and Computer Engineering Ben-Gurion University of the Negev Beer Sheva Israel

ISBN: (数字)9798350370324

ISBN: (纸本)9798350370331

We demonstrate an optical method for sensitivity enhancement for LC resonant MM’s, using meta-plate which act as plasmonic enhancer by increasing the electric field concentration in the active area of the metasurface. This makes the MM resonance ultra-sensitive to the tiny changes of particle size/concentration under test spread on the metasurface, contributing to enhanced ΔF. The meta-plate also makes the Si substrate optically lossless and transparent in THz, enabling the full effect of MM resonance without any substrate losses in the transmission spectra. This work also demonstrates an extended concept of de-coupling MM-resonance from the substrate’s Fabry-Pérot (FP) oscillations by de-trapping of the THz radiation from the MM substrate, resulting in the improved quality-factor of the MM resonance and overall plasmonic enhancement on the metasurface. The plasmonic enhancer meta-plate increases the ΔF by eight-fold compared to MM’s fabricated on conventional Si substrates.

关键词： Sensitivity Terahertz radiation Plasmons Metasurfaces Propagation losses Resonance Silicon Optical sensors Substrates Oscillators

来源：评论

学校读者我要写书评

暂无评论

Breathalyzer-based Prompt Screening and Diagnosis of Respiratory Diseases using Metamaterial Based Terahertz Impedance Spectroscopy of Viruses

Breathalyzer-based Prompt Screening and Diagnosis of Respira...

引用

International Conference on Infrared and Millimeter Waves

作者： Rudrarup Sengupta Heena Khand Gabby Sarusi Department of Photonics and Electro-Optics Engineering School of Electrical and Computer Engineering Ben-Gurion University of the Negev Beer Sheva Israel

ISBN: (数字)9798350370324

ISBN: (纸本)9798350370331

We present breathalyzer-based prompt screening technology that can detect and screen multiple respiratory diseases using exhaled breath of a patient, collected on an LC resonant metasurface, demonstrating a spectral red-shift (ΔF). We categorized ΔF for multiple respiratory diseases, which do not overlap. This was physically possible by removing the constraint of detection limit in metamaterials, along with significant sensitivity enhancement physics.

关键词： Spectroscopy Sensitivity Pulmonary diseases Metasurfaces Metamaterials Impedance Physics

来源：评论

学校读者我要写书评

暂无评论

Laguerre-Gaussian Beam Propagation through Anisotropic Turbulence: A Comparison of the Extended Huygens-Fresnel Principle and Perturbation Method

Laguerre-Gaussian Beam Propagation through Anisotropic Turbu...

引用

2024 Applications of Lasers for Sensing and Free Space Communications, LS and C 2024

作者： Ghanati, Elaheh van Iersel, Miranda University of Dayton Department of Electro-Optics and Photonics 300 College Park DaytonOH45469 United States New Mexico State University Klipsch School of Electrical and Computer Engineering 1780 E University Avenue Las CrucesNM88003 United States

A Laguerre-Gaussian beam is propagated through anisotropic atmospheric turbulence using the extended-Huygens-Fresnel principle and perturbation method. The normalized intensity profile and spot size are calculated usi... 详细信息

关键词： Gaussian beams

来源：评论

学校读者我要写书评

暂无评论

Imperceptible CMOS camera dazzle for adversarial attacks on deep neural networks

arXiv

引用

arXiv 2023年

作者： Stein, Zvi Stern, Adrian Ben Gurion University Electro-Optics Engineering Department School of Electrical and Computer Engineering Beer Sheva Israel

Despite the outstanding performance of deep neural networks, they are vulnerable to adversarial attacks. While there are many invisible attacks in the digital domain, most physical world adversarial attacks are visible. Here we present an invisible optical adversarial attack that uses a light source to dazzle a CMOS camera with a rolling shutter. We present the photopic conditions required to keep the attacking light source completely invisible while sufficiently jamming the captured image so that a deep neural network applied to it is deceived. © 2023, CC BY-SA.

关键词： Deep neural networks

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：