Sangho Jo, Ohtae Jang, Chaitali Bhattacharyya, Minjun Kim, Taeseok Lee, Yewon Jang, Haekang Song, Hyukmin Kwon, Saebyeol Do, Sungho Kim
Several studies in computer vision have examined specular removal, which is crucial for object detection and recognition. This research has traditionally been divided into two tasks: specular highlight removal, which focuses on removing specular highlights on object surfaces, and reflection removal, which deals with specular reflections occurring on glass surfaces. In reality, however, both types of specular effects often coexist, making it a fundamental challenge that has not been adequately addressed. Recognizing the necessity of integrating specular components handled in both tasks, we constructed a specular-light (S-Light) DB for training single-image-based deep learning models. Moreover, considering the absence of benchmark datasets for quantitative evaluation, the multi-scale normalized cross correlation (MS-NCC) metric, which considers the correlation between specular and diffuse components, was introduced to assess the learning outcomes.
{"title":"S-LIGHT: Synthetic Dataset for the Separation of Diffuse and Specular Reflection Images","authors":"Sangho Jo, Ohtae Jang, Chaitali Bhattacharyya, Minjun Kim, Taeseok Lee, Yewon Jang, Haekang Song, Hyukmin Kwon, Saebyeol Do, Sungho Kim","doi":"10.3390/s24072286","DOIUrl":"https://doi.org/10.3390/s24072286","url":null,"abstract":"Several studies in computer vision have examined specular removal, which is crucial for object detection and recognition. This research has traditionally been divided into two tasks: specular highlight removal, which focuses on removing specular highlights on object surfaces, and reflection removal, which deals with specular reflections occurring on glass surfaces. In reality, however, both types of specular effects often coexist, making it a fundamental challenge that has not been adequately addressed. Recognizing the necessity of integrating specular components handled in both tasks, we constructed a specular-light (S-Light) DB for training single-image-based deep learning models. Moreover, considering the absence of benchmark datasets for quantitative evaluation, the multi-scale normalized cross correlation (MS-NCC) metric, which considers the correlation between specular and diffuse components, was introduced to assess the learning outcomes.","PeriodicalId":221960,"journal":{"name":"Sensors (Basel, Switzerland)","volume":"49 3","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140771327","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Hironobu Wakabayashi, Y. Hiroi, Kenzaburo Miyawaki, Akinori Ito
This paper proposes a novel tour guide robot, “ASAHI ReBorn”, which can lead a guest by hand one-on-one while maintaining a proper distance from the guest. The robot uses a stretchable arm interface to hold the guest’s hand and adjusts its speed according to the guest’s pace. The robot also follows a given guide path accurately using the Robot Side method, a robot navigation method that follows a pre-defined path quickly and accurately. In addition, a control method is introduced that limits the angular velocity of the robot to avoid the robot’s quick turn while guiding the guest. We evaluated the performance and usability of the proposed robot through experiments and user studies. The tour-guiding experiment revealed that the proposed method that keeps distance between the robot and the guest using the stretchable arm enables the guests to look around the exhibits compared with the condition where the robot moved at a constant velocity.
{"title":"Development of a Personal Guide Robot That Leads a Guest Hand-in-Hand While Keeping a Distance","authors":"Hironobu Wakabayashi, Y. Hiroi, Kenzaburo Miyawaki, Akinori Ito","doi":"10.3390/s24072345","DOIUrl":"https://doi.org/10.3390/s24072345","url":null,"abstract":"This paper proposes a novel tour guide robot, “ASAHI ReBorn”, which can lead a guest by hand one-on-one while maintaining a proper distance from the guest. The robot uses a stretchable arm interface to hold the guest’s hand and adjusts its speed according to the guest’s pace. The robot also follows a given guide path accurately using the Robot Side method, a robot navigation method that follows a pre-defined path quickly and accurately. In addition, a control method is introduced that limits the angular velocity of the robot to avoid the robot’s quick turn while guiding the guest. We evaluated the performance and usability of the proposed robot through experiments and user studies. The tour-guiding experiment revealed that the proposed method that keeps distance between the robot and the guest using the stretchable arm enables the guests to look around the exhibits compared with the condition where the robot moved at a constant velocity.","PeriodicalId":221960,"journal":{"name":"Sensors (Basel, Switzerland)","volume":"324 12","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140779666","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Linglong Zhou, Guoxin Wu, Yunbo Zuo, Xuanyu Chen, Hongle Hu
With the rapid development of 3D reconstruction, especially the emergence of algorithms such as NeRF and 3DGS, 3D reconstruction has become a popular research topic in recent years. 3D reconstruction technology provides crucial support for training extensive computer vision models and advancing the development of general artificial intelligence. With the development of deep learning and GPU technology, the demand for high-precision and high-efficiency 3D reconstruction information is increasing, especially in the fields of unmanned systems, human-computer interaction, virtual reality, and medicine. The rapid development of 3D reconstruction is becoming inevitable. This survey categorizes the various methods and technologies used in 3D reconstruction. It explores and classifies them based on three aspects: traditional static, dynamic, and machine learning. Furthermore, it compares and discusses these methods. At the end of the survey, which includes a detailed analysis of the trends and challenges in 3D reconstruction development, we aim to provide a comprehensive introduction for individuals who are currently engaged in or planning to conduct research on 3D reconstruction. Our goal is to help them gain a comprehensive understanding of the relevant knowledge related to 3D reconstruction.
{"title":"A Comprehensive Review of Vision-Based 3D Reconstruction Methods","authors":"Linglong Zhou, Guoxin Wu, Yunbo Zuo, Xuanyu Chen, Hongle Hu","doi":"10.3390/s24072314","DOIUrl":"https://doi.org/10.3390/s24072314","url":null,"abstract":"With the rapid development of 3D reconstruction, especially the emergence of algorithms such as NeRF and 3DGS, 3D reconstruction has become a popular research topic in recent years. 3D reconstruction technology provides crucial support for training extensive computer vision models and advancing the development of general artificial intelligence. With the development of deep learning and GPU technology, the demand for high-precision and high-efficiency 3D reconstruction information is increasing, especially in the fields of unmanned systems, human-computer interaction, virtual reality, and medicine. The rapid development of 3D reconstruction is becoming inevitable. This survey categorizes the various methods and technologies used in 3D reconstruction. It explores and classifies them based on three aspects: traditional static, dynamic, and machine learning. Furthermore, it compares and discusses these methods. At the end of the survey, which includes a detailed analysis of the trends and challenges in 3D reconstruction development, we aim to provide a comprehensive introduction for individuals who are currently engaged in or planning to conduct research on 3D reconstruction. Our goal is to help them gain a comprehensive understanding of the relevant knowledge related to 3D reconstruction.","PeriodicalId":221960,"journal":{"name":"Sensors (Basel, Switzerland)","volume":"32 5","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140770438","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
An on-site InSAR imaging method carried out with unmanned aerial vehicles (UAVs) is proposed to monitor terrain changes with high spatial resolution, short revisit time, and high flexibility. To survey and explore a specific area of interest in real time, a combination of a least-square phase unwrapping technique and a mean filter for removing speckles is effective in reconstructing the terrain profile. The proposed method is validated by simulations on three scenarios scaled down from the high-resolution digital elevation models of the US geological survey (USGS) 3D elevation program (3DEP) datasets. The efficacy of the proposed method and the efficiency in CPU time are validated by comparing with several state-of-the-art techniques.
提出了一种利用无人飞行器（UAVs）进行现场 InSAR 成像的方法，以高空间分辨率、短重访时间和高灵活性监测地形变化。为了实时勘测和探索感兴趣的特定区域，将最小平方相位解包技术与去除斑点的均值滤波器相结合，可有效重建地形剖面。通过对美国地质调查局（USGS）三维高程计划（3DEP）数据集的高分辨率数字高程模型缩小的三个场景进行模拟，验证了所提出的方法。通过与几种最先进的技术进行比较，验证了所提方法的功效和 CPU 时间效率。
{"title":"An On-Site InSAR Terrain Imaging Method with Unmanned Aerial Vehicles","authors":"Hsu-Yueh Chuang, Jean‐Fu Kiang","doi":"10.3390/s24072287","DOIUrl":"https://doi.org/10.3390/s24072287","url":null,"abstract":"An on-site InSAR imaging method carried out with unmanned aerial vehicles (UAVs) is proposed to monitor terrain changes with high spatial resolution, short revisit time, and high flexibility. To survey and explore a specific area of interest in real time, a combination of a least-square phase unwrapping technique and a mean filter for removing speckles is effective in reconstructing the terrain profile. The proposed method is validated by simulations on three scenarios scaled down from the high-resolution digital elevation models of the US geological survey (USGS) 3D elevation program (3DEP) datasets. The efficacy of the proposed method and the efficiency in CPU time are validated by comparing with several state-of-the-art techniques.","PeriodicalId":221960,"journal":{"name":"Sensors (Basel, Switzerland)","volume":"70 19","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140795262","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Leonardo A. Fagundes, Alexandre G. Caldeira, Matheus B. Quemelli, Felipe N. Martins, A. Brandão
In mobile robotics, LASER scanners have a wide spectrum of indoor and outdoor applications, both in structured and unstructured environments, due to their accuracy and precision. Most works that use this sensor have their own data representation and their own case-specific modeling strategies, and no common formalism is adopted. To address this issue, this manuscript presents an analytical approach for the identification and localization of objects using 2D LiDARs. Our main contribution lies in formally defining LASER sensor measurements and their representation, the identification of objects, their main properties, and their location in a scene. We validate our proposal with experiments in generic semi-structured environments common in autonomous navigation, and we demonstrate its feasibility in multiple object detection and identification, strictly following its analytical representation. Finally, our proposal further encourages and facilitates the design, modeling, and implementation of other applications that use LASER scanners as a distance sensor.
{"title":"Analytical Formalism for Data Representation and Object Detection with 2D LiDAR: Application in Mobile Robotics","authors":"Leonardo A. Fagundes, Alexandre G. Caldeira, Matheus B. Quemelli, Felipe N. Martins, A. Brandão","doi":"10.3390/s24072284","DOIUrl":"https://doi.org/10.3390/s24072284","url":null,"abstract":"In mobile robotics, LASER scanners have a wide spectrum of indoor and outdoor applications, both in structured and unstructured environments, due to their accuracy and precision. Most works that use this sensor have their own data representation and their own case-specific modeling strategies, and no common formalism is adopted. To address this issue, this manuscript presents an analytical approach for the identification and localization of objects using 2D LiDARs. Our main contribution lies in formally defining LASER sensor measurements and their representation, the identification of objects, their main properties, and their location in a scene. We validate our proposal with experiments in generic semi-structured environments common in autonomous navigation, and we demonstrate its feasibility in multiple object detection and identification, strictly following its analytical representation. Finally, our proposal further encourages and facilitates the design, modeling, and implementation of other applications that use LASER scanners as a distance sensor.","PeriodicalId":221960,"journal":{"name":"Sensors (Basel, Switzerland)","volume":"118 ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140792455","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Wenbo Zhang, Ziqian Bai, Pengfei Yan, Hongwei Liu, Li Shao
Functional electrical stimulation (FES) devices are widely employed for clinical treatment, rehabilitation, and sports training. However, existing FES devices are inadequate in terms of wearability and cannot recognize a user’s intention to move or muscle fatigue. These issues impede the user’s ability to incorporate FES devices into their daily life. In response to these issues, this paper introduces a novel wearable FES system based on customized textile electrodes. The system is driven by surface electromyography (sEMG) movement intention. A parallel structured deep learning model based on a wearable FES device is used, which enables the identification of both the type of motion and muscle fatigue status without being affected by electrical stimulation. Five subjects took part in an experiment to test the proposed system, and the results showed that our method achieved a high level of accuracy for lower limb motion recognition and muscle fatigue status detection. The preliminary results presented here prove the effectiveness of the novel wearable FES system in terms of recognizing lower limb motions and muscle fatigue status.
{"title":"Recognition of Human Lower Limb Motion and Muscle Fatigue Status Using a Wearable FES-sEMG System","authors":"Wenbo Zhang, Ziqian Bai, Pengfei Yan, Hongwei Liu, Li Shao","doi":"10.3390/s24072377","DOIUrl":"https://doi.org/10.3390/s24072377","url":null,"abstract":"Functional electrical stimulation (FES) devices are widely employed for clinical treatment, rehabilitation, and sports training. However, existing FES devices are inadequate in terms of wearability and cannot recognize a user’s intention to move or muscle fatigue. These issues impede the user’s ability to incorporate FES devices into their daily life. In response to these issues, this paper introduces a novel wearable FES system based on customized textile electrodes. The system is driven by surface electromyography (sEMG) movement intention. A parallel structured deep learning model based on a wearable FES device is used, which enables the identification of both the type of motion and muscle fatigue status without being affected by electrical stimulation. Five subjects took part in an experiment to test the proposed system, and the results showed that our method achieved a high level of accuracy for lower limb motion recognition and muscle fatigue status detection. The preliminary results presented here prove the effectiveness of the novel wearable FES system in terms of recognizing lower limb motions and muscle fatigue status.","PeriodicalId":221960,"journal":{"name":"Sensors (Basel, Switzerland)","volume":"212 ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140783011","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Philipp Schulmeyer, Manfred Weihnacht, Hagen Schmidt
Ice accumulation on infrastructure poses severe safety risks and economic losses, necessitating effective detection and monitoring solutions. This study introduces a novel approach employing surface acoustic wave (SAW) sensors, known for their small size, wireless operation, energy self-sufficiency, and retrofit capability. Utilizing a SAW dual-mode delay line device on a 64°-rotated Y-cut lithium niobate substrate, we demonstrate a solution for combined ice detection and temperature measurement. In addition to the shear-horizontal polarized leaky SAW, our findings reveal an electrically excitable Rayleigh-type wave in the X+90° direction on the same cut. Experimental results in a temperature chamber confirm capability for reliable differentiation between liquid water and ice loading and simultaneous temperature measurements. This research presents a promising advancement in addressing safety concerns and economic losses associated with ice accretion.
基础设施上的积冰会带来严重的安全风险和经济损失，因此需要有效的检测和监控解决方案。本研究介绍了一种采用表面声波（SAW）传感器的新方法，这种传感器以体积小、无线操作、能源自给自足和改造能力强而著称。通过在 64° 旋转的 Y 型切割铌酸锂基板上使用声表面波双模延迟线装置，我们展示了一种冰探测和温度测量相结合的解决方案。除了剪切水平极化泄漏声表面波外，我们的研究结果还揭示了同一切面上 X+90° 方向上的电激发雷利型波。温度室中的实验结果证实了可靠区分液态水和冰负载以及同时测量温度的能力。这项研究为解决安全问题和与冰增生相关的经济损失问题带来了希望。
{"title":"A Dual-Mode Surface Acoustic Wave Delay Line for the Detection of Ice on 64°-Rotated Y-Cut Lithium Niobate","authors":"Philipp Schulmeyer, Manfred Weihnacht, Hagen Schmidt","doi":"10.3390/s24072292","DOIUrl":"https://doi.org/10.3390/s24072292","url":null,"abstract":"Ice accumulation on infrastructure poses severe safety risks and economic losses, necessitating effective detection and monitoring solutions. This study introduces a novel approach employing surface acoustic wave (SAW) sensors, known for their small size, wireless operation, energy self-sufficiency, and retrofit capability. Utilizing a SAW dual-mode delay line device on a 64°-rotated Y-cut lithium niobate substrate, we demonstrate a solution for combined ice detection and temperature measurement. In addition to the shear-horizontal polarized leaky SAW, our findings reveal an electrically excitable Rayleigh-type wave in the X+90° direction on the same cut. Experimental results in a temperature chamber confirm capability for reliable differentiation between liquid water and ice loading and simultaneous temperature measurements. This research presents a promising advancement in addressing safety concerns and economic losses associated with ice accretion.","PeriodicalId":221960,"journal":{"name":"Sensors (Basel, Switzerland)","volume":"42 5","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140769997","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Zhaobin Duan, Zhenyang Ma, Jie Bai, Peng Wang, Ke Xu, Shun Yuan
In this manuscript, we present a novel deployment protection method aimed at safeguarding aeronautical radio altimeters (RAs) from interference caused by fifth-generation (5G) telecommunication base stations (BSs). Our methodology involves an integrated interference model for defining prohibited zones and utilizes power control and angle shutoff methods to mitigate interference. First, to ensure reliable protection, we define both horizontal and vertical prohibited zones and investigate their variations to immunize RA against 5G interference. Second, we validate the effectiveness of the model in various operational scenarios, analyzing the influence of factors such as base station types, antenna parameters, flight altitude, and aircraft attitudes to cover a wide range of real-world scenarios. Third, to mitigate interference, we propose and analyze the power control and angle shutoff methods through simulation for the RMa prohibited zone. Our results demonstrate the efficacy of the deployment protection method in safeguarding RAs from 5G interference, providing guidance for interference protection during civil aviation operations and base station deployment near airports.
在本手稿中，我们介绍了一种新型部署保护方法，旨在保护航空无线电高度计（RA）免受第五代（5G）电信基站（BS）的干扰。我们的方法涉及一个用于定义禁区的综合干扰模型，并利用功率控制和角度关闭方法来减轻干扰。首先，为了确保可靠的保护，我们定义了水平和垂直禁区，并研究了它们的变化，以增强 RA 对 5G 干扰的免疫力。其次，我们验证了该模型在各种运行场景中的有效性，分析了基站类型、天线参数、飞行高度和飞机姿态等因素的影响，以涵盖广泛的现实世界场景。第三，为减少干扰，我们提出并分析了针对 RMa 禁区的功率控制和角度关闭方法。我们的结果证明了部署保护方法在保护 RA 免受 5G 干扰方面的有效性，为民航运行和机场附近基站部署期间的干扰防护提供了指导。
{"title":"Deployment Protection for Interference of 5G Base Stations with Aeronautical Radio Altimeters","authors":"Zhaobin Duan, Zhenyang Ma, Jie Bai, Peng Wang, Ke Xu, Shun Yuan","doi":"10.3390/s24072313","DOIUrl":"https://doi.org/10.3390/s24072313","url":null,"abstract":"In this manuscript, we present a novel deployment protection method aimed at safeguarding aeronautical radio altimeters (RAs) from interference caused by fifth-generation (5G) telecommunication base stations (BSs). Our methodology involves an integrated interference model for defining prohibited zones and utilizes power control and angle shutoff methods to mitigate interference. First, to ensure reliable protection, we define both horizontal and vertical prohibited zones and investigate their variations to immunize RA against 5G interference. Second, we validate the effectiveness of the model in various operational scenarios, analyzing the influence of factors such as base station types, antenna parameters, flight altitude, and aircraft attitudes to cover a wide range of real-world scenarios. Third, to mitigate interference, we propose and analyze the power control and angle shutoff methods through simulation for the RMa prohibited zone. Our results demonstrate the efficacy of the deployment protection method in safeguarding RAs from 5G interference, providing guidance for interference protection during civil aviation operations and base station deployment near airports.","PeriodicalId":221960,"journal":{"name":"Sensors (Basel, Switzerland)","volume":"59 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140786604","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The advancement of unmanned aerial vehicles (UAVs) enables early detection of numerous disasters. Efforts have been made to automate the monitoring of data from UAVs, with machine learning methods recently attracting significant interest. These solutions often face challenges with high computational costs and energy usage. Conventionally, data from UAVs are processed using cloud computing, where they are sent to the cloud for analysis. However, this method might not meet the real-time needs of disaster relief scenarios. In contrast, edge computing provides real-time processing at the site but still struggles with computational and energy efficiency issues. To overcome these obstacles and enhance resource utilization, this paper presents a convolutional neural network (CNN) model with an early exit mechanism designed for fire detection in UAVs. This model is implemented using TSMC 40 nm CMOS technology, which aids in hardware acceleration. Notably, the neural network has a modest parameter count of 11.2 k. In the hardware computation part, the CNN circuit completes fire detection in approximately 230,000 cycles. Power-gating techniques are also used to turn off inactive memory, contributing to reduced power consumption. The experimental results show that this neural network reaches a maximum accuracy of 81.49% in the hardware implementation stage. After automatic layout and routing, the CNN hardware accelerator can operate at 300 MHz, consuming 117 mW of power.
{"title":"Implementation of Lightweight Convolutional Neural Networks with an Early Exit Mechanism Utilizing 40 nm CMOS Process for Fire Detection in Unmanned Aerial Vehicles","authors":"Yu-Pei Liang, Chen-Ming Chang, Ching-Che Chung","doi":"10.3390/s24072265","DOIUrl":"https://doi.org/10.3390/s24072265","url":null,"abstract":"The advancement of unmanned aerial vehicles (UAVs) enables early detection of numerous disasters. Efforts have been made to automate the monitoring of data from UAVs, with machine learning methods recently attracting significant interest. These solutions often face challenges with high computational costs and energy usage. Conventionally, data from UAVs are processed using cloud computing, where they are sent to the cloud for analysis. However, this method might not meet the real-time needs of disaster relief scenarios. In contrast, edge computing provides real-time processing at the site but still struggles with computational and energy efficiency issues. To overcome these obstacles and enhance resource utilization, this paper presents a convolutional neural network (CNN) model with an early exit mechanism designed for fire detection in UAVs. This model is implemented using TSMC 40 nm CMOS technology, which aids in hardware acceleration. Notably, the neural network has a modest parameter count of 11.2 k. In the hardware computation part, the CNN circuit completes fire detection in approximately 230,000 cycles. Power-gating techniques are also used to turn off inactive memory, contributing to reduced power consumption. The experimental results show that this neural network reaches a maximum accuracy of 81.49% in the hardware implementation stage. After automatic layout and routing, the CNN hardware accelerator can operate at 300 MHz, consuming 117 mW of power.","PeriodicalId":221960,"journal":{"name":"Sensors (Basel, Switzerland)","volume":"53 13","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140788415","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
F. L. Tofoli, Thaís Martins Jajah Carlos, A. Morais
The cascaded connection of power converters extends conversion ranges but requires careful consideration due to high component count and efficiency concerns, as power is processed redundantly. Furthermore, using several active switches that must be turned on simultaneously may introduce significant drive and control complexity. To overcome this limitation, single-switch quadratic DC-DC converters have been proposed in the literature as a prominent choice for various applications, such as light-emitting diode (LED) drivers. Nevertheless, the motivation behind the conception of such topologies, beyond extending the conversion ratio, remains unclear. Another unexplored issue is the possibility of obtaining single-switch versions of cascaded converters consisting of multiple stages. In this context, this work investigates the synthesis of single-switch non-isolated DC-DC converters for achieving high step-down and/or high step-up based on the graft scheme. Key issues such as the voltage gain, additional stresses on the active switches, component count, and behavior of the input current and output stage current are addressed in detail. An in-depth discussion is presented to identify potential advantages and shortcomings of the resulting structures.
{"title":"Review, Properties, and Synthesis of Single-Switch Non-Isolated DC-DC Converters with a Wide Conversion Range","authors":"F. L. Tofoli, Thaís Martins Jajah Carlos, A. Morais","doi":"10.3390/s24072264","DOIUrl":"https://doi.org/10.3390/s24072264","url":null,"abstract":"The cascaded connection of power converters extends conversion ranges but requires careful consideration due to high component count and efficiency concerns, as power is processed redundantly. Furthermore, using several active switches that must be turned on simultaneously may introduce significant drive and control complexity. To overcome this limitation, single-switch quadratic DC-DC converters have been proposed in the literature as a prominent choice for various applications, such as light-emitting diode (LED) drivers. Nevertheless, the motivation behind the conception of such topologies, beyond extending the conversion ratio, remains unclear. Another unexplored issue is the possibility of obtaining single-switch versions of cascaded converters consisting of multiple stages. In this context, this work investigates the synthesis of single-switch non-isolated DC-DC converters for achieving high step-down and/or high step-up based on the graft scheme. Key issues such as the voltage gain, additional stresses on the active switches, component count, and behavior of the input current and output stage current are addressed in detail. An in-depth discussion is presented to identify potential advantages and shortcomings of the resulting structures.","PeriodicalId":221960,"journal":{"name":"Sensors (Basel, Switzerland)","volume":"259 ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140790205","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：