检索结果-内蒙古大学图书馆

Complex environment localization system using complementary ceiling and ground map information

Autonomous Robots 2023年第6期47卷 669-683页

作者： Yu, Chee-An Chen, Hao-Yun Wang, Chun-Chieh Fu, Li-Chen Department of Electrical Engineering National Taiwan University Taipei Taiwan NTU Center for Artificial Intelligence and Advanced Robotics Taipei Taiwan

This paper proposes a robust localization system using complementary information extracted from ceiling and ground plans, particularly applicable to dynamic and complex environments. The ceiling perception provides the robot with stable and time-invariant environmental features independent of the dynamic changes on the ground, whereas the ground perception allows the robot to navigate in the ground plane while avoiding stationary obstacles. We propose an architecture to fuse ground 2D LiDAR scan and ceiling 3D LiDAR scan with our enhanced mapping algorithm associating perception from both sources efficiently. The localization ability and the navigation performance can be promisingly secured even in a harsh environment with our complementary sensed information from the ground and ceiling. The salient feature of our work is that our system can simultaneously map both the ceiling and ground plane efficiently without extra efforts of deploying articulated landmarks and apply such hybrid information effectively, which facilitates the robot to travel through any indoor environment with human crowds without getting lost. © 2023, The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature.

关键词： Ceilings

来源：评论

学校读者我要写书评

暂无评论

Spatial Graph-Based Localization and Navigation on Scaleless Floorplan

引用

IEEE robotics and Automation Letters 2024年第4期9卷 3932-3939页

作者： Ewe, Zu Lin Chang, Fu-Hao Huang, Yi-Shiang Fu, Li-Chen National Taiwan University Department of Electrical Engineering Taipei100 Taiwan National Taiwan University NTU Center for Artificial Intelligence and Advanced Robotics Taipei100 Taiwan

Effective navigation in unfamiliar environments remains a critical challenge for successful deployment. Current navigation methods, which rely on autonomous or teleoperated exploration and map building, pose technical difficulties for inexperienced end-users. In contrast, humans can effectively navigate using abstract floorplans, suggesting the potential for service robots to leverage similar techniques. The practical application of floorplan-based navigation;however, is currently limited by methods that require pre-exploration or floorplan with accurate measurements or scale. This letter aims to address the aforementioned challenges and investigate the feasibility of floorplan-based navigation in unfamiliar environments. Specifically, we propose a novel scale-invariant floorplan localization method, enabling navigation without relying on precise scale information. Furthermore, we introduce an incremental graph augmentation approach that enriches the floorplan representation with traversability information derived from robot observations. Finally, we develop an efficient navigation framework capable of utilizing both the inherent structure of the floorplan and real-time observations. Experimental results demonstrate that our scale-invariant floorplan localization method outperforms baseline methods in most cases when floorplan scale information is unavailable, and our graph-based navigation system exhibits superior success and efficiency as compared to grid-based counterparts. The outcomes of this research contribute to the advancement of service robot deployment in unfamiliar environments, particularly in scenarios where extensive exploration and map building may be impractical or technically challenging for end-users. © 2016 IEEE.

关键词： Navigation systems

来源：评论

学校读者我要写书评

暂无评论

Data Augmentation via Face Morphing for Recognizing Intensities of Facial Emotions

引用

IEEE Transactions on Affective Computing 2023年第2期14卷 1228-1235页

作者： Huang, Tsung-Ren Hsu, Shin-Min Fu, Li-Chen National Taiwan University Department of Psychology Center for Artificial Intelligence & Advanced Robotics Taipei106319 Taiwan National Taiwan University Center for Artificial Intelligence & Advanced Robotics Department of Electrical Engineering Department of Computer Science and Information Engineering Taipei106319 Taiwan

Being able to recognize emotional intensity is a desirable feature for a facial emotional recognition (FER) system. However, the development of such a feature is hindered by the paucity of intensity-labeled data for model training. To ameliorate the situation, the present study proposes using face morphing as a novel way of data augmentation to synthesize faces that express different degrees of a designated emotion. Such an approach has been successfully validated on humans and machines. Specifically, humans indeed perceived different levels of intensified emotions in these parametrically synthesized faces, and FER systems based on neural networks indeed showed improved sensitivities to intensities of different emotions when additionally trained on the synthesized faces. Overall, the proposed data augmentation method is not only simple and effective but also useful for building FER systems that recognize facial expressions of mixed emotions. © 2010-2012 IEEE.

关键词： Face recognition

来源：评论

学校读者我要写书评

暂无评论

Enhancing Healthcare: IoT-LoRaWAN System for Stress Level Monitoring

Enhancing Healthcare: IoT-LoRaWAN System for Stress Level Mo...

引用

2024 International Conference on Information Technology and Computing, ICITCOM 2024

作者： Hayati, Nur Riki Aldi, Iqbal Widyasmoro Center of Artificial Intelligence Robotics Studies Universitas Muhammadiyah Yogyakarta Department of Electrical Engineering Bantul Indonesia Universitas Muhammadiyah Yogyakarta Department of Electrical Engineering Bantul Indonesia

ISBN: (纸本)9798350379839

Indonesia is grappling with a significant mental health issue as the number of suicides continues to rise each year. This phenomenon aligns with the global trend highlighted by the World Health Organization (WHO), which reports approximately 700,000 suicides occurring worldwide annually. Impaired mental health frequently plays a significant role in cases of suicide. Thus, prioritizing mental well-being is crucial in preventing suicide. Experiencing high levels of stress can significantly worsen one's mental well-being. In order to address the issue at hand, we require a creative approach to monitoring stress levels. This research project focuses on designing and building a stress level monitoring system using IoT-LoRa Wantechnology. The system is capable of monitoring patient stress levels in real-time. Based on the test results, it is evident that the system has achieved an impressive average accuracy of 97% in monitoring stress levels over 20 data collection times. We conducted a thorough analysis to evaluate the precision of the data obtained from the sensor by comparing it with a pulse oximeter. © 2024 IEEE.

关键词： Electronic health record

来源：评论

学校读者我要写书评

暂无评论

Data augmentation and denoising of computed tomography scan images in training deep learning models for rapid COVID-19 detection

引用

International Journal of Business intelligence and Data Mining 2024年第2期24卷 203-216页

作者： Mubarak, Auwalu Saleh Serte, Sertan Al-Turjman, Fadi Ameen, Zubaida Sa'Id Artificial Intelligence Engineering Department AI and Robotics Institute Near East University Mersin10 Turkey Research Center for AI and IoT Faculty of Engineering University of Kyrenia Mersin10 Turkey Department of Electrical and Electronics Engineering Near East University Mersin10 Turkey

The deadly respiratory disease corona virus-2 (COVID-19) which was declared a pandemic by the World Health Organization (WHO) has resulted in over a million deaths around the world within less than a year. With the rapid spread of the virus, the currently adopted COVID-19 test by the WHO is the reverse transcription polymerase chain reaction (RT-PCR) test, which is expensive, time-consuming and not accessed by underdeveloped countries. Computed tomography (CT) scan images that were used in profiling suspected COVID-19 patients can serve as an alternative to the RT PCR test method. In this study, two different pre-Trained deep learning models ResNet-50 and ResNet-101 were trained to classify positive COVID-19 scan images. The best model which was trained on the augmented CT scan images achieved an accuracy of 98.3%, a sensitivity of 0.984, specificity of 0.983. © 2024 Inderscience Enterprises Ltd.. All rights reserved.

关键词： COVID-19

来源：评论

学校读者我要写书评

暂无评论

Predicting PM2.5 Levels using AIoT Device in Smart Campus Environment

Predicting PM2.5 Levels using AIoT Device in Smart Campus En...

引用

2023 International Conference on artificial intelligence robotics, Signal and Image Processing, AIRoSIP 2023

作者： Putra, Karisma Trinanda Jamal, Agus Mujahid, Faaris Ayuningtias, Adela Nafis, Raihan Fauzan Hafizah, Annisa Universitas Muhammadiyah Yogyakarta Center of Artificial Intelligence and Robotics Studies Department of Electrical Engineering Bantul Indonesia Universitas Muhammadiyah Yogyakarta Department of Electrical Engineering Bantul Indonesia

ISBN: (纸本)9798350327472

Integrating an environmental prediction system within a smart campus framework is crucial as it empowers campus management to proactively address the detrimental impacts of pollution on the campus environment and the health and well-being of its occupants. Existing prediction systems that rely on centralized server and micro-sensory nodes often struggle with accuracy when forecasting extensive environmental measurements. Moreover, this edge sensory nodes are not capable of performing prediction models e.g., artificial intelligence (AI) models, thus real-time predictions cannot be achieved. To address this, our research introduces a novel artificial intelligence of things (AIoT) model called long short-term memory (LSTM) lite, which leverages a regular LSTM architecture to effectively capture intricate temporal patterns using a comprehensive PM2.5 dataset spanning one month. We compare our model against naive AI approaches, including linear regression (LR) and nearest neighbor regression (NNR), to evaluate its performance. Through supervised learning, we carefully process sequential data to forecast future events in the data series. Additionally, our model incorporates a flexible architecture that allows for the extraction of multivariate features, enabling the identification of crucial correlations across multiple dimensions. Experimental results demonstrate the superior performance of our proposed model, with prediction accuracies approximately 25.4% higher than LR and 3.7% higher than NNR, as measured by the R-squared metric. Overall, our findings affirm the feasibility and effectiveness of utilizing an AI lite model for handling data in time series prediction tasks with dramatically lower size from its regular model. © 2023 IEEE.

关键词： Prediction models

来源：评论

学校读者我要写书评

暂无评论

Continuous Glucose Monitoring: A Non-Invasive Approach for Improved Daily Healthcare 3

Continuous Glucose Monitoring: A Non-Invasive Approach for I...

引用

3rd International Conference on Electronic and electrical engineering and Intelligent System, ICE3IS 2023

作者： Putra, Karisma Trinanda Surahmat, Indar Nazilah Chamim, Anna Nur Ramadhan, Muhammad Zhafran Wicaksana, Darma Dhea Namyra Alissa, Raden Ajeng Center of Artificial Intelligence and Robotics Studies Universitas Muhammadiyah Yogyakarta Department of Electrical Engineering Bantul Indonesia Universitas Muhammadiyah Yogyakarta Department of Electrical Engineering Bantul Indonesia

ISBN: (纸本)9798350327762

Diabetes mellitus is a prevalent health condition affecting approximately one in eight individuals worldwide. Insufficient management of this disease can lead to various medical complications, including vision loss and limb amputation. Continuous glucose monitoring (CGM) is crucial for effective diabetes management. However, current one-time glucose monitoring methods still rely on invasive techniques, causing pain and lacking continuous monitoring capability. This study aims to develop a non-invasive technology for measuring blood glucose levels that can be conveniently monitored through a smartphone using wireless communication. The device utilizes an ESP32 microprocessor and a PPG MAX 30102 sensor capable of recording pulse waves, a biomarker, at the fingertips. In this preliminary study, non-invasive data from 22 participants were used and divided into three classes according to conventional invasive measurements. artificial intelligence (AI) methods, specifically Convolutional Neural Networks (CNN) and artificial Neural Networks (ANN) models, were utilized to achieve convergence between the conventional invasive measurements and the proposed approach. The trained model demonstrated the highest accuracy of 0.944, with a minimum model loss of 0.177. In the validation phase, the proposed approach proved its reliability by achieving an average accuracy of 0.750. The non-invasive approach has the potential to provide a convenient and pain-free alternative to traditional monitoring methods, significantly enhancing diabetes management and improving the overall quality of life for those affected by this condition. © 2023 IEEE.

关键词： Glucose

来源：评论

学校读者我要写书评

暂无评论

Enhancing Indoor User Localization:An Adaptive Bayesian Approach for Multi-Floor Environments

引用

Computers, Materials & Continua 2024年第8期80卷 1889-1905页

作者： Abdulraqeb Alhammadi Zaid Ahmed Shamsan Arijit De Center for Artificial Intelligence and Robotics(CAIRO) Malaysia-Japan International Institute of TechnologyUniversiti Teknologi MalaysiaKuala Lumpur54100Malaysia Department of Electrical Engineering College of EngineeringImam Mohammad Ibn Saud Islamic University(IMSIU)Riyadh11432Saudi Arabia Department of Information Engineering Electronics and TelecommunicationsSapienza UniversityRome00185Italy

Indoor localization systems are crucial in addressing the limitations of traditional global positioning system(GPS)in indoor environments due to signal attenuation *** complex indoor spaces become more sophisticated,indoor localization systems become essential for improving user experience,safety,and operational *** localization methods based on Wi-Fi fingerprints require a high-density location fingerprint database,but this can increase the computational burden in the online *** networks,which integrate prior knowledge or domain expertise,are an effective solution for accurately determining indoor user *** networks use probabilistic reasoning to model relationships among various localization parameters for indoor environments that are challenging to *** article proposes an adaptive Bayesian model for multi-floor environments based on fingerprinting techniques to minimize errors in estimating user *** proposed system is an off-the-shelf solution that uses existing Wi-Fi infrastructures to estimate user’s *** operates in both online and offline *** the offline phase,a mobile device with Wi-Fi capability collects radio signals,while in the online phase,generating samples using Gibbs sampling based on the proposed Bayesian model and radio map to predict user’s *** results unequivocally showcase the superior performance of the proposed model when compared to other existing models and *** proposed model achieved an impressive lower average localization error,surpassing the accuracy of competing ***,this noteworthy achievement was attained with minimal reliance on reference points,underscoring the efficiency and efficacy of the proposed model in accurately estimating user locations in indoor environments.

关键词： Localization positioning Bayesian fingerprinting received signal strength(RSS)

来源：评论

学校读者我要写书评

暂无评论

Histogram of Oriented Gradients Feature Extraction in Enhanced Convolutional Neural Networks Method for Palm Vein Recognition 4

Histogram of Oriented Gradients Feature Extraction in Enhanc...

引用

4th International Conference on Electronic and electrical engineering and Intelligent System, ICE3IS 2024

作者： Wulandari, Meirista Irmawati Hayati, Nur Universitas Tarumanagara Department of Electrical Engineering Jakarta Indonesia Universitas Multimedia Nusantara Department of Information System Tangerang Indonesia Universitas Muhammadiyah Yogyakarta Center of Artificial Intelligence and Robotics Studies Department of Electrical Engineering Bantul Indonesia

ISBN: (纸本)9798350378368

Histogram of Oriented Gradients, or HOG, is an efficient method for extracting features from images that shows to be reliable in recognizing and detecting objects in a variety of conditions. HOG is a very useful tool in image analysis because it is invariant to brightness, rotation, and scale transformations and can generate detailed local descriptors. By exploiting the advantages of HOG in feature extraction, the palm vein recognition system can overcome the weaknesses of palm vein images, which are sensitive to changes in brightness and rotation. By analyzing the distribution of orientation gradients, HOG can extract important features from palm vein images generated from infrared light, enabling reliable biometric recognition. We proposed a method applying HOG into palm vein biometric system with Convolutional Neural Network (CNN) which has strong hierarchical representation capabilities, automatic feature extraction, and the ability to handle high-dimensional data. Our proposed model is tested on three public contactless palm vein datasets (CASIA and VERA). We evaluated the results using various performance biometric measures, such as correct recognition rate (CRR) accuracy, and area under curve (AUC) of the receiver operating characteristic. The simulation results showed that HOG features could improve accuracy from pretrained CNN such as AlexNet, VGG-16, ResNet-50 and EfficientNet. © 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

Classification of Cancer Cells from Mammogram Images Based on GoogLeNet and ResNet-50 Deep Learning Models 1

Classification of Cancer Cells from Mammogram Images Based o...

引用

1st International Conference on Smart Mechatronics, ICSMech 2024

作者： Jusman, Yessi Nur 'Aini, Masayu Alya Naufal, Zaidan Ahmad Ridho, Prajarino Rushartini, Hesti Wimala, Galuh Ananda, Indra Pradani, Nisha Alifa Universitas Muhammadiyah Yogyakarta Center of Artificial Intelligence and Robotics Studies Department of Electrical Engineering Bantul Indonesia Universiti Malaysia Perlis Faculty of Electrical Engineering and Technology Perlis Malaysia Universitas Muhammadiyah Yogyakarta Faculty of Engineering Department of Electrical Engineering Bantul Indonesia

ISBN: (纸本)9798350368918

Breast cancer remains one of the most common and deadly forms of cancer affecting women globally. Early detection through mammogram screening plays a vital role in improving survival rates by enabling timely diagnosis and treatment. This study explored the application of deep learning techniques, particularly GoogLeNet and ResNet-50 architectures, to classify mammogram images into benign and malignant categories. The dataset, sourced from Kaggle, consisted of 479 mammogram images. These images were preprocessed and divided into training and testing sets, with 90% allocated for training and 10% for testing. Both GoogLeNet and ResNet-50 models were trained and tested, yielding different performance metrics. GoogLeNet achieved the highest testing accuracy of 70.83%, while ResNet-50 attained 68.75% accuracy. GoogLeNet demonstrated better sensitivity at 92.85%, but ResNet-50 outperformed in terms of precision at 69.69%. Although both models displayed competitive results, GoogLeNet exhibited superior performance in classifying cancer cells based on sensitivity, making it more effective in identifying true positives. These findings demonstrated the potential of deep learning models to enhance breast cancer detection and classification using mammogram images, contributing to improved diagnostic accuracy and patient outcomes. © 2024 IEEE.

关键词： Lung cancer

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：