检索结果-内蒙古大学图书馆

2023 International Conference on Cyber Management and engineering, CyMaEn 2023

作者： Salau, Ayodeji Olalekan Abeje, Bekalu Tadele Faisal, Abeer Naser Asfaw, Tesfa Tegegne Afe Babalola University Department of Electrical/Electronics and Computer Engineering Ado-Ekiti Nigeria Haramaya University College of Computing and Informatics Department of Information Technology Ethiopia Institute of Technology Faculty of Computing Bahir Dar University Department of Computer Science Ethiopia University of Sumer College of Computer Science and Information Technology Al-Rifai Iraq

ISBN: (纸本)9781665493291

Faba bean (Vicia faba L.) is an important cash crop for animal and human consumption in many countries, especially Ethiopia due to its high protein content and high rate of production. It also improves soil fertility with the help of bacteria, which forms nodules on its roots. Ethiopia is the second largest faba bean producer;however, according to research, the yield is generally low. Among the constraints contributing to low productivity of the crop, disease is the most important. Even though there are countermeasures such as the use of chemical sprays, which is used to reduce the damage or eliminate the disease, an effective disease identification technique is still required. Chocolate spot, faba bean gal, and rust are the most common diseases of faba beans. To effectively detect the disease, in this paper, we propose an end-to-end CNN model for training before and after the dataset is preprocessed. The CNN end-to-end model is trained with raw images and achieves 92.1% training accuracy. Also, the model is trained with preprocessed data and it achieves a training accuracy of 98.14% for the end-to-end CNN classifier. The results show that the proposed approach is suitable for faba bean disease detection and can be employed to detect diseases of other crops. © 2023 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

RELATIONAL SURROGATE LOSS LEARNING 10

RELATIONAL SURROGATE LOSS LEARNING

引用

10th International Conference on Learning Representations, ICLR 2022

作者： Huang, Tao Li, Zekang Lu, Hua Shan, Yong Yang, Shusheng Feng, Yang Wang, Fei You, Shan Xu, Chang School of Computer Science Faculty of Engineering The University of Sydney Australia SenseTime Research China Huazhong University of Science and Technology China University of Science and Technology of China China

Evaluation metrics in machine learning are often hardly taken as loss functions, as they could be non-differentiable and non-decomposable, e.g., average precision and F1 score. This paper aims to address this problem by revisiting the surrogate loss learning, where a deep neural network is employed to approximate the evaluation metrics. Instead of pursuing an exact recovery of the evaluation metric through a deep neural network, we are reminded of the purpose of the existence of these evaluation metrics, which is to distinguish whether one model is better or worse than another. In this paper, we show that directly maintaining the relation of models between surrogate losses and metrics suffices, and propose a rank correlation-based optimization method to maximize this relation and learn surrogate losses. Compared to previous works, our method is much easier to optimize and enjoys significant efficiency and performance gains. Extensive experiments show that our method achieves improvements on various tasks including image classification and neural machine translation, and even outperforms state-of-the-art methods on human pose estimation and machine reading comprehension tasks. Code is available at: https://***/hunto/ReLoss. © 2022 ICLR 2022 - 10th International Conference on Learning Representationss. All rights reserved.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Spatial-temporal Consistency Constraint for Depth and Ego-motion Estimation of Laparoscopic Images

Spatial-temporal Consistency Constraint for Depth and Ego-mo...

引用

2024 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2024

作者： Zhang, Qin Nan, Xiangling Fan, Yingfang Zhou, Yihang Liu, Yaoqun Jia, Fucang Luo, Huoling South China University of Technology School of Computer Science and Engineering Guangzhou China Third Affiliated Hospital Southern Medical University Guangzhou China Chinese University of Hong Kong Shenzhen School of Science and Engineering Shenzhen China Chinese Academy of Sciences Shenzhen Institute of Advanced Technology Shenzhen China Shenzhen Institute of Information Technology School of Digital Media Shenzhen China

ISBN: (纸本)9798350386226

Estimating depth and ego-motion are crucial tasks for laparoscopic navigation and robotic-assisted surgery. Most current self-supervised methods involve warping one frame onto an adjacent frame using the estimated depth and camera pose. The photometric loss between the estimated and original frames then serves as the training signal. However, these methods encounter major challenges due to non-Lambertian reflection regions and the textureless surfaces of organs, leading to significant performance degradation and scale ambiguity in monocular depth estimation. In this paper, we introduce a network that predicts depth and ego-motion using spatial-temporal consistency constraints. Spatial consistency is derived from the left and right views of the stereo laparoscopic image pairs, while temporal consistency comes from consecutive frames. To enhance the understanding of semantic information in surgical scenes, we employ the Swin Transformer as the encoder and decoder for depth estimation, due to its superior semantic segmentation capabilities. To address issues of illumination variance and scale ambiguity, we incorporate a SIFT loss term to eliminate oversaturated regions in laparoscopic images. Our method is evaluated on the SCARED dataset and shows remarkable results. The code is publicly available at https://***/nanasylum/spatialtemporal. © 2024 IEEE.

关键词： Laparoscopy

来源：评论

学校读者我要写书评

暂无评论

Early Detecting the Abnormality of Heart Function via *** System (Made by Viet Nam) 9th

Early Detecting the Abnormality of Heart Function via *** Sy...

引用

9th International Conference on the Development of Biomedical engineering in Vietnam, BME 2022

作者： Thang, Bui Quoc Duy, Phan Dinh The Thien, Vu Trong Department of Scientific Research Training Center Cho Ray Hospital Ho Chi Minh City Viet Nam Department of General Surgery Medical Pharmacy Faculty Tra Vinh University TraVinh Viet Nam Faculty of Computer Science and Engineering HCMC University of Technology Ho Chi Minh City Viet Nam

ISBN: (纸本)9783031446290

Most recent non-invasive blood pressure measurements only show systolic and diastolic pressure;some devices can alert the abnormal heart rhythm. This report aimed to show how the cuff pressure waveforms of a non-invasive blood pressure measurement device combined with an IoT system detects the abnormal heart function related to some common cardiovascular disease. We developed a blood pressure machine based on an IoT system for collecting patient data. Heart rhythm and function abnormality can be detected via the cuff pressure waveforms. The cardiac output can also be calculated following the result of treatment via heart function changes. Cuff pressure waveforms of blood pressure machines based on medical knowledge, platform IoT and data science can be used to detect abnormality of heart function and follow up the treatment result. The device was given a personal notification based on the users’ medical data. The information was also collected, saved on server, and then sent to the user’s family and their family doctor via the web interface. We recently designed, produced and implemented 30 prototypes in Vietnam from June 2022 to test their stability, accuracy, and usefulness. © The Author(s), under exclusive license to Springer Nature Switzerland AG. 2024.

关键词： Blood pressure

来源：评论

学校读者我要写书评

暂无评论

Stereo Matching for Face Images Based on Feature Points

引用

Chinese Journal of Electronics 2023年第2期19卷 241-244页

作者： Yinghui Wang Bing Xiao Qiongfang Wang Faculty of Computer Science and Engineering Xi'an University of Technology Xi'an China Department of Computer Science Shaanxi Normal University Xi'an China Faculty of of Automation and Information Engineering Xi'an University of Technology Xi'an China

Stereo matching is the key of reconstructing 3D face model from 2D face images and Barnard's algorithm is one of the classical stereo matching algorithms based on feature points. In order to enhance the stability of extracting feature points, reduce computational complexity and minimize the effect of subjective factors, in this paper, Barnard's algorithm of edge point matching was improved from two aspects: edge point sets of left and right face images were extracted with dyadic wavelet transform, rather than comparing gray values of adjacent pixels; on the other hand, edge points in binocular stereo face images were matched under epipolar constraint in parallel binocular stereo matching. Experimental results demonstrate that stability and efficiency of stereo matching algorithm are both improved.

关键词： Wavelet transforms Technological innovation Solid modeling Three-dimensional displays Image edge detection Education medical services

来源：评论

学校读者我要写书评

暂无评论

Analysis and Treatment of Missing Data in Happiness Score Using Machine Learning 21

Analysis and Treatment of Missing Data in Happiness Score Us...

引用

21st IEEE International Conference on Smart Communities: Improving Quality of Life using AI, Robotics and IoT, HONET 2024

作者： Touati, Wassim El Touati, Yamen Hadj-Alouane, Nejib Ben Jemai, Abderrazek Polytechnic School of Tunisia Carthage University SERCOM Laboratory Tunis Tunisia Northern Border University Faculty of Computing and Information Technology Department of Computer Science Saudi Arabia ENIT University of Tunis EI Manar OASIS Laboratory Tunisia American University in Dubai Electrical Engineering and Computer Engineering Department Dubai United Arab Emirates

ISBN: (纸本)9798350378078

Machine learning, a vital part of artificial intelli-gence, improves our ability to make predictions from complex data. The success of these predictions relies heavily on the model's fit with its data and the data's consistency. Accurate predictions depend on closely matching the model to the data patterns. However, missing data can seriously disrupt the accuracy of predictions, causing errors. To address this, we recommend adopting a proposed framework that enhances predictions by effectively filling in missing data. Our objective is to refine the accuracy of our predictions, especially in scenarios where the data might be incomplete and features exhibit patterns of dependency. In the context of missing data, our approach concentrates on enhancing regression performance to minimize the effects of missing data as effectively as possible. © 2024 IEEE.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Sine-Squared Pulse Approximation for Matched Filter Design Using Generalized Bessel Polynomials and Particle Swarm Optimization 11

Sine-Squared Pulse Approximation for Matched Filter Design U...

引用

2022 11th International Conference on Information and Electronics engineering, ICIEE 2022

作者： Chutchavong, V. Anuwongpinit, T. Pumee, T. Benjangkaprasert, C. Janchitrapongvej, K. Department of Computer Engineering School of Engineering King Mongkut's Institute of Technology Ladkrabang Bangkok Thailand Faculty of Science and Technology South East Bangkok College Bangkok Thailand

This paper presents the study of mathematical characteristics of generalized Bessel polynomial that can be applied to approximate a sine-squared pulse for designing matched filters in communication systems. The proposed pulse can be designed by using the transfer function, in which the numerator is the five pairs of a transmissions zero pairs, and the generalized Bessel polynomial is used as the denominator. A parameters of generalized Bessel polynomials can be adjusted by particle swarm optimization to find the best parameter value. From the simulation results can be found that a parameter can be adjusted. A proposed pulse is close to the ideal response in mainlobe, and one sidelobe to four sidelobes with stability, which outperformed previous research. © Published under licence by IOP Publishing Ltd.

关键词： Matched filters

来源：评论

学校读者我要写书评

暂无评论

Simulation and Evaluation of a Cascade Multilevel Inverter Using STATCOM Strategy

Simulation and Evaluation of a Cascade Multilevel Inverter U...

引用

2024 International Conference on Sustainable Computing and Integrated Communication in Changing Landscape of AI, ICSCAI 2024

作者： Kumar, Neeraj Jain, Upma Almusawi, Muntather Kumar, Yogendra Deepak, A. Dutt, Amit Department of Electrical Engineering Doon Institute of Engineering and Technology Uttarakhand Rishikesh India Department of Computer Science & Engineering Graphic Era Deemed to be University Dehradun India Department of Computers Techniques Engineering College of Technical Engineering The Islamic University Najaf Iraq Department of Electrical Engineering GLA University Mathura India Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Saveetha University Tamilnadu Chennai India Lovely Professional University Phagwara India

ISBN: (纸本)9798350355338

In this paper analysis the Cascade multilevel inverters (CMI) are superior to other multilevel inverter configurations in STATCOM applications owing to their distinguishing features, such as structural flexibility and control design feasibility in the service of lower device losses and high value of AC output voltage waveform improve. Voltage output stabilization and pf correction in case of common connection are two exceedingly difficulties that might lead to the system's dominance. This study presents the system's general control architecture and details how the 9-level CMI might be used as STATCOM to perform reactive power compensation and DC network voltage equalization. The superiority of the system under varying loads and fault circumstances has been confirmed by computer modeling. © 2024 IEEE.

关键词： Cascade multilevel inverter (CMI) Line ground (LG) fault Point of common coupling (PCC)

来源：评论

学校读者我要写书评

暂无评论

Optimization Effects of Rotating Transmitarray for Direction-of-Arrival Estimation with a Single Antenna

引用

IEICE Communications Express 2025年第6期14卷 210-213页

作者： Ryusuke Sunagawa Hibiki Shiiba Makoto Sano Ryuji Kuse Takeshi Fukusako Dept. of Computer Science and Electrical Engineering Kumamoto University Kumamoto Japan Faculty of Advanced Science and Technology Kumamoto University Kumamoto Japan

This paper discusses the optimization effects of a transmitarray on a direction-of-arrival (DOA) estimation using the transmitarray and a single receiving antenna. The transmitarray consists of 1-bit elements with 0° and 180° transmission phases. The transmission phase distribution was optimized using a genetic algorithm and compared to a random distribution. The optimized transmitarray achieves a DOA estimation accuracy comparable to that of a conventional system using a 3-element linear array antenna. The root mean square error (RMSE) of the optimized system exhibits a decreasing trend as the signal-to-noise ratio (SNR) increases. Notably, at an SNR of 20 dB, the RMSE is 20° lower than that of the 3-element linear array antenna.

关键词： Estimation Direction-of-arrival estimation Linear antenna arrays Accuracy Antenna radiation patterns Transmitting antennas Optimization Signal to noise ratio Receiving antennas Vectors

来源：评论

学校读者我要写书评

暂无评论

FGRL-Net: Fine-Grained Personalized Patient Representation Learning for Clinical Risk Prediction Based on EHRs

FGRL-Net: Fine-Grained Personalized Patient Representation L...

引用

2023 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2023

作者： Chio, KaKit Zhu, Wenhao He, Lihua Zhang, Dian Yang, Xu Luo, Wuman Macao Polytechnic University Faculty of Applied Sciences S.A.R Macao China School of Information and Communication Engineering University of Electronic Science and Technology of China Sichuan Province China Shenzhen University Department of Computer Science and Software Engineering Shenzhen China

ISBN: (纸本)9798350337020

Personalized patient representation learning (PPRL) is a critical element in clinical risk prediction. It aims to obtain a complete portrait of each patient based on Electronic Health Records (EHR). Although existing works have achieved remarkable progress in healthcare prediction, there are still three major issues. First, feature correlation is crucial for risk prediction, but it has not yet been fully exploited by existing works. Second, variation pattern of dynamic feature contains useful information about patient's physical status, but adaptive pattern recognition is still a challenge. Third, existing works usually adopt a two-stage embedding process to process each dimension of the EHR data. However, some useful low-level information for PPRL will be lost. To address these issues, in this paper, we propose a fine-grained PPRL architecture named FG RL- N et for clinical risk prediction based on EHR. Specifically, we propose a medical Feature Correlation Detection Module (FCM) to effectively learn the feature correlations for each patient and a Temporal Variation Pattern Recognition Module (TVM) to effectively detect the variation patterns of each dynamic feature. Moreover, we design a Fine-Grained Representation Mechanism (FGRM) to preserve the low-level information (from both feature and visit dimensions) useful for risk prediction. In addition, in the stage of data preprocessing, We utilize generic medical classification knowledge to classify numerical dynamic data. We conduct the in-hospital mortality experiment and the decompensation experiment on a real-world dataset. The experiment results show that the FGRL-Net outperforms state-of-the-art approaches. The source code is provided in github https://***/JackyChio/FGRL-Net. © 2023 IEEE.

关键词：

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：