检索结果-内蒙古大学图书馆

2nd International Conference on Robotics, Automation and Intelligent Control, ICRAIC 2022

作者： Li, He Jia, Huimei Zhang, Rui Yang, Shuaipeng Qi, Qinglei Liu, Tianyang Henan Costar Group Co. Ltd Nanyang China College of Information Engineering Henan University of Science and Technology Luoyang China Henan Engineering Research Center of Intelligent Processing for Big Data of Digital Image School of Computer Science and Technology Nanyang Normal University Nanyang China Sydney Business School Shanghai University Shanghai China

For the detection of objects floating in the river, most of the traditional intelligent video monitoring methods are used to monitor through manual viewing. During the traditional video data transmission and processing, the massive amount of data will put pressure on the background such as network and computing power. The new effective method is to crop and optimize YOLOV5s, add a specific image pre-processing module, and deploy it by edge computing, embed a SOC (System on Chip) chip in the web camera for real-time processing of video data. Simulation experiments prove that the above method can improve the accuracy and speed of the target detection network. © Published under licence by IOP Publishing Ltd.

关键词： Edge computing

来源：评论

学校读者我要写书评

暂无评论

CNN based Emotion Detection in Cross Linguistic Children speech 24

CNN based Emotion Detection in Cross Linguistic Children spe...

引用

24th International Congress on Acoustics, ICA 2022

作者： Lyakso, Elena Frolova, Olga Ruban, Nersisson Mekala, A. Mary Joseph Raj, Alex Noel Matveev, Anton Matveev, Yuri The Child Speech Research Group St. Petersburg State University St. Petersburg Russia School of Electrical Engineering VIT Vellore India School of Information Technology & Engineering VIT Vellore India Key Laboratory of Digital Signal and Image Processing of Guangdong Province Department of Electronics Engineering College of Engineering Shantou University China

The goal of this cross-linguistic study is to analyze the speech of children in the age group of 8-12 years and automatic classification of children's emotions from various speech data in Russian and Tamil languages. A standardized approach for the collection of speech (spontaneous and acting speech - emotional words, phrases, and meaningless texts) was used. Two emotional child speech corpora include the emotional speech of 95 Russian children and 40 Indian children were created. Annotation of the emotional speech was carried out in four categories "joy - neutral - sadness – anger" by two speech specialists of the same nationality of the child. The set of 2505 labeled audio files of Russian children and 418 labeled audio files of Tamil children were used. The SVM classifier shows slightly better results in Russian language, and the MLP classifier in Tamil. Inter-Cultural approach (mixed dataset of Russian and Indian speech) revealed that the accuracy of recognition of all emotions remains higher. In Cross-Cultural approach, on samples of Tamil speech, anger state is recognized better, and in samples of Russian speech – sad state. Using 5 Layered CNN Model, the accuracy was revealed for Russian speech is higher than for Indian children. © 2022 Proceedings of the International Congress on Acoustics. All rights reserved.

关键词： Emotion Recognition

来源：评论

学校读者我要写书评

暂无评论

Speaker Diarization based on Deep Learning Techniques: A Review 6

Speaker Diarization based on Deep Learning Techniques: A Rev...

引用

6th International Symposium on Multidisciplinary Studies and Innovative Technologies, ISMSIT 2022

作者： Al-Hadithy, Thaer M. Frikha, Mondher Maseer, Zaidoon Kamil ATISP Research Unit Ecole Nationale D 'Electronique et de Telecommunications de Sfax University Sfax Tunisia Directeur de l'Unite de Recherche: Advanced Technologies for Image and Signal Processing University Sfax Tunisia Technical University of Malaysia Malacca Faculty of Information and Communications Technology Malaysia

ISBN: (数字)9781665470131

ISBN: (纸本)9781665470131

Speaker diarization is a function that recognizes 'who was speaking at the phase' by organizing video and audio recordings with sets that correspond to the presenter's personality. Speaker diarization approaches for multi-speaker audio recordings in the domain of speech recognition were developed in the first few years to allow speaker adaptive technology. For a long time, the self-contained use of these technologies grew in relevance as a means of facilitating meta-information for downstream speaker-specific difficulties, such as audio retrieval. Speaker diarization has made rapid progress in recent years, thanks to the advent of deep learning technology, which has led to ground-breaking shifts in research and practices in the field of speech implementation. The goal of this work is to provide an in-depth review of the studies published in the last 20 years. In the proposed paper, a total of 85 studies have been selected. In this paper, we tried to highlight recent improvements in speaker diarization and challenges to help the researchers who are working in this field. © 2022 IEEE.

关键词： Audio recordings

来源：评论

学校读者我要写书评

暂无评论

Exploring the Latest Highlights in Medical Natural Language processing across Multiple Languages: A Survey

引用

Yearbook of Medical Informatics 2023年第1期32卷 230-243页

作者： Shaitarova, Anastassia Zaghir, Jamil Lavelli, Alberto Krauthammer, Michael Rinaldi, Fabio Department of Computational Linguistics University of Zurich Zurich Switzerland Division of Medical Information Sciences University Hospitals of Geneva Geneva Switzerland Department of Radiology and Medical Informatics University of Geneva Geneva Switzerland Natural Language Processing Research Unit Center for Digital Health and Wellbeing Fondazione Bruno Kessler Trento Italy Department of Quantitative Biomedicine University of Zurich Zurich Switzerland Biomedical Informatics University Hospital Zurich Zurich Switzerland Dalle Molle Institute for Artificial Intelligence Research Lugano Switzerland Swiss Institute of Bioinformatics Switzerland

OBJECTIVES: This survey aims to provide an overview of the current state of biomedical and clinical Natural Language processing (NLP) research and practice in Languages other than English (LoE). We pay special attention to data resources, language models, and popular NLP downstream tasks. METHODS: We explore the literature on clinical and biomedical NLP from the years 2020-2022, focusing on the challenges of multilinguality and LoE. We query online databases and manually select relevant publications. We also use recent NLP review papers to identify the possible information lacunae. RESULTS: Our work confirms the recent trend towards the use of transformer-based language models for a variety of NLP tasks in medical domains. In addition, there has been an increase in the availability of annotated datasets for clinical NLP in LoE, particularly in European languages such as Spanish, German and French. Common NLP tasks addressed in medical NLP research in LoE include information extraction, named entity recognition, normalization, linking, and negation detection. However, there is still a need for the development of annotated datasets and models specifically tailored to the unique characteristics and challenges of medical text in some of these languages, especially low-resources ones. Lastly, this survey highlights the progress of medical NLP in LoE, and helps at identifying opportunities for future research and development in this field. IMIA and Thieme. This is an open access article published by Thieme under the terms of the Creative Commons Attribution-NonDerivative-NonCommercial License, permitting copying and reproduction so long as the original work is given appropriate credit. Contents may not be used for commercial purposes, or adapted, remixed, transformed or built upon. (https://***/licenses/by-nc-nd/4.

关键词： datasets as topic language models Multilingualism natural language processing shared tasks

来源：评论

学校读者我要写书评

暂无评论

Optimal Training Positive Sample Size Determination for Deep Learning Caries Recognition Model Establishment with Cbct images

SSRN

引用

SSRN 2024年

作者： Wang, Yanlin Li, Gang Zhang, Xinyue Wang, Yue Zhang, Zhenhao Li, Jupeng Ma, Junqi Wang, Lingxing Department of Oral and Maxillofacial Radiology Peking University School Hospital of Stomatology National Center for Stomatology National Clinical Research Center for Oral Diseases National Engineering Research Center of Oral Biomaterials and Digital Medical Device Beijing Key Laboratory of Digital Stomatology NHC Key Laboratory of Digital Stomatology Beijing China Signal and image processing laboratory School of Electronic Information Engineering Beijing Jiao tong University Beijing China Algorithm Department YOFO Medical Technology Co. Ltd Anhui China

Background and objective When performing model training, it is necessary to consider the balance between the impact of sample size, actual resource and time constraints. To do so, the single-arm objective performance criteria for deep learning optimal training positive sample size determination was proposed for caries recognition and identified with three network models. Methods The expected sensitivity of 0.6 and specificity of 0.85 were applied to the single-arm objective performance criteria calculation formula, resulting in an optimal training set with 263 positive samples. According to a ratio of 8:1:1 for the distribution of training set, validation set and test set, the U-Net, YOLOv5n and CariesDetectNet were trained and validated on the clinically self-collected cone-beam computed tomography (CBCT) images. To evaluate performance, an additional set of test sets was used to test the three models. The accuracy of the three best-performed models for caries detection was evaluated through comparisons with the detection accuracy of caries by two dental radiologists. Results When the number of carious teeth reached about 250, the models reached the optimal level of performance. Among the three models, the U-Net performed best, with accuracy, sensitivity, specificity, F1-Score, Dice similarity coefficients, and intersection over union of 0.9929, 0.9307, 0.9989, 0.9590, 0.9435 and 0.9008, respectively. And the three models performed higher accuracy than dental radiologists for caries recognition. Conclusions This study indicated that the optimal number of images with caries in the training set was predictable and it could be calculated through the single-arm objective performance criteria. © 2024, The Authors. All rights reserved.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

A method for detecting floating objects on water based on edge computing

A method for detecting floating objects on water based on ed...

引用

2023 IEEE International Symposium on Broadband Multimedia Systems and Broadcasting, BMSB 2023

作者： Li, He Yang, Shuaipeng Liu, Jinjiang Fang, Honglin Fu, Zhumu Zhang, Rui Jia, Huimei Lv, Lianmeng Henan Costar Group Co. Ltd Henan Nanyang473003 China Henan University of Science and Technology College of Information Engineering Henan Luoyang471000 China Nanyang Normal University Henan Engineering Research Center of Intelligent Processing for Big Data of Digital Image School of Computer Science and Technology Nanyang473061 China Beijing University of Posts and Telecommunication State Key Laboratory of Networking and Switching Technology Beijing100876 China Xi'an Hengpin Electronic Technology Co. Ltd Xi'an710100 China

ISBN: (纸本)9798350321524

With the development and application of computer vision, many target detection networks are applied to the detection of floating objects in rivers. For the detection problems such as small targets easily missed and misdetected in water surface floating object detection tasks and difficult to deploy models. An edge computing-oriented approach to river floater detection is proposed. First, a four-fold down-sampling feature layer is added to the YOLOv5 network which enhances more target detail features and improves the detection capability of small objects. Second, CA (Coordinate Attention) is added to the Backbone to suppress background noise interference, and different pooling is used to accommodate different hierarchical features. Then, a bilinear interpolation method is adopted for up-sampling to avoid the loss of small object features. Design a data enhancement algorithm for small targets based on Mosaic to increase the number of small objects and enrich the training background. Finally, for the edge computing architecture platform, the channel pruning algorithm is used to prune and compress the model structure to adapt to the computing capability of edge devices. The experimental results show that the method can effectively improve the detection capability of the network for floating objects on the water surface. The detection accuracy can reach 93.6%, and the detection speed can be maintained at 36 frames per second, which can achieve high-precision real-time detection of floating objects on the water surface. © 2023 IEEE.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

A Shape-Supervised Feature Fusion U-Net for Tubular Structure Segmentation

SSRN

引用

SSRN 2024年

作者： Yue, Jinghua Jin, Shuo Wang, Siyuan Zeng, Jianping Shan, Siqiao Liu, Bo Jiang, Nan Zhou, Fugen Image Processing Center Beihang University Beijing China Hepatopancreatobiliary Center Beijing Tsinghua Changgung Hospital Key Laboratory of Digital Intelligence Hepatology Ministry of Education School of Clinical Medicine Tsinghua University Beijing China Research Unit of Precision Hepatobiliary Surgery Paradigm Chinese Academy of Medical Sciences Beijing China Institute for Precision Medicine Tsinghua University Beijing China

Accurate segmentation of tubular structures, such as blood vessels and bile ducts, is pivotal for clinical diagnosis and subsequent treatment. However, challenges arise from their thin elongated structures, complex morphologies, and complex backgrounds, making it difficult to extract detailed and shape features and consequently posing challenges for automated segmentation. This paper introduces a novel approach, the shape-supervised feature fusion U-Net (SFU-Net), to address the segmentation of tubular structures. The proposed network extends the U-Net architecture by integrating a feature fusion module, effectively fuses different semantic information in both the encoder and decoder paths to facilitate the acquisition of detailed geometric representations of tubular structures. Furthermore, a novel distance-based tubularity loss is designed to capture boundary information and supervise segmentation integrity by imposing shape constraints. We conducted comprehensive experiments on three tubular structure segmentation tasks in medical imaging, including dilated biliary tree, hepatic vessel, and pulmonary artery. The experimental results illustrate the superior performance of our proposed method compared to other state-of-the-art approaches, underscoring its effectiveness in medical image segmentation with tubular structures. © 2024, The Authors. All rights reserved.

关键词： Blood vessels

来源：评论

学校读者我要写书评

暂无评论

Hyperspectral Emissivity Estimation and Application Based on Machine Learning Method

Hyperspectral Emissivity Estimation and Application Based on...

引用

IEEE International Symposium on Geoscience and Remote Sensing (IGARSS)

作者： Xiujuan Li Hua Wu José Antonio Sobrino State Key Laboratory of Resources and Environmental Information System Institute of Geographic Sciences and Natural Resources Research Chinese Academy of Sciences Beijing China Global Change Unit Image Processing Laboratory University of Valencia Valencia Spain University of Chinese Academy of Sciences Beijing China School of Resources and Environment University of Electronic Science and Technology of China Chengdu China

ISBN: (数字)9798350360325

ISBN: (纸本)9798350360332

Land surface emissivity (LSE) can reflect the unique spectral characteristics of land surfaces, which holds significant research value and promising applications. However, due to the limitations of spaceborne hyperspectral thermal infrared (TIR) sensors, there is currently a lack of high spatial resolution LSE imagery. Therefore, this study proposes a methodology based on machine learning and mathematical statistical approaches to extend multispectral emissivity values to hyperspectral LSE spectrums. Furthermore, the Spectral Angle Mapper (SAM) method was used for target identification. The results indicated that the proposed method can estimate hyperspectral LSE curves in high accuracy. Moreover, the target ground class can be identified accurately by matching the spectral angle with the measured LSE curve. This has crucial implications for applications such as land cover classification, mineral identification, geological exploration and soil property research.

关键词： Training Accuracy Databases Land surface Estimation Machine learning Thermal sensors

来源：评论

学校读者我要写书评

暂无评论

A method for detecting floating objects on water based on edge computing

A method for detecting floating objects on water based on ed...

引用

IEEE International Symposium on Broadband Multimedia Systems and Broadcasting (BMSB)

作者： He Li Shuaipeng Yang Jinjiang Liu Honglin Fang Zhumu Fu Rui Zhang Huimei Jia Lianmeng Lv Henan Costar Group Co. Ltd Nanyang Henan China College of Information Engineering Henan University of Science and Technology Luoyang Henan China Henan Engineering Research Center of Intelligent Processing for Big Data of Digital Image School of Computer Science and Technology Nanyang Normal University Nanyang China State Key Laboratory of Networking and Switching Technology Beijing University of Posts and Telecommunication Beijing China Xi'an Hengpin Electronic Technology Co. Ltd Xi’an China

关键词：

来源：评论

学校读者我要写书评

暂无评论

Reversible adversarial information hiding based on the interpretation of neural network

引用

digital Signal processing: A Review Journal 2025年 165卷

作者： Li, Ruiping Wang, Xi Cui, Qingchen Li, Ming Information Construction and Management Office Henan Normal University Xinxiang453007 China School of Artificial Intelligence and Software Engineering Nanyang Normal University Nanyang473061 China Henan Engineering Research Center of Intelligent Processing for Big Data of Digital Image Henan Nanyang473000 China College of Information Engineering Zhoukou Vocational College of Arts and Science Zhoukou466000 China College of Computer and Information Engineering Henan Normal University Xinxiang453007 China Key Laboratory of Artificial Intelligence and Personalized Learning in Education of Henan Province Xinxiang453007 China

The emerging technology of adversarial information hiding can generate adversarial example by embedding useful information instead of meaningless noise into the host image. The obtained image can deceive classification models and therefore equip the function of privacy protection. However, existing research mainly focuses on the effectiveness of attack rather than preserving the integrity of the protected image. It cannot meet the requirements in the fields such as medical imaging, financial transactions, and copyright protection where users' raw data should not be damaged. Therefore, we propose a novel reversible adversarial information hiding based on interpretability of neural networks. The Grad-CAM-generated heatmap is employed to identify a minimal set of high-impact pixels for embedding, ensuring that minor modifications can induce significant misclassification. Then, the user-defined secret bits are embedded into the identified pixels by using difference expansion. After extraction, the original image can be perfectly restored. The technologies of adversarial example and reversible information hiding are combined to accommodate wider applications. The proposed method is tested on CIFAR-10 with three different neural architectures (NiN, AlexNet, ResNet). Experimental results show that the proposed method can completely restore the cover image while ensuring the extraction of the embedded data and misleading the neural network. © 2025 Elsevier Inc.

关键词： Adversarial attacks Adversarial information hiding Neural network interpretability Reversibility

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：