检索结果-内蒙古大学图书馆

2024 China Automation Congress, CAC 2024

作者： Fan, Xing-Rong Chen, Hui-Ting Deng, Yu-He Wan, Hui Cai, Zong-Rong Peng, Qiu-Ju Engineering Research Centre for Waste Oil Recovery Technology and Equipment Ministry of Education Chongqing Technology and Business University Chongqing China Chongqing Key Laboratory of Intelligent Perception and Block Chain Technology Chongqing Technology and Business University Chongqing China School of Artificial Intelligence School of Computer Science and Information Engineering Chongqing Technology and Business University Chongqing China

ISBN: (纸本)9798350368604

Accurate identification and classification of bird species are crucial for advancing research and conservation efforts, enhancing our understanding of population dynamics, and implementing targeted conservation strategies. Despite its importance, this task is challenging due to the wide variation in bird appearances and environmental noise, which can lead to poor generalization and increased complexity. To address these challenges, this paper proposes a novel bird species recognition method, EN-BLS, which integrates the EfficientNet-B0 deep learning architecture with a broad learning system. EfficientNet-B0 is used as the backbone network for feature extraction from bird images, with the final feature map layer serving as input for the broad learning system. The resulting hybrid model, EN-BLS, is designed for bird species classification. The model's performance is evaluated using the Birds 525 Species dataset, the latest publicly available benchmark dataset with the highest number of categories collected from diverse real-world environments. Experimental results show that EN-BLS achieves a recognition accuracy of 93.79% with lower computational complexity compared to existing methods, demonstrating its effectiveness for bird species diversity monitoring and conservation efforts. © 2024 IEEE.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

AE-BEGAN based Synthetic Data Augmentation for Sample-Limited High-Dimensional Problems with Application to NIR Spectral Data

AE-BEGAN based Synthetic Data Augmentation for Sample-Limite...

引用

International Petroleum and Petrochemical technology Conference 2023, IPPTC 2023

作者： Luo, Xin-Yue Fan, Xing-Rong Zhang, Xian-Ming Chen, Tian-Yu Huang, Chun-Jie Engineering Research Centre for Waste Oil Recovery Technology and Equipment Ministry of Education Chongqing Technology and Business University Chongqing China Chongqing Key Laboratory of Intelligent Perception and Block Chain Technology Chongqing Technology and Business University Chongqing China Chongqing Intelligent Supply Chain Engineering and Technology Research Center Chongqing Jiaotong University Chongqing China Chongqing Haolongyu Technology Co. Ltd. Chongqing China

Synthetic data augmentation holds substantial research and application value in scenarios characterized by limited samples and high dimensions. It enhances the analytical ability and efficiency of spectral analysis models. This paper proposes Autoencoder-Combined Boundary Equilibrium Generative Adversarial Networks (AE-BEGAN) as a new method for augmenting synthetic data in scenarios with limited samples and high dimensions, with a specific emphasis on near-infrared (NIR) spectral data. The spectral data first undergoes preprocessing procedures that encompass advanced noise reduction algorithms and techniques for removing abnormal samples, guaranteeing elimination of unwanted disturbances and outliers. Then, the pre-processed data is utilized to train the AE-BEGAN model, which generates augmented synthetic samples. Finally, real NIR spectral data obtained from lubricant samples exhibiting different water contents were employed to validate and test the performance of the model. The experimental results demonstrate that the AE-BEGAN model outperforms other GANs in generating synthetic data of high quality and diversity, as quantified by two evaluation metrics, α-Precision and β-Recall with scores of approximately 0.86 and 0.28, respectively. The application case study confirms that the AE-BEGAN model exhibits the capability to generate derived NIR spectra and expand the number of spectra in scenarios with limited samples and high dimensions. © Published under licence by IOP Publishing Ltd.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Valley edge states and chiral selective transport in reconfigurable phononic crystals

引用

The European Physical Journal B 2023年第11期96卷 1-8页

作者： Xi, Feng Tang, Yuxia Hu, Li School of Computer Science and Information Engineering Chongqing Technology and Business University Chongqing China Chongqing Key Laboratory of Intelligent Perception and Block Chain Technology Chongqing Technology and Business University Chongqing China College of Optoelectronic Engineering Chongqing University Chongqing China

Valley topological edge state has been extended from condensed matter physics to acoustics and rapidly developed, in which one-way propagation in valley topological phononic crystal is achieved breaking the spatial reversal symmetry. Here, we present a 2D hexagonal lattice phononic crystal with the reconfigurable scatters composed of three rectangle rods. By rotating its three-legged rods, two distinct valley edge states can be realised and the valley Hall topological transition phase also is triggered. We numerically simulate the projected band structures of the supercells including zigzag and armchair interface and verify the existence of valley edge states. We demonstrate the acoustic topological transport in the straight interface and sharply curved path. Then, in an across-interface, acoustic splitting and emerging are also observed. Furthermore, we design the complicated interfaces constructed by phononic crystals with distinct valley Hall topological phase. We verify the chiral selective transports in the oriented interface if the excited point source positioned at different interfaces. The chiral selective valley topological propagation of acoustic wave provides many potential applications in designing acoustic switch, splitter and acoustic waveguide.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Bird Species Recognition by Integrating EfficientNet-B0 Deep Learning Architecture and Broad Learning System

Bird Species Recognition by Integrating EfficientNet-B0 Deep...

引用

Chinese Automation Congress (CAC)

作者： Xing-Rong Fan Hui-Ting Chen Yu-He Deng Hui Wan Zong-Rong Cai Qiu-Ju Peng Engineering Research Centre for Waste Oil Recovery Technology and Equipment Ministry of Education Chongqing Technology and Business University Chongqing China Chongqing Key Laboratory of Intelligent Perception and Block Chain Technology Chongqing Technology and Business University Chongqing China School of Artificial Intelligence & School of Computer Science and Information Engineering Chongqing Technology and Business University Chongqing China

ISBN: (数字)9798350368604

ISBN: (纸本)9798350368611

Accurate identification and classification of bird species are crucial for advancing research and conservation efforts, enhancing our understanding of population dynamics, and implementing targeted conservation strategies. Despite its importance, this task is challenging due to the wide variation in bird appearances and environmental noise, which can lead to poor generalization and increased complexity. To address these challenges, this paper proposes a novel bird species recognition method, EN-BLS, which integrates the EfficientNet-B0 deep learning architecture with a broad learning system. EfficientNet-B0 is used as the backbone network for feature extraction from bird images, with the final feature map layer serving as input for the broad learning system. The resulting hybrid model, EN-BLS, is designed for bird species classification. The model's performance is evaluated using the Birds 525 Species dataset, the latest publicly available benchmark dataset with the highest number of categories collected from diverse real-world environments. Experimental results show that EN-BLS achieves a recognition accuracy of 93.79 % with lower computational complexity compared to existing methods, demonstrating its effectiveness for bird species diversity monitoring and conservation efforts.

关键词： Learning systems Accuracy Computational modeling Working environment noise Deep architecture Birds Feature extraction Environmental management Protection Monitoring

来源：评论

学校读者我要写书评

暂无评论

EmMixformer: Mix transformer for eye movement recognition

arXiv

引用

arXiv 2024年

作者： Qin, Huafeng Zhu, Hongyu Jin, Xin Song, Qun El-Yacoubi, Mounim A. Gao, Xinbo Chongqing intelligence perception and block chain technology key laboratory The School of Computer Science and Information Engineering Chongqing Technology and Business University Chongqing400067 China SAMOVAR Telecom SudParis Institute Polytechnique de Paris Palaiseau91120 France Chongqing Key Laboratory of Image Cognition Chongqing University of Posts and Telecommunications Chongqing400065 China

Eye movement (EM) is a new highly secure biometric behavioral modality that has received increasing attention in recent years. Although deep neural networks, such as convolutional neural network (CNN), have recently achieved promising performance, current solutions fail to capture local and global temporal dependencies within eye movement data. To overcome this problem, we propose in this paper a mixed transformer termed EmMixformer to extract time and frequency domain information for eye movement recognition. To this end, we propose a mixed block consisting of three modules, transformer, attention Long short-term memory (attention LSTM), and Fourier transformer. We are the first to attempt leveraging transformer to learn long temporal dependencies within eye movement. Second, we incorporate the attention mechanism into LSTM to propose attention LSTM with the aim to learn short temporal dependencies. Third, we perform self attention in the frequency domain to learn global features. As the three modules provide complementary feature representations in terms of local and global dependencies, the proposed EmMixformer is capable of improving recognition accuracy. The experimental results on our eye movement dataset and two public eye movement datasets show that the proposed EmMixformer outperforms the state of the art by achieving the lowest verification error. Copyright © 2024, The Authors. 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

Adversarial Masking Contrastive Learning for vein recognition

arXiv

引用

arXiv 2024年

作者： Qin, Huafeng Wu, Yiquan El-Yacoubi, Mounim A. Wang, Jun Yang, Guangxiang Chongqing Engineering Laboratory of intelligent perception and block-chain School of Computer Science and Information Engineering Chongqing Technology and Business University Chongqing400067 China SAMOVAR Telecom SudParis CNRS University Paris-Saclay Paris91011 France School of Information and Control Engineering China University of Mining and Technology Xuzhou China

Vein recognition has received increasing attention due to its high security and privacy. Recently, deep neural networks such as Convolutional neural networks (CNN) and Transformers have been introduced for vein recognition and achieved state-of-the-art performance. Despite the recent advances, however, existing solutions for finger-vein feature extraction are still not optimal due to scarce training image samples. To overcome this problem, in this paper, we propose an adversarial masking contrastive learning (AMCL) approach, that generates challenging samples to train a more robust contrastive learning model for the downstream palm-vein recognition task, by alternatively optimizing the encoder in the contrastive learning model and a set of latent variables. First, a huge number of masks are generated to train a robust generative adversarial network (GAN). The trained generator transforms a latent variable from the latent variable space into a mask space. Then, we combine the trained generator with a contrastive learning model to obtain our AMCL, where the generator produces challenging masking images to increase the contrastive loss and the contrastive learning model is trained based on the harder images to learn a more robust feature representation. After training, the trained encoder in the contrastive learning model is combined with a classification layer to build a classifier, which is further fine-tuned on labeled training data for vein recognition. The experimental results on three databases demonstrate that our approach outperforms existing contrastive learning approaches in terms of improving identification accuracy of vein classifiers and achieves state-of-the-art recognition results. Copyright © 2024, The Authors. All rights reserved.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：