检索结果-内蒙古大学图书馆

Target flow field reconstruction based on dual tree complex wavelet transform and optical flow

Target flow field reconstruction based on dual tree complex ...

2023 Applied Optics and Photonics China: Laser technology and Applications;and Optoelectronic Devices and Integration, AOPC 2023

作者： Xu, Xinggui Li, Hong Zhang, Yue Ren, Weihe Zhang, Tao School of Information Yunnan University of Finance and Economics Kunming650032 China Beijing Institute of Space Mechanics Electricity Beijing100039 China School of Information Engineering and Automation Kunming University of Science and Technology Kunming650032 China

ISBN: (纸本)9781510672246

Background oriented system (BOS) technology is widely used for measuring flow field density information. The fingerprint information of high-speed target flow field can be calculated with the digital image correction (DIC) method. However, these traditional DIC algorithms are unable to obtain large-scale features of target flow fields and costing high computational complexity. To deal with those problems, a method combination of dual tree complex wavelet transform and optical flow (DTCWT-OF) is proposed. The proposed method adopts a much sparser gradient divergence regularization to obtain much more sparse statistical characteristics of the target flow field, and utilize the prior knowledge of the target flow field fingerprint information. Meanwhile, the reconstruction method is processed in the wavelet transform domain. Compared to traditional DIC methods, the experimental results show that the proposed method improves the SNR by 5dB and can achieve quasi real-time reconstruction. © 2023 SPIE.

关键词： Wavelet transforms

来源：评论

学校读者我要写书评

暂无评论

A Weighted Graph-based Method with Mapping of Material Composition to Target Performance

A Weighted Graph-based Method with Mapping of Material Compo...

引用

information Systems and Computing technology (ISCTech), International Conference on

作者： Chunxiu Wang Jiancheng Yin Tao Shen Yingli Liu Faculty of Information Engineering and Automation Kunming University of Science and Technology Kunming China Faculty of Materials Science and Engineering Kunming University of Science and Technology Kunming China

Materials informatics is a method that combines statistics, machine learning techniques, and high-throughput computing to promote materials discovery and design. In this study, we propose a novel framework based on weighted graph for mapping material compositions to properties. The framework utilizes distributed representations of elemental compositions and incorporates soft attention mechanisms to capture the importance of different elements in the materials. By leveraging attention mechanism and residual network, the model can effectively extract features and generate fixed-length representations for materials, enabling efficient information aggregation. Experimental results on diverse datasets from the Material Project database demonstrate the effectiveness of our proposed method, achieving less errors compared to other methods that do not consider the material structure. This data-driven deep learning method showcases the potential of materials informatics in providing accurate predictions for various material properties.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Soft sensor method for endpoint carbon temperature of BOF steelmaking based on FCM-WPPCR collaborative measure

引用

Kongzhi Lilun Yu Yingyong/Control Theory and Applications 2025年第5期42卷 1026-1038页

作者： Zhao, An Liu, Hui Chen, Fu-Gang Faculty of Information Engineering & Automation Kunming University of Science and Technology Yunnan Kunming650500 China Yunnan Key Laboratory of Artificial Intelligence Kunming University of Science and Technology Yunnan Kunming650500 China Yunnan Kunming Steel Electronic Information Technology Co. Ltd Yunnan Kunming650500 China

The accurate detection of the endpoint carbon temperature in basic oxygen furnace (BOF) steelmaking is the key to determining the final stage of steelmaking, and the nonlinear and multiworking condition data characteristics are the difficulties in effectively predicting the endpoint carbon temperature using soft sensing. A fuzzy C-means clustering weighted posterior probability and cluster relation (FCM-WPPCR) co metric strategy is proposed to address the problem of inaccurate sample matching caused by the difficulty of directly measuring relevant samples in just-in-time learning (JITL) under complex industrial data and the relatively single measurement basis. This strategy is used to select the optimal sample subset for JITL. This strategy uses fuzzy C-means clustering to partition multicondition sub clusters, and introduces the Earth mover’s distance (EMD) criterion to construct a posterior probability pattern to characterize the membership degree of the tested sample in each cluster;Secondly, after determining the correlation between other sub clusters and the maximum membership sub cluster through the maximum mean difference, a collaborative measurement mechanism is constructed by calculating the mean of the EMD distance between the test sample and each cluster sample to construct a weighted posterior probability and cluster relationship, in order to determine the execution criteria for selecting the test sample in each cluster;Finally, selecting relevant samples from each cluster to construct the optimal sample subset, and establishing a local regression model to predict the end-point carbon temperature. Through simulation of actual steelmaking production process data, the prediction accuracy of carbon content reaches 92.60% within the error range of ±0.02%, and the prediction accuracy of temperature reaches 93.30% within the error range of ±10◦C. © 2025 South China university of technology. All rights reserved.

关键词： Spray steelmaking

来源：评论

学校读者我要写书评

暂无评论

Cascaded information enhancement and cross-modal attention feature fusion for multispectral pedestrian detection

arXiv

引用

arXiv 2023年

作者： Yang, Yang Xu, Kaixiong Wang, Kaizheng Faculty of Information Engineering and Automation Kunming University of Science and Technology Kunming China Faculty of Electrical Engineering Kunming University of Science and Technology Kunming China

Multispectral pedestrian detection is a technology designed to detect and locate pedestrians in Color and Thermal images, which has been widely used in automatic driving, video surveillance, etc. So far most available multispectral pedestrian detection algorithms only achieved limited success in pedestrian detection because of the lacking take into account the confusion of pedestrian information and background noise in Color and Thermal images. Here we propose a multispectral pedestrian detection algorithm, which mainly consists of a cascaded information enhancement module and a cross-modal attention feature fusion module. On the one hand, the cascaded information enhancement module adopts the channel and spatial attention mechanism to perform attention weighting on the features fused by the cascaded feature fusion block. Moreover, it multiplies the single-modal features with the attention weight element by element to enhance the pedestrian features in the single-modal and thus suppress the interference from the background. On the other hand, the cross-modal attention feature fusion module mines the features of both Color and Thermal modalities to complement each other, then the global features are constructed by adding the cross-modal complemented features element by element, which are attentionally weighted to achieve the effective fusion of the two modal features. Finally, the fused features are input into the detection head to detect and locate pedestrians. Extensive experiments have been performed on two improved versions of annotations (sanitized annotations and paired annotations) of the public dataset KAIST. The experimental results show that our method demonstrates a lower pedestrian miss rate and more accurate pedestrian detection boxes compared to the comparison method. Additionally, the ablation experiment also proved the effectiveness of each module designed in this paper. Copyright © 2023, The Authors. All rights reserved.

关键词： Color

来源：评论

学校读者我要写书评

暂无评论

Phrase Decoupling Cross-Modal Hierarchical Matching and Progressive Position Correction for Visual Grounding

arXiv

引用

arXiv 2024年

作者： Xie, Minghong Wang, Mengzhao Li, Huafeng Zhang, Yafei Tao, Dapeng Yu, Zhengtao Faculty of Information Engineering and Automation Kunming University of Science and Technology Kunming650500 China FIST LAB School of Information Science and Engineering Yunnan University Kunming650091 China

Visual grounding has attracted wide attention thanks to its broad application in various visual language tasks. Although visual grounding has made significant research progress, existing methods ignore the promotion effect of the association between text and image features at different hierarchies on cross-modal matching. This paper proposes a Phrase Decoupling Cross-Modal Hierarchical Matching and Progressive Position Correction Visual Grounding method. It first generates a mask through decoupled sentence phrases, and a text and image hierarchical matching mechanism is constructed, highlighting the role of association between different hierarchies in cross-modal matching. In addition, a corresponding target object position progressive correction strategy is defined based on the hierarchical matching mechanism to achieve accurate positioning for the target object described in the text. This method can continuously optimize and adjust the bounding box position of the target object as the certainty of the text description of the target object improves. This design explores the association between features at different hierarchies and highlights the role of features related to the target object and its position in target positioning. The proposed method is validated on different datasets through experiments, and its superiority is verified by the performance comparison with the state-of-the-art methods. The source code of the proposed method is available at https://***/X7J92/VGNet. Copyright © 2024, The Authors. All rights reserved.

关键词： Visual languages

来源：评论

学校读者我要写书评

暂无评论

Adaptive Sensitivity Amplification Control with Q-learning for Cable-Driven Lower-Limb Exoskeletons

Adaptive Sensitivity Amplification Control with Q-learning f...

引用

第42届中国控制会议

作者： Linzhen Zhong Xiaodong Wang Guanbin Gao Yashan Xing Faxiang Zhang Faculty of Information Engineering & Automation Kunming University of Science and Technology Faculty of Mechanical & Electrical Engineering Kunming University of Science & Technology

This paper proposes a sensitivity amplification control(SAC) based on Q-learning for the cable-driven lower-limb exoskeleton(CDLEX) to guarantee the tracking performance and achieve comfortable human-robot interaction(HRI). There is no requirement for direct measurement of HRI in the SAC scheme, hence the force/torque sensors are avoided. For the purpose of accommodating the uncertainties of the interaction environment, such as the individual gaits and suddenly changing walking trajectories, the Q-learning algorithm is utilized to realize online parameter tuning of SAC. Finally, comparative numerical simulations are included to verify the effectiveness of the control scheme.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Knowledge-Guided Reinforcement Learning for Low-Resource Unsupervised Syllabification

Knowledge-Guided Reinforcement Learning for Low-Resource Uns...

引用

International Joint Conference on Neural Networks (IJCNN)

作者： Wenjun Wang Ling Dong Zhengtao Yu Yuxin Huang Junjun Guo Shengxiang Gao School of Information Engineering and Automation Kunming University of Science and Technology Kunming China

ISBN: (数字)9798350359312

ISBN: (纸本)9798350359329

Syllabification is a crucial task in natural language processing, and syllables also play a significant role as modeling units in speech processing. While deep learning methods have shown remarkable progress in syllabification, they face challenges in low-resource languages where ready-made segmentation datasets or rules are lacking. Large language models (LLMs) are mostly unsuitable for these low-resource languages as well. To address these challenges, this paper proposes an unsupervised syllabification approach that incorporates logical reasoning into the reinforcement learning training process, achieving knowledge-guided syllabification. By introducing logical reasoning knowledge and modeling the interaction between the Agent and the knowledge base (KB), the model gains a better understanding of language structures and patterns. The study primarily focuses on low-resource Lao language, with experiments conducted on a publicly available English dataset to validate the effectiveness of the proposed method.

关键词： Training Deep learning Large language models Knowledge based systems Neural networks Reinforcement learning Natural language processing

来源：评论

学校读者我要写书评

暂无评论

Single-Image HDR Reconstruction Assisted Ghost Suppression and Detail Preservation Network for Multi-Exposure HDR Imaging

arXiv

引用

arXiv 2024年

作者： Li, Huafeng Yang, Zhenmei Zhang, Yafei Tao, Dapeng Yu, Zhengtao The Faculty of Information Engineering and Automation Kunming University of Science and Technology Kunming650500 China FIST LAB School of Information Science and Engineering Yunnan University Kunming650091 China

The reconstruction of high dynamic range (HDR) images from multi-exposure low dynamic range (LDR) images in dynamic scenes presents significant challenges, especially in preserving and restoring information in oversaturated regions and avoiding ghosting artifacts. While current methods often struggle to address these challenges, our work aims to bridge this gap by developing a multi-exposure HDR image reconstruction network for dynamic scenes, complemented by single-frame HDR image reconstruction. This network, comprising single-frame HDR reconstruction with enhanced stop image (SHDR-ESI) and SHDR-ESI-assisted multi-exposure HDR reconstruction (SHDR-A-MHDR), effectively leverages the ghost-free characteristic of single-frame HDR reconstruction and the detail-enhancing capability of ESI in oversaturated areas. Specifically, SHDR-ESI innovatively integrates single-frame HDR reconstruction with the utilization of ESI. This integration not only optimizes the single image HDR reconstruction process but also effectively guides the synthesis of multi-exposure HDR images in SHDR-A-MHDR. In this method, the single-frame HDR reconstruction is specifically applied to reduce potential ghosting effects in multi-exposure HDR synthesis, while the use of ESI images assists in enhancing the detail information in the HDR synthesis process. Technically, SHDR-ESI incorporates a detail enhancement mechanism, which includes a self-representation module and a mutual-representation module, designed to aggregate crucial information from both reference image and ESI. To fully leverage the complementary information from non-reference images, a feature interaction fusion module is integrated within SHDR-A-MHDR. Additionally, a ghost suppression module, guided by the ghost-free results of SHDR-ESI, is employed to suppress the ghosting artifacts. Experimental results on four public datasets demonstrate the efficacy and superiority of the proposed method. The code is available at https://github.c

关键词： Image reconstruction

来源：评论

学校读者我要写书评

暂无评论

Decoupling-Enhanced Vietnamese Speech Recognition Accent Adaptation Supervised by Prosodic Domain information

Decoupling-Enhanced Vietnamese Speech Recognition Accent Ada...

引用

International Joint Conference on Neural Networks (IJCNN)

作者： Yanwen Fang Wenjun Wang Ling Dong Shengxiang Gao Hua Lai Zhengtao Yu Faculty of Information Engineering and Automation Kunming University of Science and Technology Kunming China Yunnan Key Laboratory of Artificial Intelligence Kunming University of Science and Technology Kunming China

ISBN: (数字)9798350359312

ISBN: (纸本)9798350359329

The northern and southern accents of the Vietnamese exhibit pronunciation differences in terms of tones and rhythms. Existing Vietnamese pre-trained speech models show a bias in accent representation, leading to a lack of generalization capability for southern accents in Vietnamese speech recognition models and a noticeable decrease in recognition accuracy. In this paper, we propose a decoupled-enhanced adaptive representation strategy guided by prosody and accent domain label information. Through Domain Adversarial Training (DAT), the pre-trained speech model decouples domain-invariant content features. Simultaneously, by incorporating prosodic features, we enhance the pronunciation information of Vietnamese, enabling adaptive representation of the distinctive characteristics of northern and southern Vietnamese accents. Evaluating our proposed method on a dataset of Vietnamese accents, the results demonstrate its superiority over existing approaches, mitigating the performance degradation in Vietnamese speech recognition models caused by accent differences.

关键词： Training Degradation Adaptation models Accuracy Scalability Neural networks Focusing

来源：评论

学校读者我要写书评

暂无评论

Knowledge Graph Construction: State-of-the Art Techniques and Future Opportunities

Knowledge Graph Construction: State-of-the Art Techniques an...

引用

information Systems and Computing technology (ISCTech), International Conference on

作者： Qingwang Wang Chaohui Li Yinxing Zhang Yi Liu Hao Xiong Tao Shen School of Information Engineering and Automation Kunming University of Science and Technology Kunming China

ISBN: (数字)9798350379860

ISBN: (纸本)9798350379877

Knowledge graphs represent structured mappings of human knowledge playing a pivotal role in the iterative development and deepening of domain understandings. Methods for knowledge graph construction are continually evolving. Advancements in artificial intelligence, particularly with the advent of Generative Pre-trained Transformers (GPT), have paved the way for end-to-end automated methods for constructing knowledge graphs. This study provides a comprehensive evaluation of three conventional methodologies for constructing knowledge graphs: Top-Down, Bottom-Up and Joint approaches. We survey the associated technologies and present classic knowledge graph cases. We then elucidate the integration of large language models into the automation of knowledge graph construction, including their functions, benefits and limitations. Finally, we offer a prospective outlook on future research trajectories in this field.

关键词： Training Surveys Adaptation models automation Large language models Computational modeling Knowledge graphs Transformers Trajectory Security

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：