检索结果-内蒙古大学图书馆

CycleMatch: Cyclic pseudo-labeling distillation in semi-supervised medical image segmentation☆

PATTERN RECOGNITION LETTERS 2025年 193卷 135-141页

作者： Wang, Chunshi Xiong, Chuan Hao, Bin Z. Ding, Shuxue Guilin Univ Elect Technol Sch Artificial Intelligence Guangxi Coll & Univ Key Lab AI Algorithm Engn Guilin 541004 Guangxi Zhuang Peoples R China

In this study, we present a semi-supervised medical image segmentation framework called CycleMatch, which aims to tackle the dependency of fully supervised methods on a large amount of labeled data. By integrating a cyclic pseudo-label distillation mechanism with image-level and feature-level perturbations, CycleMatch effectively leverages unlabeled data to enhance model performance and robustness. Experimental results demonstrate that CycleMatch outperforms existing semi-supervised methods across various data annotation ratios, particularly excelling in scenarios with limited labeled data. Additionally, an in-depth analysis of feature perturbation types and parameter choices further validates CycleMatch's effectiveness and adaptability in handling different medical image datasets. Overall, CycleMatch offers a new solution for medical image segmentation, showcasing the potential for achieving efficient and accurate segmentation even with limited data.

关键词： Semi-supervised learning Self-knowledge distillation Medical image segmentation

来源：评论

学校读者我要写书评

暂无评论

Low hysteresis, water retention, anti-freeze multifunctional hydrogel strain sensor for human-machine interfacing and real-time sign language translation

引用

MATERIALS HORIZONS 2024年第16期11卷 3856-3866页

作者： Zhou, Lijuan Zhao, Bin Liang, Jingye Lu, Fangying Yang, Weiping Xu, Jishuai Zheng, Jingxuan Liu, Yong Wang, Run Liu, Zunfeng Tiangong Univ Sch Text Sci & Engn 399 West Binshui Rd Tianjin 300387 Peoples R China Guangxi Coll Univ Key Lab AI Algorithm Engn Guangxi Coll & Univ Key Lab AI Algorithm Engn Sch Artificial Intelligence Guilin 541004 Guangxi Peoples R China Nankai Univ Frontiers Sci Ctr New Organ Matter State Key Lab Med Chem Biol Key Lab Funct Polymer MatColl Chem Tianjin 300071 Peoples R China

Hydrogel strain sensors have received increasing attention due to their potential applications in human-machine interfaces and flexible electronics. However, they usually suffer from both mechanical and electrical hysteresis and poor water retention, which limit their practical applications. To address this challenge, a poly(acrylic acid-co-acrylamide) hydrogel crosslinked by silica nanoparticles is fabricated via photo polymerization and salting-out of hydrophilic ions for the strain sensor. The resulting hydrogel strain sensor possessed low electrical hysteresis (1.6%), low mechanical hysteresis (<7%), high cycle stability (>10 000 cycles), high durability, water retention and anti-freezing ability. Moreover, this strain sensor can be used as a wearable sensor for real-time control of robotic hands and hand gesture recognition. Finally, a sign language translation system has been demonstrated with the aid of machine learning, achieving recognition rates of over 98% for 15 different sign languages. This work offers a promising prospect for human-machine interfaces, smart wearable devices, and the Internet of Things.

关键词： Hydrogels

来源：评论

学校读者我要写书评

暂无评论

Foggy Drone Teacher: Domain Adaptive Drone Detection Under Foggy Conditions

引用

DRONES 2025年第2期9卷 146-146页

作者： Zheng, Guida Tan, Benying Wu, Jingxin Qin, Xiao Li, Yujie Ding, Shuxue Guilin Univ Elect Technol Guangxi Coll & Univ Key Lab AI Algorithm Engn Sch Artificial Intelligence Guilin 541004 Peoples R China Guangxi Key Lab Human Machine Interact & Intellige Nanning 530007 Peoples R China

With the growing use of drones, efficient detection algorithms are crucial, especially under adverse weather conditions. Most existing drone detection algorithms perform well only in clear weather, resulting in significant performance drops in foggy conditions. This study focuses on improving drone detection in foggy environments using the Mean Teacher framework for domain adaptation. The Mean Teacher framework's performance relies on the quality of the teacher model's pseudo-labels. To enhance the quality of the pseudo-labels from the teacher model, we introduce Foggy Drone Teacher (FDT), which includes three key components: (1) Adaptive Style and Context Augmentation to reduce domain shift and improve pseudo-label quality;(2) Simplified Domain Alignment with a novel adversarial strategy to boost domain adaptation;and (3) Progressive Domain Adaptation Training, a two-stage process that helps the teacher model produce more stable and accurate pseudo-labels. In addition, owing to the lack of publicly available data, we created Foggy Drone Dataset (FDD) to support this research. Extensive experiments show that our model achieves a 21.1-point increase in AP0.5 compared to the baseline and outperforms state-of-the-art models. This method significantly improves drone detection accuracy in foggy conditions.

关键词： drone detection object detection domain adaptation foggy condition

来源：评论

学校读者我要写书评

暂无评论

An intelligent retrievable object-tracking system with real-time edge inference capability

引用

IET IMAGE PROCESSING 2025年第1期19卷

作者： Li, Yujie Wang, Yifu Ma, Zihang Wang, Xinghe Tan, Benying Ding, Shuxue Guilin Univ Elect Technol Guangxi Coll Sch Artificial Intelligence Guilin Peoples R China Univ Key Lab AI Algorithm Engn Guilin Univ Elect Technol Guilin Peoples R China

An intelligent retrievable object-tracking system assists users in quickly and accurately locating lost objects. However, challenges such as real-time processing on edge devices, low image resolution, and small-object detection significantly impact the accuracy and efficiency of video-stream-based systems, especially in indoor home environments. To overcome these limitations, a novel real-time intelligent retrievable object-tracking system is designed. The system incorporates a retrievable object-tracking algorithm that combines DeepSORT and sliding window techniques to enhance tracking capabilities. Additionally, the YOLOv7-small-scale model is proposed for small-object detection, integrating a specialized detection layer and the convolutional batch normalization LeakyReLU spatial-depth convolution module to enhance feature capture for small objects. TensorRT and INT8 quantization are used for inference acceleration on edge devices, doubling the frames per second. Experiments on a Jetson Nano (4 GB) using YOLOv7-small-scale show an 8.9% improvement in recognition accuracy over YOLOv7-tiny in video stream processing. This advancement significantly boosts the system's performance in efficiently and accurately locating lost objects in indoor home settings.

关键词： computer vision image processing object detection object tracking

来源：评论

学校读者我要写书评

暂无评论

Automatic clustering of single-molecule break junction data through task-oriented representation learning

引用

RARE METALS 2025年第5期44卷 3244-3257页

作者： Zhao, Yi-Heng Pang, Shen-Wen Huang, Heng-Zhi Wu, Shao-Wen Sun, Shao-Hua Liu, Zhen-Bing Pan, Zhi-Chao Guilin Univ Elect Technol Sch Artificial Intelligence Guilin 541004 Peoples R China Guilin Univ Elect Technol Guangxi Coll & Univ Key Lab AI Algorithm Engn Guilin 541004 Peoples R China

Clustering is a pivotal data analysis method for deciphering the charge transport properties of single molecules in break junction experiments. However, given the high dimensionality and variability of the data, feature extraction remains a bottleneck in the development of efficient clustering methods. In this regard, extensive research over the past two decades has focused on feature engineering and dimensionality reduction in break junction conductance. However, extracting highly relevant features without expert knowledge remains an unresolved challenge. To address this issue, we propose a deep clustering method driven by task-oriented representation learning (CTRL) in which the clustering module serves as a guide for the representation learning (RepL) module. First, we determine an optimal autoencoder (AE) structure through a neural architecture search (NAS) to ensure efficient RepL;second, the RepL process is guided by a joint training strategy that combines AE reconstruction loss with the clustering objective. The results demonstrate that CTRL achieves excellent performance on both the generated and experimental data. Further inspection of the RepL step reveals that joint training robustly learns more compact features than the unconstrained AE or traditional dimensionality reduction methods, significantly reducing misclustering possibilities. Our method provides a general end-to-end automatic clustering solution for analyzing single-molecule break junction data. (sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic).(sic)(sic), (sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic), (sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic).(sic)(sic)(sic)(sic)(sic)(sic), (sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)

关键词： Single-molecule conductance Break junction Deep clustering Representation learning Neural architecture search

来源：评论

学校读者我要写书评

暂无评论

Lifted proximal operator machine-based deep nonlinear dictionary learning with multilayer regularization

引用

NEUROCOMPUTING 2025年 623卷

作者： Lin, Jie Tan, Benying Li, Yujie Qin, Yang Ding, Shuxue Guilin Univ Elect Technol Guangxi Coll Sch Artificial Intelligence Guilin 541004 Peoples R China Guilin Univ Elect Technol Univ Key Lab AI Algorithm Engn Guilin 541004 Peoples R China

Nonlinear dictionary learning (NLDL) can mine nonlinear information in data better than linear models. Mainstream NLDL methods are based on kernel methods;however, their large storage requirements pose computational challenges and are difficult to scalability. Moreover, the single-layer NLDL framework cannot capture deep features and requires the reversibility of nonlinear functions, leaving room for improving nonlinear signal representation. Inspired by deep dictionary learning, we design a generalized deep nonlinear dictionary learning model called DNLDL_Sp. This model features deep and nonlinear structures. In the proposed model, data are decomposed into nonlinear mappings of several dictionary layers, and the coefficients of the deeper layers are learned via dictionary learning in the preceding layers. We impose sparsity constraints on the coefficients of each layer of DNLDL_Sp to achieve a concise and accurate representation of the data, and this representation proves to be more effective than single-layer structures in capturing deep nonlinear representations. To optimize DNLDL_Sp and prevent issues such as loss of accuracy due to the inverse operation of the nonlinear function, we employ the lifted proximal operator machine (LPOM) technology to transform the nonlinear function into an equivalent proximal operator, which is then incorporated into the model as a penalty. This method converts the nonconvex optimization problem into a series of convex subproblems, allowing us to obtain the optimal dictionaries and coefficients through alternating updates. We demonstrate the effectiveness of our algorithm via image classification and denoising experiments involving different nonlinear functions and sparse constraints.

关键词： Nonlinear dictionary learning Sparse regularization Deep features Lifted proximal operator machine Image classification and denoising

来源：评论

学校读者我要写书评

暂无评论

Advancements in Large-Scale Image and Text Representation Learning: A Comprehensive Review and Outlook

引用

IEEE ACCESS 2025年 13卷 49922-49933页

作者： Qin, Yang Ding, Shuxue Xie, Huiming Guilin Univ Elect Technol Guangxi Coll & Univ Key Lab AI Algorithm Engn Sch Artificial Intelligence Guilin 541004 Peoples R China Guilin Univ Aerosp Technol Engn Comprehens Training Ctr Guilin 541004 Peoples R China

Large-scale image and text representation learning is critical in determining the performance of multimodal tasks involving images and text, such as visual question answering and image captioning. Most existing research on large-scale image and text representation learning relies on Transformer networks for pre-training, i.e., learning generic semantic representations from large-scale image-to-text pairs. These representations are then fine-tuned and transferred to downstream multimodal tasks. This paper first provides a brief analysis of the advantages of pre-training models. It then comprehensively summarizes the relevant research on large-scale image and text representation learning based on pre-training. The focus is on pre-training model architectures, pre-training tasks, and image-text datasets. Finally, we provide a summary and outlook of large-scale image and text representation learning.

关键词： Transformers Image coding Representation learning Feature extraction Encoding Vectors Data models Visualization Analytical models Semantics Image-and-text large-scale representation learning pre-training transformer self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Multidimensional perturbed consistency learning for semi-supervised medical image segmentation

引用

INTERNATIONAL JOURNAL OF IMAGING SYSTEMS AND TECHNOLOGY 2024年第3期34卷

作者： Yuan, Enze Zhao, Bin Qin, Xiao Ding, Shuxue Guilin Univ Elect Technol Sch Artificial Intelligence Guilin Guangxi Peoples R China Guangxi Coll & Univ Key Lab AI Algorithm Engn Guilin Guangxi Peoples R China Guilin Univ Elect Technol Sch Informat & Commun Guilin Guangxi Peoples R China Guangxi Acad Sci Nanning Guangxi Peoples R China Guilin Univ Elect Technol Guangxi Coll & Univ Key Lab AI Algorithm Engn Sch Artificial Intelligence Guilin 541004 Guangxi Peoples R China

In this article, we propose a novel multidimensional perturbed consistency network (MPCNet) for more accurate semi-supervised medical image segmentation. Specifically, we develop a multidimensional perturbation by considering the noise itself, the target object and the overall spatial architecture. This type of perturbation can disrupt the propagation process of abstract features over a wide range, enabling the model to learn the distribution of comprehensive data. In addition, we design a shared encoder to extract multi-scale features. After subjecting these features to multidimensional perturbation, a consistency constraint is applied between different results output by three independent decoders. This constraint aims to minimize the statistical differences between these results and effectively leverage unlabeled data. Experimental results on the public LA, Pancreas-CT and ACDC datasets demonstrate that our method outperforms recent SOTA semi-supervised learning methods in terms of various metrics. Our code is released publicly at .

关键词： consistency learning medical image segmentation multidimensional perturbation semi-supervised learning

来源：评论

学校读者我要写书评

暂无评论

A flare removal network for night vision perception: Resistant to the interference of complex light

引用

IET IMAGE PROCESSING 2024年第12期18卷 3192-3205页

作者： Liu, Yan Huang, Guan Qi, Wenting Li, Yujie Zhengzhou Univ Light Ind Sch Comp Sci & Technol Zhengzhou Peoples R China Guilin Univ Elect Technol Sch Artificial Intelligence Guangxi Coll & Univ Key Lab AI Algorithm Engn Guilin 541004 Peoples R China

The high-precision visual perception results are easily affected by the lens flare issue when the image sensor is facing to strong light. The existing flare removal methods have poor robustness when confronted with flare interference caused by complex nighttime lighting, which has to preserve natural light source information. A simulated dataset for the removal of night flares is created to solve the problem of collecting complete paired training data, and night flare removal network (NFR-Net) is proposed to remove the interference caused by various light disturbances at night. The light source extraction module is introduced to retain light source information realistically and effectively in night vision scenes. Extensive experimental results demonstrate that the proposed method is superior to the existing related methods in the various complex night vision scenes. The proposed NFR-Net can enhance visual perception of nighttime images significantly and improve the performance of night vision tasks. aiming at the problem of strong flare interference in the image captured when the sensor points at a strong light source. This paper proposes a night flare removal network trained on synthetic data which can remove a variety of flare interference while reasonably retaining the light source information. Compared with the existing methods, this method achieves better visual perception and makes up for the lack of methods in the field of night flare removal. image

关键词： computer vision convolutional neural nets image enhancement

来源：评论

学校读者我要写书评

暂无评论

CPG-based neural control of peristaltic planar locomotion in an earthworm-like robot: evaluation of nonlinear oscillators

引用

BIOINSPIRATION & BIOMIMETICS 2025年第2期20卷 026018-026018页

作者： Zhou, Qinyan Jia, Peisen Fang, Hongbin Fudan Univ Inst AI & Robot MOE Engn Res Ctr AI & Robot State Key Lab Med Neurobiol Shanghai Peoples R China

Earthworm-like robots have excellent locomotion capability in confined environments. Central pattern generator (CPG) based controllers utilize the dynamics of coupled nonlinear oscillators to spontaneously generate actuation signals for all segments, which offer significant merits over conventional locomotion control strategies. There are a number of oscillators that can be exploited for CPG control, while their performance in controlling peristaltic locomotion has not been systematically evaluated. To advance the state of the art, this study comprehensively evaluates the performance of four widely used nonlinear oscillators-Hopf, Van der Pol (VDP), Matsuoka, and Kuramoto-in controlling the planar locomotion of metameric earthworm-like robots. Specifically, the amplitude and phase characteristics of the continuous control signals used by the robot for achieving rectilinear, sidewinding, and arcuate locomotion are first summarized. On this basis, the sufficient parametric conditions for the four oscillator networks to generate the corresponding control signals are derived. Using a six-segment earthworm-like robot prototype as a platform, experiments confirm that the signals output by these oscillator networks can effectively control the robot to achieve the specified planar motion. Furthermore, the effects of the output signal waveforms of different oscillator networks on locomotion trajectories and performance metrics, as well as the effects of transient dynamics on the smoothness of gait transitions when the parameters are varied, are analyzed. The results demonstrate that their applicability varies in terms of locomotion efficiency, trajectory modulation, and smooth gait transitions. The Matsuoka oscillator lacks explicit rules for parameter modulation, the VDP oscillator is advantageous in enhancing the average speed and turning efficiency, and the Hopf and Kuramoto oscillators are advantageous in terms of smooth gait transition. These findings provide valuable i

关键词： bio-inspired robot central pattern generator nonlinear oscillator peristaltic locomotion gait transition

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：