检索结果-内蒙古大学图书馆

1st International Conference on pattern Analysis and Machine Intelligence, ICPAMI 2024

作者： Zhu, Yu Li, Wangtao Ling, Xufeng Yang, Jie Shanghai China School of AI Shanghai Normal University Tianhua College Shanghai China Institute of Image Processing and Pattern Recognition Shanghai Jiao Tong University Shanghai China

ISBN: (纸本)9789819633487

As society increasingly focuses on health issues, the importance of the role of family doctors becomes more apparent. The question of how to use cutting-edge technology to improve the accessibility and efficiency of medical services is a critical topic. Generative large language models, with their excellent capabilities in natural language processing, offer new possibilities for the implementation of intelligent family doctor systems. This study focuses on the application of generative large language models in family doctor systems, selecting the ChatGLM2-6B model as the base model for this research. Through a custom-written data collection program, a dataset containing 170,000 high-quality medical question-and-answer entries was constructed. The LoRA method was used to tune the base model, and the performance of the original and tuned models in medical question-and-answer tasks was examined. The practical results of this study show that the generatively tuned large language model has achieved a 70% improvement in ROUGE scores compared to the untuned model, validating the effectiveness of large language model tuning in the family doctor field. Furthermore, this study has open-sourced the collected dataset and program, providing a technical reference for deploying efficient and accurate question-and-answer systems in similar vertical fields in the future. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Natural language processing systems

来源：评论

学校读者我要写书评

暂无评论

OCR4HSV: A Multi-task Learning Approach for Handwritten Signature Verification 27th

OCR4HSV: A Multi-task Learning Approach for Handwritten Sig...

引用

27th International Conference on pattern recognition, ICPR 2024

作者： Lin, Chao-Qun Wang, Da-Han Su, Yan-Fei Ge, De-Wu Zhang, Xu-Yao School of Computer and Information Engineering Xiamen University of Technology Xiamen361024 China Fujian Key Laboratory of Pattern Recognition and Image Understanding Xiamen361024 China Xiamen KEYTOP Communication Technology Co. Xiamen361024 China State Key Laboratory of Multimodal Artificial Intelligence Systems Institute of Automation of Chinese Academy of Sciences Beijing100190 China

ISBN: (纸本)9783031781186

Handwritten signature verification (HSV) models are notably recognized for their ability to discern whether a signature is forged in an offline document. Recently, HSV technology has made significant development. However, these methods have primarily focused on the shape features of the text image and overlooked the textual feature information inherent in the text itself, which makes the HSV model overfit. In this paper, we propose a novel network model named OCR4HSV to solve the above shortcomings. The proposed OCR4HSV first attempts to combine OCR and HSV to learn textual features in a multi-task learning manner. The model employs a dual-parameter approach, combining shared parameters and independent parameters. Specifically, Within the shared parameters, the Laplace attention module (LAM) is incorporated for edge information extraction. For independent parameters, CN-Mamba is utilized for sequence feature extraction in OCR, and the multi-scale global fusion block (MGFB) is designed to enhance the distinction between reference and test sample pairs. Leveraging OCR-related information and these architectural enhancements can fully mine the inherent textual feature information and significantly mitigate overfitting in the HSV task, thereby boosting verification accuracy. Our model has achieved state-of-the-art performance on the ChiSig and HanSig datasets. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Multi-task learning

来源：评论

学校读者我要写书评

暂无评论

Reflecting topology consistency and abnormality via learnable attentions for airway labeling

引用

International Journal of Computer Assisted Radiology and Surgery 2025年 1-9页

作者： Li, Chenyu Zhang, Minghui Zhang, Chuyan Gu, Yun Institute of Medical Robotics Shanghai Jiao Tong University Shanghai China Institute of Image Processing and Pattern Recognition Shanghai Jiao Tong University Shanghai China Shanghai Key Laboratory of Flexible Medical Robotics Tongren Hospital Shanghai Jiao Tong University Shanghai China

Purpose: Accurate airway anatomical labeling is crucial for clinicians to identify and navigate complex bronchial structures during bronchoscopy. Automatic airway labeling is challenging due to significant anatomical variations. Previous methods are prone to generate inconsistent predictions, hindering preoperative planning and intraoperative navigation. This paper aims to enhance topological consistency and improve the detection of abnormal airway branches. Methods: We propose a transformer-based framework incorporating two modules: the soft subtree consistency (SSC) and the abnormal branch saliency (ABS). The SSC module constructs a soft subtree to capture clinically relevant topological relationships, allowing for flexible feature aggregation within and across subtrees. The ABS module facilitates interaction between node features and prototypes to distinguish abnormal branches, preventing the erroneous features aggregation between normal and abnormal nodes. Results: Evaluated on a challenging dataset characterized by severe airway deformities, our method achieves superior performance compared to state-of-the-art approaches. Specifically, it attains an 83.7% subsegmental accuracy, along with a 3.1% increase in segmental subtree consistency, a 45.2% increase in abnormal branch recall. Notably, the method demonstrates robust performance in cases with airway deformities, ensuring consistent and accurate labeling. Conclusion: The enhanced topological consistency and robust identification of abnormal branches provided by our method offer an accurate and robust solution for airway labeling, with potential to improve the precision and safety of bronchoscopy procedures. © CARS 2025.

关键词： Airway anatomical labeling Anomaly detection Structural prior Transformer

来源：评论

学校读者我要写书评

暂无评论

CATransUnetLBP: Accurate Prediction of Protein-Ligand Binding Pockets Using a Hybrid Network

IEEE Transactions on Computational Biology and Bioinformatic...

引用

IEEE Transactions on Computational Biology and Bioinformatics 2025年第1期22卷 355-367页

作者： Cheng Cai Zhaohong Deng Andong Li Yun Zuo Haoran Chen Zhisheng Wei Lei Wang Xiaoyong Pan Hong-Bin Shen Dong-Jun Yu School of Artificial Intelligence and Computer Science Jiangnan University Wuxi China Shanghai Key Lab of Intell. Info. Processing School of CS Fudan University Shanghai China School of Biotechnology Jiangnan University Wuxi China Institute of Image Processing and Pattern Recognition Shanghai Jiao Tong University Shanghai China School of Computer Science and Engineering Nanjing University of Science and Technology Nanjing China

The development of intelligent methods capable of predicting protein-ligand binding sites has become a popular research field. Recently, deep learning based methods have been proposed as a promising solution for this task. However, some limitations still exist. For example, the network structure is not optimized for predicting protein binding pockets, which limits the model's capabilities. To address the aforementioned challenges, a novel method called CATransUnetLPB is proposed, in which a new network structure named CATransUnet is designed. The proposed CATransUnet combines CNN and Transformer models to accurately segment binding pocket regions from protein 3D structures. It outperforms existing representative methods on three test sets, demonstrating the effectiveness of optimizing the deep network model for detecting protein ligand binding pockets. Furthermore, we conduct thorough analysis on applying data augmentation to protein data structure and confirm that such technique can enhance the model's generalization ability, thereby ensuring good performance on new protein structures. Moreover, experiments show that the predicted binding pockets from our model can complement the results obtained from other methods. This suggests that integrating our method with existing approaches could further improve the prediction of protein-ligand binding pockets.

关键词： Proteins Three-dimensional displays Probes Vectors image segmentation Deep learning Atoms Surface treatment Prediction algorithms Training

来源：评论

学校读者我要写书评

暂无评论

2M3DF: Advancing 3D Industrial Defect Detection with Multi Perspective Multimodal Fusion Network

引用

IEEE Transactions on Circuits and Systems for Video technology 2025年

作者： Asad, Mujtaba Azeem, Waqar Jiang, He Mustafa, Hafiz Tayyab Yang, Jie Liu, Wei Shanghai Jiao Tong University Institute of Image Processing and Pattern Recognition Department of Automation Shanghai200240 China Lahore Garrison University Department of Software Engineering Lahore54000 Pakistan China University of Mining and Technology School of Information and Control Engineering Jiangsu Xuzhou221116 China Zhejiang Normal University School of Computer Science and Technology Jinhua321004 China

In the context of Industrial Anomaly Detection (IAD), ensuring the quality of manufactured products is critical. Traditional 2D based methods often fail to capture anomalies present in complex 3D shapes. For effective anomaly detection in 3D shapes, it is essential to incorporate global semantic context, local geometric structure, and color information of the object. To fully leverage these features, we propose a network named 2M3DF, that leverages knowledge from multi-view RGB images and corresponding point cloud information for enhanced anomaly detection performance. Our model initially employs pre-trained feature extractors that generate local features from multi-view RGB images and corresponding point clouds. The novel inter-modality feature representation and fusion module first adapts these inter-modality features and then effectively aligns and aggregates these multimodality features on a pixel-to-point basis. To learn the normality from point-wise fused multimodal features, we fit a multivariate Gaussian distribution to model the normal feature distribution. Comprehensive experimental evaluations using the MVTec3D-AD and Eyecandies dataset validate the effectiveness of our propose model and demonstrate significant improvements in comparison to existing state-of-the-art methods. Our model achieves a 96.6% mean I-AUROC while delivering real-time results. © 1991-2012 IEEE.

关键词： Normal distribution

来源：评论

学校读者我要写书评

暂无评论

Enhance the old representations’ adaptability dynamically for exemplar-free continual learning

引用

Neurocomputing 2025年 639卷

作者： Li, Kunchi Ding, Chaoyue Wan, Jun Yu, Shan School of Computer and Information Engineering of Xiamen University of Technology Fujian Xiamen361024 China Fujian Key Laboratory of Pattern Recognition and Image Understanding of Xiamen University of Technology Fujian Xiamen361024 China Institute of Automation of Chinese Academy of Sciences Beijing Beijing100190 China

In continual learning, new data often falls outside the distribution of previous data. Since the old model is trained solely on past tasks and has not encountered the new data, its learned representations lack the adaptability needed to accommodate these new inputs. This mismatch creates a conflict between the learned representations and the concepts of new classes, resulting in a significant representation shift during model updates and exacerbating the issue of catastrophic forgetting. In this work, we presents a two-stage training framework to enhance the learned representations’ adaptability dynamically for exemplar-free continual learning (EFCL), dubbed ERA-EFCL. The first stage involves synthesizing data for the old classes using DeepInversion and combining them with the new data to train an expanded module and a feature fusion network (FFN). The expanded module identifies critical features overlooked by the old model, while the FFN integrates the old representations with these newly discovered features. By incorporating additional information that distinguishes between the old and new classes, the FFN generates fused representations enriched with more transferable features for the old classes. This helps align the old representations with the new concepts, enhancing their adaptability and reducing the representation shift. Next, we introduce global task-wise knowledge distillation to ensure balanced knowledge transfer alongside other losses, thereby improving representation learning during the training of the new model. Furthermore, the new classifier is refined using a class-balanced training strategy. Extensive experiments demonstrate that ERA-EFCL achieves favorable results across four benchmark datasets. The code is available at: "https://***/CSTiger77/ExemplarFreeCL". © 2025

关键词： Continual learning Enhance representation's adaptability Knowledge distillation Representation alignment

来源：评论

学校读者我要写书评

暂无评论

Consistency-Guided Adaptive Alternating Training for Semi-Supervised Salient Object Detection

引用

IEEE Transactions on Circuits and Systems for Video technology 2025年

作者： Chen, Liyuan Liu, Wei Wang, Hua Jeon, Sang-Woon Jiang, Yunliang Zheng, Zhonglong Zhejiang Normal University School of Computer Science and Technology Jinhua321004 China Shanghai Jiao Tong University Institute of Image Processing and Pattern Recognition Department of Automation Shanghai200240 China Victoria University Institute for Sustainable Industries and Liveable Cities College of Engineering and Science MelbourneVIC8001 Australia Hanyang University Department of Electrical and Electronic Engineering Ansan Korea Republic of

This paper presents a novel approach that leverages two models to integrate features from numerous unlabeled images, addressing the challenge of semi-supervised salient object detection (SSOD). Unlike conventional methods that rely on selecting high-quality pseudo labels, our method identifies the model that produces consistent predictions for original images and their color transformation versions from two models to infer reliable pseudo labels for all unlabeled images, improving the diversity of the training set. Specifically, we propose adaptive selection indicators to quantify prediction differences and guide the updates of the two models using the unlabeled set alternatively. Initially, two models used in our framework are trained on the labeled set. Once the adaptive selection indicator conditions are satisfied, one model is designated as the proxy, generating pseudo labels, while the other serves as the saliency model, which is further trained using these pseudo labels. Subsequently, the updated saliency model optimizes the proxy model's parameters according to another adaptive selection indicator. Experimental results and ablation studies on six benchmark salient object detection datasets confirm the effectiveness and robustness of our method. Our approach achieves performance comparable to recent fully supervised methods while using only one eighth of the labeled data, demonstrating its potential for efficient and scalable SSOD. © 2025 IEEE.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

A novel enhancement method for low illumination images based on microarray camera

引用

Applied Mathematics(A Journal of Chinese Universities) 2017年第3期32卷 313-322页

作者： ZOU Jian-cheng ZHENG Wen-qi YANG Zhi-hui Institute of Image Processing and Pattern Recognition North China University of Technology Beijing 100144 China.

It is difficult but important to get clear information from the low illumination images. In recent years the research of the low illumination image enhancement has become a hot topic in image processing and computer vision. The Retinex algorithm is one of the most popular methods in the field and uniform illumination is necessary to enhance low illumination image quality by using this algorithm. However, for the different areas of an image with contrast brightness differences, the illumination image is not smooth and causes halo artifacts so that it cannot retain the detail information of the original images. To solve the problem, we generalize the multi-scale Retinex algorithm and propose a new enhancement method for the low illumination images based on the microarray camera. The proposed method can well make up for the deficiency of imbalanced illumination and significantly inhibit the halo artifacts as well. Experimental results show that the proposed method can get better image enhancement effect compared to the multi-scale Retinex algorithm of a single image enhancement. Advantages of the method also include that it can significantly inhibit the halo artifacts and thus retain the details of the original images, it can improve the brightness and contrast of the image as well. The newly developed method in this paper has application potential to the images captured by pad and cell phone in the low illumination environment.

关键词： low illumination microarray camera multi-scale Retinex image sharpening

来源：评论

学校读者我要写书评

暂无评论

Robust skeleton extraction of gray images based on level set approach

引用

Journal of Multimedia 2013年第1期8卷 24-31页

作者： Yang, Zhihui Guo, Fangfang Dong, Ping Institute of Image Processing and Pattern Recognition North China University of Technology Beijing China

The skeleton of an image object is a simplified representation, which is of great significance for the image recognition and matching. To obtain a smooth and accurate skeleton of a specified object in the gray image, this paper provides a unified framework by combining the level set idea with the gradient module method. The intrinsic procedure involves three steps. First, an energy function is given by virtue of the statistical intensity disparity between the sample points and the object, and then a novel segmentation model is proposed to extract any specified objects in the gray image by the variational method. Moreover, the segmentation model is further improved to be suited to the texture image. Second, in order to find an accurate position of the skeleton, a searching algorithm for the endpoints of the skeleton is shown based on the segmentation result. Finally, an improved skeleton extraction algorithm is given via the shortest path connection approach. Some examples show the robustness and insensitivity of the presented algorithm to the perturbation and noise, respectively. © 2013 ACADEMY PUBLISHER.

关键词： image recognition

来源：评论

学校读者我要写书评

暂无评论

LLM-PS: Empowering Large Language Models for Time Series Forecasting with Temporal patterns and Semantics

arXiv

引用

arXiv 2025年

作者： Tang, Jialiang Chen, Shuo Gong, Chen Zhang, Jing Tao, Dacheng School of Computer Science and Engineering Nanjing University of Science and Technology China College of Computing and Data Science Nanyang Technological University Singapore School of Intellgence Science and Technology Nanjing University China Department of Automation Institute of Image ProceProceedingssing and Pattern Recognition Shanghai Jiao Tong University China School of Computer Science Wuhan University China

Time Series Forecasting (TSF) is critical in many real-world domains like financial planning and health monitoring. Recent studies have revealed that Large Language Models (LLMs), with their powerful in-contextual modeling capabilities, hold significant potential for TSF. However, existing LLM-based methods usually perform suboptimally because they neglect the inherent characteristics of time series data. Unlike the textual data used in LLM pre-training, the time series data is semantically sparse and comprises distinctive temporal patterns. To address this problem, we propose LLM-PS to empower the LLM for TSF by learning the fundamental patterns and meaningful Semantics from time series data. Our LLM-PS incorporates a new multi-scale convolutional neural network adept at capturing both short-term fluctuations and long-term trends within the time series. Meanwhile, we introduce a time-to-text module for extracting valuable semantics across continuous time intervals rather than isolated time points. By integrating these patterns and semantics, LLM-PS effectively models temporal dependencies, enabling a deep comprehension of time series and delivering accurate forecasts. Intensive experimental results demonstrate that LLM-PS achieves state-of-the-art performance in both short- and long-term forecasting tasks, as well as in few- and zero-shot settings. © 2025, CC BY.

关键词： Spatio-temporal data

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：