检索结果-内蒙古大学图书馆

11th International Conference on Computing and Pattern Recognition, ICCPR 2022

作者： Weng, Kangming Du, Xia Chen, Kunze Wang, Dahan Zhu, Shunzhi School of Computer and Information Engineering Xiamen University of Technology Xiamen361024 China Fujian Key Laboratory of Pattern Recognition and Image Understanding Xiamen361024 China

ISBN: (纸本)9781450397056

The segmentation-based approach is an essential direction of scene text detection, and it can detect arbitrary or curved text, which has attracted the increasing attention of many researchers. However, extensive research has shown that the segmentation-based method will be disturbed by adjoining pixels and cannot effectively identify the text boundaries. To tackle this problem, we proposed a ResAsapp Conv based on the PSE algorithm. This convolution structure can provide different scale visual fields about the object and make it effectively recognize the boundary of texts. The method's effectiveness is validated on three benchmark datasets, CTW1500, Total-Text, and ICDAR2015 datasets. In particular, on the CTW1500 dataset, a dataset full of long curve text in all kinds of scenes, which is hard to distinguish, our network achieves an F-measure of 81.2%. © 2022 ACM.

关键词： Pixels

来源：评论

学校读者我要写书评

暂无评论

Attention-based Deep Learning Methods for Document Layout Analysis 22

Attention-based Deep Learning Methods for Document Layout An...

引用

8th International Conference on Computing and Artificial Intelligence, ICCAI 2022

作者： Sun, Hao-Yue Zhong, Ying Wang, Da-Han School of Computer and Information Engineering Xiamen University of Technology Xiamen361024 China Fujian Key Laboratory of Pattern Recognition and Image Understanding Xiamen361024 China

ISBN: (纸本)9781450396110

Deep learning has unique advantages in document layout analysis. In this paper, we propose to use the attention mechanism to improve the deep learning based methods for document layout analysis, and conducted comprehensive experiments to evaluate the effects of different models/modules/backbones. Firstly, the evaluated backbones include ResNet-50 and ResNet-101 for their high performance, and the popularly used Fast RCNN and Faster RCNN based methods. Secondly, for the loss function, the Focal loss function is also adopted to improve the performance of document layout analysis. Most importantly, we propose to integrate the attention mechanism into the deep learning-based methods to improve the performance. Specifically, when calculating the residual using this ResNet network, we use the attention to perform a dot product operation on the degraded accuracy to correct the residual error of the network calculation, which obtains a higher detection accuracy. We conducted experiments on the PubLayNet, DocBank and ICDAR-2013 datasets for evaluation. Experimental results show that, with the attention mechanism, the basic deep learning-based methods has been significantly improved in the task of document layout analysis, demonstrating the effectiveness of the proposed method. The focal loss that is designed to overcome the problem of imbalance number of training samples is also effective for document layout analysis. © 2022 Association for Computing Machinery. 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

Unsupervised Terahertz image Restoration Based on CycleGan 3

Unsupervised Terahertz Image Restoration Based on CycleGan

引用

3rd International Conference on Defence Technology, ICDT 2022

作者： Su, Zhipeng Zhang, Yixiong Qi, Feng Shi, Jianghong School of Informatics Xiamen University National Demonstrative Software School Xiamen361005 China Shenyang Institute of Automation Chinese Academy of Sciences Shenyang110016 China Institutes for Robotics and Intelligent Manufacturing Chinese Academy of Sciences Shenyang110016 China Key Laboratory of Opto-Electronic Information Process Shenyang110016 China Key Laboratory of Image Understanding and Computer Vision Shenyang110016 China

Terahertz (THz) is considered as one of the key technologies for sixth generation communications, military, medical imaging and industrial inspection. THz images are susceptible to degradation due to system noise and point spread functions during transmission. The existing deep learning methods use ground truth and input images for supervised training that can recover THz images very well. But it's difficult to obtain labeled THz data in practical application. In this paper, we propose an attentional adversarial cycle generation network for THz image restoration (CycleTHz) based on CycleGan to address this problem. The CycleTHz generates clean images firstly by an attention-guided generation network and then discriminates the quality of the generators by an attention discriminator. In addition, RGB color loss is used for image channels for constraint. To the best of our knowledge, this is the first THz dataset to be trained using an unsupervised approach. Extensive experiments show that the proposed method improves the PSNR and SSIM by 43.4% and 101.7% compared with CycleGan, which is a benchmark method for the unsupervised development in THz image restoration. The code is available at https://***/hellogry/UnsupervisedCycleTHz © Published under licence by IOP Publishing Ltd.

关键词： image reconstruction

来源：评论

学校读者我要写书评

暂无评论

image Segmentation of Endoscopic Cataract Based on Semantic Segmentation 6

Image Segmentation of Endoscopic Cataract Based on Semantic ...

引用

2022 6th International Conference on Machine vision and Information Technology, CMVIT 2022

作者： Chen, Jingshi Ye, Haili Yin, Huayi Wang, Da-Han Li, Jianmin School of Computer and Information Engineering Xiamen University of Technology Xiamen361000 China Fujian Key Laboratory of Pattern Recognition and Image Understanding Xiamen361000 China

Medical imaging technology has become one of the indispensable computer-assisted intervention methods in clinical disease diagnosis and treatment, including identifying and locating lesion areas, detecting and segmenting tissue and organ lesions in different modalities. The wide application of medical image analysis in clinical examination and medical auxiliary diagnosis has effectively improved the diagnostic efficiency of doctors. In this paper, we propose a novel and effective model to achieve semantic segmentation of cataract datasets. This model uses the self-supervised method BYOL for parameter pre-training, which improves the model's ability to extract image consistency features. In addition, we have added a lightweight Coordinate attention mechanism to the backbone network to enable the model to independently learn the correlation between the channel and the space and enhance the ability of network feature expression. Experiments are conducted on the cataract endoscopy fine-grained segmentation data set, showing the effectiveness of the proposed method for segment the organs and surgical instruments in the cataract surgical microscope image, which demonstrates the accuracy and robustness of the proposed method. © Published under licence by IOP Publishing Ltd.

关键词： Endoscopy

来源：评论

学校读者我要写书评

暂无评论

Global Convergence Guarantees of (A)GIST for a Family of Nonconvex Sparse Learning Problems

引用

IEEE Transactions on Cybernetics 2022年第5期52卷 3276-3288页

作者： Zhang, Hengmin Qian, Feng Shang, Fanhua Du, Wenli Qian, Jianjun Yang, Jian East China University of Science and Technology Key Laboratory of Advanced Control and Optimization for Chemical Processes of Ministry of Education School of Information Science and Engineering Shanghai200237 China Shanghai Institute of Intelligent Science and Technology Tongji University Shanghai200092 China Xidian University Key Laboratory of Intelligent Perception and Image Understanding of Ministry of Education School of Artificial Intelligence Xi'an710071 China Peng Cheng Laboratory Shenzhen518066 China Nanjing University of Science and Technology Pca Laboratory Key Lab. of Intelligent Percept. and Syst. for High-Dimensional Information of Ministry of Education Nanjing210094 China Nanjing University of Science and Technology Jiangsu Key Laboratory of Image and Video Understanding for Social Security School of Computer Science and Engineering Nanjing210094 China

In recent years, most of the studies have shown that the generalized iterated shrinkage thresholdings (GISTs) have become the commonly used first-order optimization algorithms in sparse learning problems. The nonconvex relaxations of the \ell _{0} -norm usually achieve better performance than the convex case (e.g., \ell _{1} -norm) since the former can achieve a nearly unbiased solver. To increase the calculation efficiency, this work further provides an accelerated GIST version, that is, AGIST, through the extrapolation-based acceleration technique, which can contribute to reduce the number of iterations when solving a family of nonconvex sparse learning problems. Besides, we present the algorithmic analysis, including both local and global convergence guarantees, as well as other intermediate results for the GIST and AGIST, denoted as (A)GIST, by virtue of the Kurdyka-Lojasiewica (KL) property and some milder assumptions. Numerical experiments on both synthetic data and real-world databases can demonstrate that the convergence results of objective function accord to the theoretical properties and nonconvex sparse learning methods can achieve superior performance over some convex ones. © 2013 IEEE.

关键词： Optimization

来源：评论

学校读者我要写书评

暂无评论

A novel clinical artificial intelligence model for disease detection via retinal imaging

引用

The Innovation 2024年第2期5卷 11-12页

作者： Yidian Fu Liang Ma Sheng Wan Shengfang Ge Zhi Yang Department of Ophthalmology Ninth People’s HospitalShanghai JiaoTong University School of MedicineShanghai 200011China Shanghai Key Laboratory of Orbital Diseases and Ocular Oncology Shanghai 200011China PCA Lab Key Lab of Intelligent Perception and Systems for High-Dimensional Information of Ministry of EducationNanjing 210000China Jiangsu Key Lab of Image and Video Understanding for Social Security School of Computer Science and EngineeringNanjing University of Science and TechnologyNanjing 210000China

Artificial intelligence(AI)is an emerging field in which computerized systems are used to carry out complex tasks in place of *** AI algorithms have been developed for disease diagnosis and prediction and treatment recommendation across various clinical data types,e.g.,chest X-rays,electrocardiograms,and other radiological images.1 In ophthalmology,particularly,great progress has been made in AI systems over the past *** fundus photography(CFP)and optical coherence tomography(OCT),which are readily available in routine clinical practice,are both mainstream and useful retinal imaging modalities in *** September 2023,the 2023 Lasker-Debakey Clinical Medical Research Award was awarded to three scientists for their work on OCT for accurate retinal disease detection.

关键词： clinical routine diagnosis

来源：评论

学校读者我要写书评

暂无评论

Fast Candidate Region Extraction for SAR Ship Target 37

Fast Candidate Region Extraction for SAR Ship Target

引用

37th Youth Academic Annual Conference of Chinese Association of Automation, YAC 2022

作者： Zhang, Panpan Luo, Haibo Xu, Zheng He, Miao Shenyang Institute of Automation Shenyang110016 China Institutes for Robotics and Intelligent Manufacturing Shenyang110016 China University of Chinese Academy of Sciences Beijing100049 China Key Laboratory of Opto-Electronic Information Processing Shenyang110016 China The Key Lab of Image Understanding and Computer Vision Shenyang110016 China

ISBN: (纸本)9781665465366

At present, deep learning technology is widely used in ship target detection in synthetic aperture radar (SAR) images. However, high-resolution remote sensing SAR images cover a larger area and have larger image sizes. To be able to use the deep learning model for training and testing, the image needs to be cropped to the appropriate size. In a high-resolution SAR ship image, the ship target usually takes only a small part of the whole image. As a result, only part of the cropped image contains the target, and most of the rest are background regions. This will cause a lot of computational redundancy in the model inference stage. To solve this problem, a fast candidate region extraction algorithm is proposed in this paper for ship target extraction. The algorithm consists of three parts: firstly, extract the saliency map of the SAR ship image, secondly, perform median filtering on the segmented image, and thirdly, extract candidate regions. The superiority of the algorithm was demonstrated by experiments on the AIR-SARShip-1.0 dataset. © 2022 IEEE.

关键词： Synthetic aperture radar

来源：评论

学校读者我要写书评

暂无评论

A Dynamic Spatio-temporal Network with Self-attention for Multi-station Passenger Flow Prediction

A Dynamic Spatio-temporal Network with Self-attention for Mu...

引用

International Conference on Awareness Science and Technology (iCAST)

作者： Fengzhi Wang Qinzhi Lv Lijuan Liu College of Computer and Information Engineering Xiamen University of Technology Xiamen China Fujian Key Laboratory of Pattern Recognition and Image Understanding

Passenger flow prediction is vitally significant for intelligent transportation systems (ITS). Most of the studies typically focus on the passenger flow prediction for an individual station, and only capture the temporal features without considering any spatial features. Constructing a passenger flow prediction model for multiple stations, or even a whole network, is more valuable for practical applications. Therefore, we develop a dynamic spatio-temporal network (DSTNet) with a self-attention (SA) mechanism for multi-station passenger flow prediction. A dynamic graph convolutional network (DGCN) is applied for the spatial feature extraction, and gated recurrent unit (GRU) is combined to learn the temporal features. SA is applied to further assign the weights for the extracted spatio-temporal features. The Experiment has been conducted on the passenger flow in the Xiamen bus rapid transit (BRT). The results demonstrate that the proposed DSTNet with SA (SA-DSTNet) outperforms the baselines in the multi-station passenger flow prediction task.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Federated Multitask Learning for HyperFace

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2022年第5期3卷 788-797页

作者： Zhang, Chen Bai, Hang Zhang, Yonggang Niu, Xinyi Yu, Bin Gao, Yuan Xie, Yu Xidian University School of Computer Science and Technology Xi'an710071 China Shaanxi Province Science and Technology Resources Coordination Center Xi'an710071 China Xidian University Key Laboratory of Intelligent Perception and Image Understanding Ministry of Education Xi'an710071 China Shanxi University Key Laboratory of Computational Intelligence and Chinese Information Processing Ministry of Education Taiyuan030006 China

Multitask learning (MTL) is a promising field in machine learning owing to its capability to improve the generalization performance of all tasks by sharing knowledge among the related tasks. MTL has attracted a large amount of attention in the multitask-related community. In recent years, with the rapid development of distributed machine learning, MTL in a distributed environment becomes a hot research topic. Although MTL can reap great fruit in a distributed environment, the sensitive information, such as photos and voice recordings of the owner, in distributed data may be leaked during training. Consequently, in the distributed settings, one prominent challenge of MTL is to prevent the disclosure of sensitive private data. In this article, we propose a novel approach, federated multitask learning (FMTL), which applies federated learning to MTL in a distributed environment in order to protect the MTL model from being leaked and to better optimize the utility of the MTL model. In FMTL, all the participants independently train their local models on their own dataset in parallel and only transmit their model updates to the central server for aggregation at every epoch. As a result, their learning accuracy is enhanced beyond what is achieved only on their own training data. In this manner, it can achieve the best tradeoff between the utility and privacy: participants not only preserve the privacy of their own data, but also benefit from the models of other participants. The experimental results on annotated facial landmarks in the wild (AFLW) and annotated faces in the wild (AFW) benchmark datasets verify the effectiveness of our framework. © 2020 IEEE.

关键词： Linearization

来源：评论

学校读者我要写书评

暂无评论

A Survey of Person Re-identification Based on Deep Learning 10

A Survey of Person Re-identification Based on Deep Learning

引用

10th International Conference on Computing and Pattern Recognition, ICCPR 2021

作者： Tian, Zimin Chen, Si Wang, Da-Han Lu, Junwen School of Computer and Information Engineering Xiamen University of Technology China Fujian Key Laboratory of Pattern Recognition and Image Understanding China

ISBN: (纸本)9781450390439

Person re-identification (Re-ID) has been a popular research topic in computer vision in recent years, and it has important application value in numerous fields, such as intelligent security. The person Re-ID task is to identify whether the pedestrians appearing under different cameras are the same person. The traditional person Re-ID methods mainly rely on the characteristics of manual design, and it has difficulty in solving the problems of person occlusion, posture change, and illumination variation. With the wide application of deep learning, the person Re-ID based on deep learning has brought new ideas for solving these problems, and has been widely concerned by scholars. This paper summarizes and analyzes the latest research trends of person Re-ID based on deep learning. In our work, the recent research works of person Re-ID are coarsely categorized into the supervised learning methods and the unsupervised learning methods according to whether the pedestrian images in the training set have real labels. We then describe the representative datasets used in the person Re-ID task. Finally, we conclude and discuss the future directions of the person Re-ID based on deep learning. © 2021 ACM.

关键词： Unsupervised learning

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：