检索结果-内蒙古大学图书馆

Classification of mammography based on semi-supervised learning 7

学校读者我要写书评

暂无评论

Classification of mammography based on semi-supervised learn...

7th IEEE International Conference on Progress in Informatics and Computing, PIC 2020

作者： Sun, Lilei Wen, Jie Wang, Junqian Zhao, Yong Xu, Yong Guizhou University College of Computer Science and Technology Guiyang China Harbin Institute of Technology Shenzhen Shenzhen Key Laboratory of Visual Object Detection and Recognition Shenzhen China Shenzhen Graduate School of Peking University School of Electronic and Computer Engineering Shenzhen China

ISBN: (纸本)9781728170862

In recent years, deep learning, especially the convo-lutional neural network (CNN), has been widely used in the field of molybdenum target image classification. Generally speaking, the CNN based methods train the network with a huge number of samples in a supervised way. However, it is too expensive to collect such a large-scale labeled mammogram dataset for training. In this paper, we propose a semi-supervised mammography classification approach to address this issue. Compared with the existing method, our method can simultaneously use the labeled and unlabeled samples to train the mammogram classification network. Specifically, for the breast mass images represented by Mediolateral Oblique (MLO) and Craniocaudal (CC) views, a new classification network is designed to extract features from both views simultaneously. The compact bilinear pooling (CBP) is introduced to extract the fine-grained features of breast mass images. Moreover, considering that the samples near the decision boundary are easy to be misclassified and those misclassified samples are harmful to the model training, we develop a K-Means based strategy, which first clusters the unlabeled samples and then removes those samples far away from the clustering center during model training. The method proposed in this paper is compared with the mainstream deep learning classification methods and semi-supervised methods on the Digital Database for Screening Mammography (DDSM) public mammogram database. The experimental results show that the method proposed is superior to most of the compared methods in terms of processing time and classification performance. © 2020 IEEE.

关键词： Convolutional neural networks

Dual-stream reciprocal disentanglement learning for domain adaptation person re-identification

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Li, Huafeng Xu, Kaixiong Li, Jinxing Lu, Guangming Xu, Yong Yu, Zhengtao Zhang, David the Kunming University of Science and Technology Kunming China Harbin Institute of Technology Shenzhen China Shenzhen Key Laboratory of Visual Object Detection and Recognition Shenzhen China also with the Shenzhen Research Institute of Big Data Shenzhen China

—Since human-labeled samples are free for the target set, unsupervised person re-identification (Re-ID) has attracted much attention in recent years, by additionally exploiting the source set. However, due to the differences on camera styles, illumination and backgrounds, there exists a large gap between source domain and target domain, introducing a great challenge on cross-domain matching. To tackle this problem, in this paper we propose a novel method named Dual-stream Reciprocal Disentanglement Learning (DRDL), which is quite efficient in learning domain-invariant features. In DRDL, two encoders are first constructed for id-related and id-unrelated feature extractions, which are respectively measured by their associated classifiers. Furthermore, followed by an adversarial learning strategy, both streams reciprocally and positively effect each other, so that the id-related features and id-unrelated features are completely disentangled from a given image, allowing the encoder to be powerful enough to obtain the discriminative but domain-invariant features. In contrast to existing approaches, our proposed method is free from image generation, which not only reduces the computational complexity remarkably, but also removes redundant information from id-related features. Extensive experiments substantiate the superiority of our proposed method compared with the state-of-the-arts. The source code has been released in https://***/lhf12278/DRDL. Copyright © 2021, The Authors. All rights reserved.

关键词： Signal encoding

A Survey on Incomplete Multi-view Clustering

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Wen, Jie Zhang, Zheng Fei, Lunke Zhang, Bob Xu, Yong Zhang, Zhao Li, Jinxing Shenzhen Key Laboratory of Visual Object Detection and Recognition Harbin Institute of Technology Shenzhen Shenzhen518055 China School of Computer Science and Technology Guangdong University of Technology Guangzhou China Department of Computer and Information Science University of Macau Taipa China School of Computer Science School of Artificial Intelligence Hefei University of Technology Hefei230000 China

Conventional multi-view clustering seeks to partition data into respective groups based on the assumption that all views are fully observed. However, in practical applications, such as disease diagnosis, multimedia analysis, and recommendation system, it is common to observe that not all views of samples are available in many cases, which leads to the failure of the conventional multi-view clustering methods. Clustering on such incomplete multi-view data is referred to as incomplete multi-view clustering. In view of the promising application prospects, the research of incomplete multi-view clustering has noticeable advances in recent years. However, there is no survey to summarize the current progresses and point out the future research directions. To this end, we review the recent studies of incomplete multi-view clustering. Importantly, we provide some frameworks to unify the corresponding incomplete multi-view clustering methods, and make an in-depth comparative analysis for some representative methods from theoretical and experimental perspectives. Finally, some open problems in the incomplete multi-view clustering field are offered for researchers. The related codes are released at https://***/DarrenZZhang/Survey_IMC. Copyright © 2022, The Authors. All rights reserved.

关键词： Surveys

Graph embedded incomplete multi-view clustering method with proximity relation estimation

学校读者我要写书评

暂无评论

Graph embedded incomplete multi-view clustering method with ...

3rd Asian Conference on Artificial Intelligence Technology, ACAIT 2019

作者： Chen, Runze Wen, Jie Chen, Xiaoyue Xu, Yong Bio-Computing Research Centre Harbin Institute of Technology Shenzhen Shenzhen China Shenzhen Key Laboratory of Visual Object Detection and Recognition Shenzhen China School of Electric Power South China University of Technology Guangzhou China

More and more importance has been attached to multi-view clustering due to its outstanding performance against single-view clustering. Existing studies are on the basis of the assumption that every sample exists in all views. However, in reality, this is often not the case and the real data are always with incomplete views, which lead to the failure of the conventional methods. To address the issue, the authors present a novel method, referred to as incomplete multi-view spectral clustering with proximity relation estimation (IMSC_PRE), which exploits the proximity relations of samples in different views to complete their affinity graphs. Different from existing methods, IMSC_PRE fully excavates the manifold structures of all views to estimate the proximity relations of missing samples in each view, which presents a brand-new strategy to exploit the proximity information lying in the whole samples with good interpretation. Experimental results on four datasets demonstrate the advantages of the proposed method in comparison with the state-of-the-art methods. © 2020 Institution of Engineering and Technology. All rights reserved.

关键词：

Task transformer network for joint MRI reconstruction and super-resolution

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Feng, Chun-Mei Yan, Yunlu Fu, Huazhu Chen, Li Xu, Yong Shenzhen Key Laboratory of Visual Object Detection and Recognition Harbin Institute of Technology Shenzhen China Inception Institute of Artificial Intelligence United Arab Emirates First Affiliated Hospital with Nanjing Medical University China

The core problem of Magnetic Resonance Imaging (MRI) is the trade off between acceleration and image quality. Image recon- struction and super-resolution are two crucial techniques in Magnetic Resonance Imaging (MRI). Current methods are designed to perform these tasks separately, ignoring the correlations between them. In this work, we propose an end-to-end task transformer network (T2Net) for joint MRI reconstruction and super-resolution, which allows represen- tations and feature transmission to be shared between multiple task to achieve higher-quality, super-resolved and motion-artifacts-free im- ages from highly undersampled and degenerated MRI data. Our frame- work combines both reconstruction and super-resolution, divided into two sub-branches, whose features are expressed as queries and keys. Specifically, we encourage joint feature learning between the two tasks, thereby transferring accurate task information. We first use two separate CNN branches to extract task-specific features. Then, a task transformer module is designed to embed and synthesize the relevance between the two tasks. Experimental results show that our multi-task model signif- icantly outperforms advanced sequential methods, both quantitatively and qualitatively. © 2021, CC BY.

关键词： Magnetic resonance imaging

Cross-Modal Retrieval: A Systematic Review of Methods and Future Directions

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Wang, Tianshi Li, Fengling Zhu, Lei Li, Jingjing Zhang, Zheng Shen, Heng Tao School of Electronic and Information Engineering Tongji University Shanghai201804 China School of Information Science and Engineering Shandong Normal University Jinan250358 China School of Computer Science and Engineering University of Electronic Science and Technology of China Chengdu611731 China Australian Artificial Intelligence Institute University of Technology Sydney SydneyNSW2007 Australia Shenzhen Key Laboratory of Visual Object Detection and Recognition Harbin Institute of Technology Shenzhen518055 China

With the exponential surge in diverse multi-modal data, traditional uni-modal retrieval methods struggle to meet the needs of users seeking access to data across various modalities. To address this, cross-modal retrieval has emerged, enabling interaction across modalities, facilitating semantic matching, and leveraging complementarity and consistency between heterogeneous data. Although prior literature has reviewed the field of cross-modal retrieval, it suffers from numerous deficiencies in terms of timeliness, taxonomy, and comprehensiveness. This paper conducts a comprehensive review of cross-modal retrieval's evolution, spanning from shallow statistical analysis techniques to vision-language pre-training models. Commencing with a comprehensive taxonomy grounded in machine learning paradigms, mechanisms, and models, the paper delves deeply into the principles and architectures underpinning existing crossmodal retrieval methods. Furthermore, it offers an overview of widely-used benchmarks, metrics, and performances. Lastly, the paper probes the prospects and challenges that confront contemporary cross-modal retrieval, while engaging in a discourse on potential directions for further progress in the field. To facilitate the ongoing research on cross-modal retrieval, we develop a user-friendly toolbox and an open-source repository at https://***. Copyright © 2023, The Authors. All rights reserved.

关键词： Semantics

Partition-A-Medical-Image: Extracting Multiple Representative Sub-regions for Few-shot Medical Image Segmentation

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Zhu, Yazhou Wang, Shidong Xin, Tong Zhang, Zheng Zhang, Haofeng The School of Computer Science and Engineering Nanjing University of Science and Technology Nanjing210094 China The School of Engineering Newcastle University Newcastle upon TyneNE1 7RU United Kingdom The School of Computing Newcastle University Newcastle upon TyneNE1 7RU United Kingdom The Shenzhen Key Laboratory of Visual Object Detection and Recognition Harbin Institute of Technology Shenzhen518055 China

Few-shot Medical Image Segmentation (FSMIS) is a more promising solution for medical image segmentation tasks where high-quality annotations are naturally scarce. However, current mainstream methods primarily focus on extracting holistic representations from support images with large intra-class variations in appearance and background, and encounter difficulties in adapting to query images. In this work, we present an approach to extract multiple representative sub-regions from a given support medical image, enabling fine-grained selection over the generated image regions. Specifically, the foreground of the support image is decomposed into distinct regions, which are subsequently used to derive region-level representations via a designed Regional Prototypical Learning (RPL) module. We then introduce a novel Prototypical Representation Debiasing (PRD) module based on a two-way elimination mechanism which suppresses the disturbance of regional representations by a self-support, Multi-direction Self-debiasing (MS) block, and a support-query, Interactive Debiasing (ID) block. Finally, an Assembled Prediction (AP) module is devised to balance and integrate predictions of multiple prototypical representations learned using stacked PRD modules. Results obtained through extensive experiments on three publicly accessible medical imaging datasets demonstrate consistent improvements over the leading FSMIS methods. The source code is available at https://***/YazhouZhu19/PAMI. © 2023, CC BY.

关键词： Image segmentation

DS-TransUNet: Dual swin transformer U-net for medical image segmentation

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Lin, Ailiang Chen, Bingzhi Xu, Jiayu Zhang, Zheng Lu, Guangming Shenzhen Medical Biometrics Perception and Analysis Engineering Laboratory Harbin Institute of Technology Shenzhen518055 China Bio-Computing Research Center Harbin Institute of Technology Shenzhen518055 China Shenzhen Key Laboratory of Visual Object Detection and Recognition Shenzhen518055 China

Automatic medical image segmentation has made great progress benefit from the development of deep learning. However, most existing methods are based on convolutional neural networks (CNNs), which fail to build long-range dependencies and global context connections due to the limitation of receptive field in convolution operation. Inspired by the success of Transformer whose self-attention mechanism has the powerful abilities of modeling the long-range contextual information, some researchers have expended considerable efforts in designing the robust variants of Transformer-based U-Net. Moreover, the patch division used in vision transformers usually ignores the pixel-level intrinsic structural features inside each patch. To alleviate these problems, in this paper, we propose a novel deep medical image segmentation framework called Dual Swin Transformer U-Net (DS-TransUNet), which might be the first attempt to concurrently incorporate the advantages of hierarchical Swin Transformer into both encoder and decoder of the standard U-shaped architecture to enhance the semantic segmentation quality of varying medical images. Unlike many prior Transformer-based solutions, the proposed DS-TransUNet first adopts dual-scale encoder subnetworks based on Swin Transformer to extract the coarse and fine-grained feature representations of different semantic scales. As the core component for our DS-TransUNet, a well-designed Transformer Interactive Fusion (TIF) module is proposed to effectively establish global dependencies between features of different scales through the self-attention mechanism, in order to make full use of these obtained multi-scale feature representations. Furthermore, we also introduce the Swin Transformer block into decoder to further explore the long-range contextual information during the up-sampling process. Extensive experiments across four typical tasks for medical image segmentation demonstrate the effectiveness of DS-TransUNet, and show that our approac

关键词： Semantic Segmentation

Discrete semantic embedding hashing for scalable cross-modal retrieval

学校读者我要写书评

暂无评论

Discrete semantic embedding hashing for scalable cross-modal...

IEEE International Conference on Systems, Man and Cybernetics

作者： Junjie Liu Lunke Fei Wei Jia Shuping Zhao Jie Wen Shaohua Teng Wei Zhang School of Computer Science and Technology Guangdong University of Technology Guangzhou China School of Computer and Information Hefei University of Technology Hefei China Shenzhen Key Laboratory of Visual Object Detection and Recognition Harbin Institute of Technology Shenzhen China

ISBN: (纸本)9781665442084

Cross-modal hashing has attracted much attention for cross-modal retrieval and achieved promising performance due to its powerful capacity. Some existing cross-modal hashing methods construct pairwise similarities to represent the relationship of heterogeneous data, which require much computation time and storage space, making them unscalable for large-scale retrieval tasks. In this paper, we propose a novel supervised Discrete Semantic Embedding Hashing (DSEH) for cross-modal retrieval. Specifically, we first learn the common representation of heterogeneous data by embedding the semantic labels into a collective matrix factorization, such that both intra- and inter-modality similarities can be well captured. Then, we learn the hash codes in the discrete space based on the learned common representation via an orthogonal rotation technique. Moreover, we learn the multi-modal hash functions that can efficiently convert out-of-sample instances into unified hash codes. Extensive experimental results on three widely used benchmark databases demonstrate the superiority of the proposed DSEH compared with previous state-of-the-arts.

关键词： Hash functions Training Databases Semantics Information services Binary codes Benchmark testing