检索结果-内蒙古大学图书馆

Pick of the Bunch: Detecting Infrared Small Targets Beyond Hit-Miss Trade-Offs via Selective Rank-Aware Attention

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Dai, Yimian Pan, Peiwen Qian, Yulei Li, Yuxuan Li, Xiang Yang, Jian Wang, Huan PCA Lab Key Lab of Intelligent Perception and Systems for High-Dimensional Information of Ministry of Education Jiangsu Key Lab of Image Video Understanding for Social Security China The department of intelligence science School of Computer Science and Engineering Nanjing University of Science and Technology Nanjing China Nanjing Marine Radar Institute Nanjing China VCIP CS Nankai University China The NKIARI Futian Shenzhen China

Infrared small target detection faces the inherent challenge of precisely localizing dim targets amidst complex background clutter. Traditional approaches struggle to balance detection precision and false alarm rates. To break this dilemma, we propose SeRankDet, a deep network that achieves high accuracy beyond the conventional hit-miss trade-off, by following the "Pick of the Bunch" principle. At its core lies our Selective Rank-Aware Attention (SeRank) module, employing a non-linear Top-K selection process that preserves the most salient responses, preventing target signal dilution while maintaining constant complexity. Furthermore, we replace the static concatenation typical in U-Net structures with our Large Selective Feature Fusion (LSFF) module, a dynamic fusion strategy that empowers SeRankDet with adaptive feature integration, enhancing its ability to discriminate true targets from false alarms. The network’s discernment is further refined by our Dilated Difference Convolution (DDC) module, which merges differential convolution aimed at amplifying subtle target characteristics with dilated convolution to expand the receptive field, thereby substantially improving target-background separation. Despite its lightweight architecture, the proposed SeRankDet sets new benchmarks in state-of-the-art performance across multiple public datasets. The code is available at https://***/GrokCV/SeRankDet. Copyright © 2024, The Authors. All rights reserved.

关键词： Semantic Segmentation

Generative Subgraph Contrast for Self-Supervised Graph Representation Learning

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Han, Yuehui Hui, Le Jiang, Haobo Qian, Jianjun Xie, Jin Key Lab of Intelligent Perception and Systems for High-Dimensional Information of Ministry of Education Jiangsu Key Lab of Image and Video Understanding for Social Security PCA Lab School of Computer Science and Engineering Nanjing University of Science and Technology China

Contrastive learning has shown great promise in the field of graph representation learning. By manually constructing positive/negative samples, most graph contrastive learning methods rely on the vector inner product based similarity metric to distinguish the samples for graph representation. However, the handcrafted sample construction (e.g., the perturbation on the nodes or edges of the graph) may not effectively capture the intrinsic local structures of the graph. Also, the vector inner product based similarity metric cannot fully exploit the local structures of the graph to characterize the graph difference well. To this end, in this paper, we propose a novel adaptive subgraph generation based contrastive learning framework for efficient and robust self-supervised graph representation learning, and the optimal transport distance is utilized as the similarity metric between the subgraphs. It aims to generate contrastive samples by capturing the intrinsic structures of the graph and distinguish the samples based on the features and structures of subgraphs simultaneously. Specifically, for each center node, by adaptively learning relation weights to the nodes of the corresponding neighborhood, we first develop a network to generate the interpolated subgraph. We then construct the positive and negative pairs of subgraphs from the same and different nodes, respectively. Finally, we employ two types of optimal transport distances (i.e., Wasserstein distance and Gromov-Wasserstein distance) to construct the structured contrastive loss. Extensive node classification experiments on benchmark datasets verify the effectiveness of our graph contrastive learning method. Source code is available at https://***/yh-han/***. Copyright © 2022, The Authors. All rights reserved.

关键词： Classification (of information)

3D Siamese Transformer Network for Single Object Tracking on Point Clouds

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Hui, Le Wang, Lingpeng Tang, Linghua Lan, Kaihao Xie, Jin Yang, Jian Key Lab of Intelligent Perception and Systems for High-Dimensional Information of Ministry of Education Jiangsu Key Lab of Image and Video Understanding for Social Security PCA Lab School of Computer Science and Engineering Nanjing University of Science and Technology China

Siamese network based trackers formulate 3D single object tracking as cross-correlation learning between point features of a template and a search area. Due to the large appearance variation between the template and search area during tracking, how to learn the robust cross correlation between them for identifying the potential target in the search area is still a challenging problem. In this paper, we explicitly use Transformer to form a 3D Siamese Transformer network for learning robust cross correlation between the template and the search area of point clouds. Specifically, we develop a Siamese point Transformer network to learn shape context information of the target. Its encoder uses self-attention to capture non-local information of point clouds to characterize the shape information of the object, and the decoder utilizes cross-attention to upsample discriminative point features. After that, we develop an iterative coarse-to-fine correlation network to learn the robust cross correlation between the template and the search area. It formulates the cross-feature augmentation to associate the template with the potential target in the search area via cross attention. To further enhance the potential target, it employs the ego-feature augmentation that applies self-attention to the local k-NN graph of the feature space to aggregate target features. Experiments on the KITTI, nuScenes, and Waymo datasets show that our method achieves state-of-the-art performance on the 3D single object tracking task. Source code is available at https://***/fpthink/STNet. Copyright © 2022, The Authors. All rights reserved.

关键词： Iterative methods

A multi-stage semi-supervised learning for ankle fracture classification on CT images

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Liu, Hongzhi Li, Guicheng Nie, Jiacheng Tang, Hui Yang, Chunfeng Feng, Qianjin Xu, Hailin Chen, Yang The School of Computer Science and Engineering Southeast University Nanjing210096 China The Peking University People's Hospital Beijing100000 China The Guangdong Provincial Key Laboratory of Medical Image Processing School of Biomedical Engineering Southern Medical University Guangzhou510515 China The Key Laboratory of New Generation Artificial Intelligence Technology and its Interdisciplinary Applications Southeast University Ministry of Education China Jiangsu Provincial Joint International Research Laboratory of Medical Information Processing School of Computer Science and Engineering Southeast University Nanjing210096 China Jiangsu Key Laboratory of Molecular and Functional Imaging Department of Radiology Zhongda Hospital Southeast University Nanjing210009 China

Because of the complicated mechanism of ankle injury, it is very difficult to diagnose ankle fracture in clinic. In order to simplify the process of fracture diagnosis, an automatic diagnosis model of ankle fracture was proposed. Firstly, a tibia-fibula segmentation network is proposed for the joint tibiofibular region of the ankle joint, and the corresponding segmentation dataset is established on the basis of fracture data. Secondly, the image registration method is used to register the bone segmentation mask with the normal bone mask. Finally, a semi-supervised classifier is constructed to make full use of a large number of unlabeled data to classify ankle fractures. Experiments show that the proposed method can segment fractures with fracture lines accurately and has better performance than the general method. At the same time, this method is superior to classification network in several indexes. Copyright © 2024, The Authors. All rights reserved.

关键词： Image classification

FedDADP: A Privacy-Risk-Adaptive Differential Privacy Protection Method for Federated Android Malware Classifier

学校读者我要写书评

暂无评论

FedDADP: A Privacy-Risk-Adaptive Differential Privacy Protec...

International Joint Conference on Neural Networks (IJCNN)

作者： Changnan Jiang Chunhe Xia Mengyao Liu Rui Fang Pengfei Li Huacheng Li Tianbo Wang Beijing Key Lab. of Network Technology Beihang University Beijing China Guangxi Collaborative Innovation Center of Multi-Source Information Integration and Intelligent Processing Guangxi Normal University Guilin China School of Cyber Science and Technology Beihang University Beijing China SNationa Innovation Center of Intelligent and Connected Vehicles Beijing China Shanghai Key Laboratory of Computer Software Evaluating and Testing China

ISBN: (数字)9798350359312

ISBN: (纸本)9798350359329

The federated Android malware classifier has attracted much attention owing to its advantages of privacy protection and multi-party joint modeling. However, the research indicates that the gradient transmitted within the federated classifier still encodes the user's sensitive information, exposing it to indirect privacy inference threats from curious servers. Differential privacy is a recognized and effective way to address this privacy breach threat by adding noise to the user's model parameters to limit the attacker's inference of sensitive information. However, the protection effect of existing differential privacy methods is at the cost of significantly reducing the model's classification accuracy, and it cannot be reasonably balanced. To address this challenge, we propose a privacy protection method, FedDADP. FedDADP performs adaptive, lightweight privacy configuration in its training time dimension and model space dimension according to the privacy risk distribution law in the federated Android malware classifier to protect users' privacy while maintaining the model's utility. Numerous experiments on the Androzoo dataset and multiple baseline classifiers show that FedDADP protects users' sensitive information better (7% more effectiveness against adversaries' inference) than baseline differential privacy methods and achieves better model utility (classification accuracy improves by about 8%) with the same privacy budget.

关键词： Training Privacy Adaptation models Differential privacy Accuracy Privacy breach Malware

Improving Adversarial Transferability with Neighbourhood Gradient information

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Guo, Haijing Wang, Jiafeng Chen, Zhaoyu Jiang, Kaixun Hong, Lingyi Guo, Pinxue Li, Jinglun Zhang, Wenqiang Shanghai Key Lab of Intelligent Information Processing School of Computer Science Fudan University Shanghai200433 China Shanghai Engineering Research Center of AI & Robotics Academy for Engineering & Technology Fudan University Shanghai200433 China Engineering Research Center of Robotics Ministry of Education Academy for Engineering & Technology Fudan University Shanghai200433 China

Deep neural networks (DNNs) are known to be susceptible to adversarial examples, leading to significant performance degradation. In black-box attack scenarios, a considerable attack performance gap between the surrogate model and the target model persists. This work focuses on enhancing the transferability of adversarial examples to narrow this performance gap. We observe that the gradient information around the clean image, i.e. Neighbourhood Gradient information, can offer high transferability. Leveraging this, we propose the NGI-Attack, which incorporates Example Backtracking and Multiplex Mask strategies, to use this gradient information and enhance transferability fully. Specifically, we first adopt Example Backtracking to accumulate Neighbourhood Gradient information as the initial momentum term. Multiplex Mask, which forms a multi-way attack strategy, aims to force the network to focus on non-discriminative regions, which can obtain richer gradient information during only a few iterations. Extensive experiments demonstrate that our approach significantly enhances the adversarial transferability. Especially, when attacking numerous defense models, we achieve an average attack success rate of 95.8%. Notably, our method can plugin with any off-the-shelf algorithm to improve their attack performance without additional time cost. Copyright © 2024, The Authors. All rights reserved.

关键词：

A Survey on RGB, 3D, and Multimodal Approaches for Unsupervised Industrial Image Anomaly Detection

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Lin, Yuxuan Chang, Yang Tong, Xuan Yu, Jiawen Liotta, Antonio Huang, Guofan Song, Wei Zeng, Deyu Wu, Zongze Wang, Yan Zhang, Wenqiang Shanghai Key Lab of Intelligent Information Processing School of Computer Science Fudan University Shanghai200438 China Shanghai Engineering Research Center of AI & Robotics Academy for Engineering & Technology Fudan University Shanghai200438 China Faculty of Computer Science Free University of Bozen-Bolzano Bolzano39100 Italy College of Information Technology Shanghai Ocean University Shanghai201306 China College of Mechatronics and Control Engineering Shenzhen University Shenzhen518060 China

In the advancement of industrial informatization, unsupervised anomaly detection technology effectively overcomes the scarcity of abnormal samples and significantly enhances the automation and reliability of smart manufacturing. As an important branch, industrial image anomaly detection focuses on automatically identifying visual anomalies in industrial scenarios (such as product surface defects, assembly errors, and equipment appearance anomalies) through computer vision techniques. With the rapid development of Unsupervised industrial Image Anomaly Detection (UIAD), excellent detection performance has been achieved not only in RGB setting but also in 3D and multimodal (RGB and 3D) settings. However, existing surveys primarily focus on UIAD tasks in RGB setting, with little discussion in 3D and multimodal settings. To address this gap, this artical provides a comprehensive review of UIAD tasks in the three modal settings. Specifically, we first introduce the task concept and process of UIAD. We then overview the research on UIAD in three modal settings (RGB, 3D, and multimodal), including datasets and methods, and review multimodal feature fusion strategies in multimodal setting. Finally, we summarize the main challenges faced by UIAD tasks in the three modal settings, and offer insights into future development directions, aiming to provide researchers with a comprehensive reference and offer new perspectives for the advancement of industrial informatization. Corresponding resources are available at https://***/Sunny5250/Awesome-Multi-Setting-UIAD. Copyright © 2024, The Authors. All rights reserved.

关键词： Smart manufacturing

3D EAGAN: 3D edge-aware attention generative adversarial network for prostate segmentation in transrectal ultrasound images

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Liu, Mengqing Shao, Xiao Jiang, Liping Wu, Kaizhi School of Information Engineering Nanchang Hangkong University Jiangxi Nanchang China School of Computer Science Nanjing University of Information Science and Technology Jiangsu Nanjing China The First Affiliated Hospital of Nanchang University Nanchang University Jiangxi Nanchang China Key Laboratory of Jiangxi Province for Image Processing and Pattern Recognition Nanchang Hangkong University Nanchang China

Background: Segment prostates from transrectal ultrasound (TRUS) images plays an essential role in the diagnosis and treatment of prostate cancer. However, traditional segmentation methods are time-consuming and laborious. To address this issue, there is an urgent need to develop computer algorithms that can automatically segment prostates from TRUS images, which makes it become the direction and form of future development. Purpose: Automatic prostate segmentation in TRUS images has always been a challenging problem, since prostates in TRUS images have ambiguous boundaries and inhomogeneous intensity distribution. Although many prostate segmentation methods have been proposed, they still need to be improved due to the lack of sensibility to edge information. Consequently, the objective of this study is to devise a highly effective prostate segmentation method that overcomes these limitations and achieves accurate segmentation of prostates in TRUS images. Methods: A 3D edge-aware attention generative adversarial network (3D EAGAN)-based prostate segmentation method is proposed in this paper, which consists of an edge-aware segmentation network (EASNet) that performs the prostate segmentation and a discriminator network that distinguishes predicted prostates from real prostates. The proposed EASNet is composed of an encoder-decoder-based U-Net backbone network, a detail compensation module, four 3D spatial and channel attention modules, an edge enhance module, and a global feature extractor. The detail compensation module is proposed to compensate for the loss of detailed information caused by the down-sampling process of the encoder. The features of the detail compensation module are selectively enhanced by the 3D spatial and channel attention module. Furthermore, an edge enhance module is proposed to guide shallow layers in the EASNet to focus on contour and edge information in prostates. Finally, features from shallow layers and hierarchical features from the decod

关键词： Generative adversarial networks

On Approximate Opacity of Stochastic Control Systems

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Liu, Siyuan Yin, Xiang Dimarogonas, Dimos V. Zamani, Majid Division of Decision and Control Systems KTH Royal Institute of Technology Stockholm Sweden Department of Automation Shanghai Jiao Tong University Key Lab of System Control & Information Processing Ministry of Education Shanghai China Computer Science Department University of Colorado BoulderCO80309 United States Computer Science Department Ludwig Maximilian University of Munich Germany

This paper investigates an important class of information-flow security property called opacity for stochastic control systems. Opacity captures whether a system’s secret behavior (a subset of the system’s behavior that is considered to be critical) can be kept from outside observers. Existing works on opacity for control systems only provide a binary characterization of the system’s security level by determining whether the system is opaque or not. In this work, we introduce a quantifiable measure of opacity that considers the likelihood of satisfying opacity for stochastic control systems modeled as general Markov decision processes (gMDPs). We also propose verification methods tailored to the new notions of opacity for finite gMDPs by using value iteration techniques. Then, a new notion called approximate opacity-preserving stochastic simulation relation is proposed, which captures the distance between two systems’ behaviors in terms of preserving opacity. Based on this new system relation, we show that one can verify opacity for stochastic control systems using their abstractions (modeled as finite gMDPs). We also discuss how to construct such abstractions for a class of gMDPs under certain stability conditions. Copyright © 2024, The Authors. All rights reserved.

关键词： Stochastic control systems