检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Jiang, Kai Huang, Jiaxing Xie, Weiying Lei, Jie Li, Yunsong Shao, Ling Lu, Shijian State Key Laboratory of Integrated Services Networks Xidian University Xi'An710071 China S-lab School of Computer Science and Engineering Nanyang Technological University Singapore UCAS-Terminus AI Lab University of Chinese Academy of Sciences China School of Electrical and Data Engineering The University of Technology Sydney Australia

Camera-only Bird's Eye View (BEV) has demonstrated great potential in environment perception in a 3D space. However, most existing studies were conducted under a supervised setup which cannot scale well while handling various new data. Unsupervised domain adaptive BEV, which effective learning from various unlabelled target data, is far under-explored. In this work, we design DA-BEV, the first domain adaptive camera-only BEV framework that addresses domain adaptive BEV challenges by exploiting the complementary nature of image-view features and BEV features. DA-BEV introduces the idea of query into the domain adaptation framework to derive useful information from image-view and BEV features. It consists of two query-based designs, namely, query-based adversarial learning (QAL) and query-based self-training (QST), which exploits image-view features or BEV features to regularize the adaptation of the other. Extensive experiments show that DA-BEV achieves superior domain adaptive BEV perception performance consistently across multiple datasets and tasks such as 3D object detection and 3D scene segmentation. Copyright © 2024, The Authors. All rights reserved.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Flooded Cropland Mapping Based on GF-3 and Mapbox Imagery Using Semantic Segmentation: A Case Study of Typhoon Siamba in Western Guangdong in July 2022

Flooded Cropland Mapping Based on GF-3 and Mapbox Imagery Us...

引用

2022 International Conference on Geographic Information and Remote Sensing Technology, GIRST 2022

作者： Ku, Mengjun Jiang, Hao Li, Dan Wang, Chongyang Key Lab of Guangdong for Utilization of Remote Sensing and Geographical Information System Guangdong Open Laboratory of Geospatial Information Technology and Application Research Center of Guangdong Province for Engineering Technology Application of Remote Sensing Big Data Guangzhou Institute of Geography Guangdong Academy of Sciences Guangzhou510070 China Guangdong University of Technology Guangzhou China

ISBN: (纸本)9781510662186

Typhoon Siamba made landfall in Western Guangdong on July 2, 2022, causing great losses to crops in western Guangdong. Radar remote sensing can penetrate through clouds and fog, is suitable for identifying flooded areas before or after typhoons and in rainy weather. However, radar flooded waterbody mapping faces two major problems: distinguishing between flooded areas and natural waterbody, and the other is noise interference from confusing ground objects. Aiming at these problems, the study proposes a method of high-precision cropland information combined with waterbody identification. Based on GF-3 and Mapbox data, this paper first uses watershed semantic segmentation to extract initial waterbody, then uses SegFormer deep learning technology to identify cropland, finally realizes flooded cropland mapping based on cropland information. This study concluded that the affected cropland in Zhanjiang and Maoming City, Guangdong Province, China is 75.437 km2 and 31.175 km2 respectively. The cropland extraction accuracy and Intersection over Union (IoU) are 96.65% and 92.64% respectively. The study shows that flood monitoring combined with cropland identification information can effectively avoid noise interference and accurately extract the flood range, and achieve high-precision flooded cropland mapping. © 2023 SPIE.

关键词： Watersheds

来源：评论

学校读者我要写书评

暂无评论

NumbOD: A Spatial-Frequency Fusion Attack Against Object Detectors

arXiv

引用

arXiv 2024年

作者： Zhou, Ziqi Li, Bowen Song, Yufei Yu, Zhifei Hu, Shengshan Wan, Wei Zhang, Leo Yu Yao, Dezhong Jin, Hai National Engineering Research Center for Big Data Technology and System China Services Computing Technology and System Lab China Cluster and Grid Computing Lab Hubei Engineering Research Center on Big Data Security China Hubei Key Laboratory of Distributed System Security China School of Computer Science and Technology Huazhong University of Science and Technology China School of Cyber Science and Engineering Huazhong University of Science and Technology China School of Information and Communication Technology Griffith University Australia

With the advancement of deep learning, object detectors (ODs) with various architectures have achieved significant success in complex scenarios like autonomous driving. Previous adversarial attacks against ODs have been focused on designing customized attacks targeting their specific structures (e.g., NMS and RPN), yielding some results but simultaneously constraining their scalability. Moreover, most efforts against ODs stem from image-level attacks originally designed for classification tasks, resulting in redundant computations and disturbances in object-irrelevant areas (e.g., background). Consequently, how to design a model-agnostic efficient attack to comprehensively evaluate the vulnerabilities of ODs remains challenging and unresolved. In this paper, we propose NumbOD, a brand-new spatial-frequency fusion attack against various ODs, aimed at disrupting object detection within images. We directly leverage the features output by the OD without relying on its internal structures to craft adversarial examples. Specifically, we first design a dual-track attack target selection strategy to select high-quality bounding boxes from OD outputs for targeting. Subsequently, we employ directional perturbations to shift and compress predicted boxes and change classification results to deceive ODs. Additionally, we focus on manipulating the high-frequency components of images to confuse ODs' attention on critical objects, thereby enhancing the attack efficiency. Our extensive experiments on nine ODs and two datasets show that NumbOD achieves powerful attack performance and high stealthiness. Copyright © 2024, The Authors. All rights reserved.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Suppress Content Shift: Better Diffusion Features via Off-the-Shelf Generation Techniques 38

Suppress Content Shift: Better Diffusion Features via Off-th...

引用

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Meng, Benyuan Xu, Qianqian Wang, Zitai Yang, Zhiyong Cao, Xiaochun Huang, Qingming Institute of Information Engineering CAS China School of Cyber Security University of Chinese Academy of Sciences China Key Lab. of Intelligent Information Processing Institute of Computing Technology CAS China Peng Cheng Laboratory China School of Computer Science and Tech. University of Chinese Academy of Sciences China School of Cyber Science and Tech. Shenzhen Campus of Sun Yat-sen University China Key Laboratory of Big Data Mining and Knowledge Management CAS China

Diffusion models are powerful generative models, and this capability can also be applied to discrimination. The inner activations of a pre-trained diffusion model can serve as features for discriminative tasks, namely, diffusion feature. We discover that diffusion feature has been hindered by a hidden yet universal phenomenon that we call content shift. To be specific, there are content differences between features and the input image, such as the exact shape of a certain object. We locate the cause of content shift as one inherent characteristic of diffusion models, which suggests the broad existence of this phenomenon in diffusion feature. Further empirical study also indicates that its negative impact is not negligible even when content shift is not visually perceivable. Hence, we propose to suppress content shift to enhance the overall quality of diffusion features. Specifically, content shift is related to the information drift during the process of recovering an image from the noisy input, pointing out the possibility of turning off-the-shelf generation techniques into tools for content shift suppression. We further propose a practical guideline named GATE to efficiently evaluate the potential benefit of a technique and provide an implementation of our methodology. Despite the simplicity, the proposed approach has achieved superior results on various tasks and datasets, validating its potential as a generic booster for diffusion features. Our code is available at this url. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Method of Extracting Discipline Inspection Cases Based on Deep Learning 3

A Method of Extracting Discipline Inspection Cases Based on ...

引用

3rd International Conference on Education, Knowledge and Information Management, ICEKIM 2022

作者： Liu, XiaoLing Chen, JunJie Gao, Jing Fan, Hao Dong, JianMin Inner Mongolia Agricultural University College of Computer and Information Engineering China Inner Mongolia Autonomous Reg. Key Lab. of Large Data Res. and Applic. of Agric. and Anim. Husbandry China Inner Mongolia Discipline Inspection and Supervision Big Data Laboratory Hohhot China

ISBN: (纸本)9781665416665

Event extraction is one of the key tasks in information extraction tasks, and it has been widely used in various fields in recent years. In the field of disciplinary inspection and supervision, disciplinary inspection case text has the characteristics of multiple types, multiple professional terms, and strong correlation between event types and event arguments. In the face of huge data, relying on manual analysis by manpower has seriously affected the efficiency of disciplinary inspection. However, there is no corpus in the field of discipline inspection currently available. This article uses BIO annotation to construct a corpus of discipline inspection to lay the foundation for subsequent work. Proposed BERT-BiGRU-CRF event joint extraction model. Use the BERT model to train the discipline inspection corpus, and Combine BiGRU network and CRF network to realize event type recognition and argument extraction. The experimental results show that the model can effectively extract event information in the discipline inspection field. At the same time, in order to better facilitate the work of disciplinary inspection personnel, a disciplinary inspection and supervision event extraction system is constructed to systematically extract all event information contained in the case. © 2022 IEEE.

关键词： Inspection

来源：评论

学校读者我要写书评

暂无评论

PS3: Precise Patch Presence Test Based on Semantic Symbolic Signature

PS3: Precise Patch Presence Test Based on Semantic Symbolic ...

引用

International Conference on Software engineering (ICSE)

作者： Qi Zhan Xing Hu Zhiyang Li Xin Xia David Lo Shanping Li The State Key Laboratory of Blockchain and Data Security Zhejiang University Hangzhou China Zhejiang University Hangzhou China Software Engineering Application Technology Lab Huawei China Singapore Management University Singapore

ISBN: (数字)9798400702174

ISBN: (纸本)9798350382143

During software development, vulnerabilities have posed a significant threat to users. Patches are the most effective way to combat vulnerabilities. In a large-scale software system, testing the presence of a security patch in every affected binary is crucial to ensure system security. Identifying whether a binary has been patched for a known vulnerability is challenging, as there may only be small differences between patched and vulnerable versions. Existing approaches mainly focus on detecting patches that are compiled in the same compiler options. However, it is common for developers to compile programs with very different compiler options in different situations, which causes inaccuracy for existing methods. In this paper, we propose a new approach named $PS^{3}$ , referring to precise patch presence test based on semantic-level symbolic signature. $PS^{3}$ exploits symbolic emulation to extract signatures that are stable under different compiler options. Then $PS^{3}$ can precisely test the presence of the patch by comparing the signatures between the reference and the target at semantic level. To evaluate the effectiveness of our approach, we constructed a dataset consisting of 3,631 (CVE, binary) pairs of 62 recent CVEs in four C/C++ projects. The experimental results show that $PS^{3}$ achieves scores of 0.82, 0.97, and 0.89 in terms of precision, recall, and F1 score, respectively. $PS^{3}$ outperforms the state-of-the-art baselines by improving 33% in terms of F1 score and remains stable in different compiler options.

关键词： Program processors Semantics Emulation Software systems Security Testing Software engineering

来源：评论

学校读者我要写书评

暂无评论

Machine Learning is All You Need: A Simple Token-Based Approach for Effective Code Clone Detection

Machine Learning is All You Need: A Simple Token-Based Appro...

引用

International Conference on Software engineering (ICSE)

作者： Siyue Feng Wenqi Suo Yueming Wu Deqing Zou Yang Liu Hai Jin Huazhong University of Science and Technology China National Engineering Research Center for Big Data Technology and System Services Computing Technology and System Lab Hubei Key Laboratory of Distributed System Security Hubei Engineering Research Center on Big Data Security Cluster and Grid Computing Lab Jinyinhu Laboratory Wuhan China School of Cyber Science and Engineering HUST Wuhan China Nanyang Technological University Singapore School of Computer Science and Technology HUST Wuhan China

ISBN: (数字)9798400702174

ISBN: (纸本)9798350382143

As software engineering advances and the code demand rises, the prevalence of code clones has increased. This phenomenon poses risks like vulnerability propagation, underscoring the growing importance of code clone detection techniques. While numerous code clone detection methods have been proposed, they often fall short in real-world code environments. They either struggle to identify code clones effectively or demand substantial time and computational resources to handle complex clones. This paper introduces a code clone detection method namely Toma using tokens and machine learning. Specifically, we extract token type sequences and employ six similarity calculation methods to generate feature vectors. These vectors are then input into a trained machine learning model for classification. To evaluate the effectiveness and scalability of Toma, we conduct experiments on the widely used BigCloneBench dataset. Results show that our tool outperforms token-based code clone detectors and most tree-based clone detectors, demonstrating high effectiveness and significant time savings.

关键词： Codes Scalability Computational modeling Cloning Machine learning Detectors Feature extraction

来源：评论

学校读者我要写书评

暂无评论

Not All Diffusion Model Activations Have Been Evaluated as Discriminative Features 38

Not All Diffusion Model Activations Have Been Evaluated as D...

引用

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Meng, Benyuan Xu, Qianqian Wang, Zitai Cao, Xiaochun Huang, Qingming Institute of Information Engineering CAS China School of Cyber Security University of Chinese Academy of Sciences China Key Lab. of Intelligent Information Processing Institute of Computing Technology CAS China Peng Cheng Laboratory China School of Cyber Science and Tech. Shenzhen Campus of Sun Yat-sen University China School of Computer Science and Tech. University of Chinese Academy of Sciences China Key Laboratory of Big Data Mining and Knowledge Management CAS China

Diffusion models are initially designed for image generation. Recent research shows that the internal signals within their backbones, named activations, can also serve as dense features for various discriminative tasks such as semantic segmentation. Given numerous activations, selecting a small yet effective subset poses a fundamental problem. To this end, the early study of this field performs a large-scale quantitative comparison of the discriminative ability of the activations. However, we find that many potential activations have not been evaluated, such as the queries and keys used to compute attention scores. Moreover, recent advancements in diffusion architectures bring many new activations, such as those within embedded ViT modules. Both combined, activation selection remains unresolved but overlooked. To tackle this issue, this paper takes a further step with a much broader range of activations evaluated. Considering the significant increase in activations, a full-scale quantitative comparison is no longer operational. Instead, we seek to understand the properties of these activations, such that the activations that are clearly inferior can be filtered out in advance via simple qualitative evaluation. After careful analysis, we discover three properties universal among diffusion models, enabling this study to go beyond specific models. On top of this, we present effective feature selection solutions for several popular diffusion models. Finally, the experiments across multiple discriminative tasks validate the superiority of our method over the SOTA competitors. Our code is available at this url. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Service composition based on hierarchical state machine

Service composition based on hierarchical state machine

引用

2021 International Conference on Computer Technology and Media Convergence Design, CTMCD 2021

作者： Zhang, Sa Zhu, Xuefeng Beijing Key Lab of Petroleum Data Mining Department of Computer Science College of Information Science and Engineering China University of Petroleum Beijing Beijing China

ISBN: (纸本)9781665448567

There have been many research methods on Web service composition, and state machine is one of them. The service composition based on the state machine is easier to understand, more intuitive and clearer than other methods. However, as the number of states increases, the jump between states will also become complicated. In this regard, a solution to the service composition based on a hierarchical state machine is proposed, which uses the service state machine to transform into a service rule library, and finally obtains the hierarchical state machine by adding default states and hierarchical generation algorithms. The reachability and equivalence verification proves that the method is feasible. Secondly, in the process of Web service composition, there are situations where multiple sub-services meet the user's functional requirements. In this case, in order to meet other non-functional requirements of users, it is necessary to filter the sub-services in the set of candidate sub-services. In this regard, an improved firework particle swarm algorithm is proposed. The good point set method is introduced to initialize the particle population, which improves the diversity of the particle population at the beginning of the algorithm, and uses the particle position instead of the fitness value to improve the explosion radius formula and expand the search range of the algorithm. Improve convergence speed. In order to verify the effectiveness of the proposed algorithm, the experimental data adopts the QWS data set, and compares the improved firework particle swarm algorithm with IDPSO and FWPSO algorithms. The experimental results show that the proposed improved firework particle swarm algorithm can solve Web service composition more effectively Optimization. © 2021 IEEE

关键词： Quality of service

来源：评论

学校读者我要写书评

暂无评论

The Investigation and Prediction of Voriconazole-Associated Hepatotoxicity under Therapeutic Drug Monitoring*

The Investigation and Prediction of Voriconazole-Associated ...

引用

Annual International Conference of the IEEE engineering in Medicine and Biology Society (EMBC)

作者： Jing Ma Yu Wang Shuang Ma Jingsong Li Research Center for Healthcare Data Science Zhejiang Lab Hangzhou China Engineering Research Center of EMR and Intelligent Expert Systems Ministry of Education Key Laboratory for Biomedical Engineering of Ministry of Education College of Biomedical Engineering and Instrument Science Zhejiang University Hangzhou China

Voriconazole is commonly used as the first-line agent to treat invasive fungal infections (IFIs), but the induction of hepatotoxicity limits its use. To improve the treatment outcomes and minimize toxicity, doctors often administer Therapeutic Drug Monitoring (TDM) to patients receiving voriconazole treatment. Here, we conducted a real-world clinical investigation of voriconazole-treated patients and found significant differences between the TDM (n=318) and non-TDM cohort (n=6,379), and such incidence of hepatotoxicity showed 10.6% in the non-TDM cohort, compared with 21.5% in the TDM cohort. Based on our previous investigation, we presented and compared several machine learning models (including AdaBoost, decision tree, GBDT, logistic regression, neural networks, and random forest) for the early warning of voriconazole-associated hepatoxicity. Through the five-fold cross validation, the logistic model outperformed other models with a mean AUC of 0.7933±0.0934. Our findings offer important insights into the safe and effective application of voriconazole.

关键词：

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：