检索结果-内蒙古大学图书馆

12th International Conference on Computing and Pattern Recognition, ICCPR 2023

作者： Yang, Peixuan Cao, Yichao Lu, Xiaobo College of Software Engineering Southeast University JiangSu NanJing China Key Laboratory of Measurement and Control of Complex Systems of Engineering Ministry of Education JiangSu NanJing China School of Automation Southeast University JiangSu NanJing China

ISBN: (纸本)9798400707988

Change Detection is a classic and fundamental task in remote sensing(RS), and has accomplish remarkable achievement on bitemporal image based methods. However, bitemporal images are very expensive and needs professional researchers to align and label the paired images. Furthermore, most bitemporal image based methods are only semantic-aware not instance-aware, which indicates these methods only use paired images' semantic information. Thus, we propose a method that only needs unpaired RS images with semantic labels. In this paper, we introduce multi-task learning into change detection progress, which comprises four major parts: Boundary-Aware Module, Contextual-Aware Module, Segmentation Module, Change Detection Module. At first, we construct Boundary-Aware Module and Contextual-Aware Module to extract sharp boundary and dense contextual features. Then we feed the feature maps to Change Detection Module and Segmentation Module to accomplish the subtasks. The comprehensive experiments on a single temporal image dataset demonstrate the efficiency of our approach over other bitemporal image based methods, and the experimental results shows our approach is competitive with other methods while all methods was trained on bitemporal image dataset. © 2023 ACM.

关键词： Change detection

来源：评论

学校读者我要写书评

暂无评论

On the properties of kullback-leibler divergence between multivariate Gaussian distributions 23

On the properties of kullback-leibler divergence between mul...

引用

Proceedings of the 37th International Conference on Neural Information Processing systems

作者： Yufeng Zhang Jialu Pan Kenli Li Wanwei Liu Zhenbang Chen Xinwang Liu Ji Wang College of Computer Science and Electronic Engineering Hunan University Changsha China College of Computer Key Laboratory of Software Engineering for Complex Systems National University of Defense Technology Changsha China College of Computer State Key Laboratory for High Performance Computing Key laboratory of Software Engineering for Complex Systems National University of Defense Technology Changsha China

Kullback-Leibler (KL) divergence is one of the most important measures to calculate the difference between probability distributions. In this paper, we theoretically study several properties of KL divergence between multivariate Gaussian distributions. Firstly, for any two n-dimensional Gaussian distributions Ɲ1 and Ɲ2, we prove that when KL(Ɲ2‖Ɲ1) ≤ ε (ε > 0) the supremum of KL (Ɲ1‖Ɲ2) is (1/2) ((-W0 (-e-(1+2ε)))-1 + log(-W0 (-e-(1+2ε))) - 1), where W0 is the principal branch of Lambert W function. For small ε, the supremum is ε + 2ε1.5 + O (ε2). This quantifies the approximate symmetry of small KL divergence between Gaussian distributions. We further derive the infimum of KL(Ɲ1‖Ɲ2) when KL(Ɲ2‖Ɲ1) ≥ M (M > 0). We give the conditions when the supremum and infimum can be attained. Secondly, for any three n-dimensional Gaussian distributions Ɲ1, Ɲ2, and Ɲ3, we theoretically show that an upper bound of KL (Ɲ1‖Ɲ3) is 3ε1 + 3ε2 + 2√ε1ε2 + o(ε1)+ o(ε2) when KL (Ɲ1‖Ɲ2) ≤ ε1 and KL(Ɲ2‖Ɲ3) ≤ ε2 (ε1, ε2 ≥ 0). This reveals that KL divergence between Gaussian distributions follows a relaxed triangle inequality. Note that, all these bounds in the theorems presented in this work are independent of the dimension n. Finally, we discuss several applications of our theories in deep learning, reinforcement learning, and sample complexity research.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Reversible Data Hiding With Secret Encrypted Image Sharing and Adaptive Coding

引用

IEEE Internet of Things Journal 2025年第13期12卷 23933-23945页

作者： Fang, Guangtian Wang, Feng Zhao, Chenbin Qin, Chuan Chang, Ching-Chun Chang, Chin-Chen Fujian University of Technology College of Computer Science and Mathematics Fuzhou350118 China Fujian University of Technology College of Computer Science and Mathematics Key Laboratory of Big Data Mining and Applications Fuzhou350118 China Monash University Faculty of IT Department of Software Systems and Cybersecurity VIC3800 Australia Ministry of Education School of Cyber Science and Engineering Key Laboratory of Aerospace Information Security and Trusted Computing Wuhan430072 China University of Shanghai for Science and Technology School of Optical-Electrical and Computer Engineering Shanghai200093 China Feng Chia University Information and Communication Security Research Center Taichuang 40724 Taiwan Feng Chia University Department of Information Engineering and Computer Science Taichung40724 Taiwan

To ensure the security of image information and facilitate efficient management in the cloud, the utilization of reversible data hiding in encrypted images (RDHEI) has emerged as pivotal. However, most existing RDHEI schemes suffer from lower security and limited embedding capacity. To tackle these challenges, we propose a reversible data hiding with secret encrypted image sharing and adaptive coding scheme. Specifically, in the encryption phase, we introduce an improved secret sharing (SS) encryption method based on the chinese remainder theorem for polynomials (CRTP). This method not only improves the security of encrypted images but also vacates a larger room for embedding. In the embedding phase, we introduce an adaptive coding embedding approach using XOR preservation (XORP) and huffman coding, which provides high embedding capacity. Experimental results and security analysis demonstrate that our proposed encryption method achieves optimal values in security indicators for encrypted images, such as information entropy, histograms, NPCR and UACI. The proposed embedding method is superior to some state-of-the-art schemes in terms of embedding capacity. Furthermore, in datasets BOSSBase and BOWS2, the average embedding rates of the proposed embedding approach can reach 2.1745 bpp and 2.0656 bpp, respectively. © 2025 IEEE.

关键词： Cryptography Encryption Security Data mining Adaptive coding Correlation Cloud computing Polynomials Image coding Streaming media

来源：评论

学校读者我要写书评

暂无评论

Autonomous Environmental Exploration and Target Tracking Control for Unmanned Surface Vehicles in Uncertain Environment

Autonomous Environmental Exploration and Target Tracking Con...

引用

2024 IEEE International Conference on Real-Time Computing and Robotics, RCAR 2024

作者： Zhu, Kaiwei Kong, Shihan Wu, Fang Sun, Jinan Yu, Junzhi College of Engineering Peking University State Key Laboratory for Turbulence and Complex Systems Department of Advanced Manufacturing and Robotics Beijing100871 China SPIC Nuclear Energy Co. Ltd. China Peking University National Engineering Research Center for Software Engineering Beijing100871 China

ISBN: (纸本)9798350372601

In this paper, a set of methods of autonomous obstacle exploration, avoiding, and target-tracking control in the uncertain environment is proposed. First, a secure and reliable obstacle exploration framework via multi-sensors information fusion is developed for USV to autonomously perceive obstacles in unknown environments. Second, a modified dynamic artificial potential field-based path planning method is proposed to dynamically generate a safe and feasible path even without complete prior obstacle position information. Third, a robust sliding mode control law with extended state observer is designed to track expectational target in the dynamical and uncertain environment. Finally, the target-tracking control law has been validated through the simulation experiments. Besides, the proposed methods of obstacle exploration, path planning, and target-tracking control are comprehensively evaluated through the simulation experiments in WEBOTS software, which is closer to the real world. © 2024 IEEE.

关键词： Target tracking

来源：评论

学校读者我要写书评

暂无评论

Generating Targeted Universal Adversarial Perturbation against Automatic Speech Recognition via Phoneme Tailoring

Generating Targeted Universal Adversarial Perturbation again...

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Zhang, Yujun Chen, Yanqu Wang, Jiakai Hu, Jin Tao, Renshuai Liu, Xianglong State Key Laboratory of Complex & Critical Software Environment Beihang University China School of Computer Science and Engineering Beihang University China College of Computer Science Beijing University of Technology China Zhongguancun Laboratory China School of Computer and Information Technology Beijing Jiaotong University China Institute of Data Space Hefei Comprehensive National Science Center China

ISBN: (纸本)9798350368741

There is a growing concern about adversarial attacks against automatic speech recognition (ASR) systems. Although research into targeted universal adversarial examples (AEs) has progressed, current methods are constrained by inefficient exploitation of audio features, demonstrating insufficient attack ability and robustness in the physical world. To solve this problem, we propose a phoneme-tailored attack (PTA) to improve the quality of the generated AEs. Specifically, to improve attack ability, we propose a Diverse Audio Composition Enrichment method, which enhances the utilization of audio features through phoneme-level slicing and recombination. To adapt AEs to complex environments, we propose a Natural Noise Pattern Guidance method to align AEs with natural noise patterns to improve their robustness. Experiments show that our method achieves an average accuracy of more than 72.34% and 98% with and without a norm constraint, and also demonstrates excellent performance in terms of generalization across datasets and resilience to MP3 compression. © 2025 IEEE.

关键词： Automatic speech recognition natural noise pattern phoneme-level targeted universal adversarial examples

来源：评论

学校读者我要写书评

暂无评论

A Dynamic Selective Parameter Sharing Mechanism Embedded with Multi-Level Reasoning Abstractions 26

A Dynamic Selective Parameter Sharing Mechanism Embedded wit...

引用

26th European Conference on Artificial Intelligence, ECAI 2023

作者： Liu, Yan He, Ying Ming, Zhong Yu, F. Richard Shenzhen Key Lab of Digital & Intelligent Technologies & Systems College of Computer Science and Software Engineering Shenzhen University China State Key Laboratory of Integrated Services Networks Xidian University China Shenzhen China

ISBN: (纸本)9781643684369

Cooperative multi-agent reinforcement learning (Co-MARL) commonly employs different parameter sharing mechanisms, such as full and partial sharing. However, imprudent application of these mechanisms can potentially constrain policy diversity and limit cooperation flexibility. Recent methods that group agents into distinct sharing categories often exhibit poor performance due to challenges in precisely differentiating agents and neglecting the issue of promoting cooperation among these categories. To address these issues, we introduce a dynamic selective parameter sharing mechanism embedded with multi-level reasoning abstractions (DSPS-MA). Our approach uses self-comparison sequences to infer agents' abstract concepts, defining the differences between agents and allowing them to dynamically select partners to share parameters based on these abstract concepts. We also design an intrinsic reward to offer comprehensive collaboration guidance for agents, and introduce a policy cosine similarity regularization term to ensure sufficient policy diversity. Empirical evaluations demonstrate that our approach yields higher returns and faster convergence than state-of-the-art methods. © 2023 The Authors.

关键词： Dynamics

来源：评论

学校读者我要写书评

暂无评论

BPH-YOLOv5: Improved yolov5 based on biformer prediction head for small target cigatette detection

BPH-YOLOv5: Improved yolov5 based on biformer prediction hea...

引用

2023 Jiangsu Annual Conference on Automation, JACA 2023

作者： Zheng, Xinyi Lu, Xiaobo College of Software Engineering Southeast University Suzhou215123 China School of Automation Southeast University Nanjing210096 China Key Laboratory of Measurement and Control of Complex Systems of Engineering Ministry of Education Southeast University Nanjing210096 China

ISBN: (纸本)9781839539848

Small object detection in the cigarette detection field is a recent popular task. As cigarette targets in surveillance are tiny, the object scale leads to a great challenge of object detection. To solve the issue mentioned above, we propose BPH-YOLOv5 based on YOLOv5. We add one more prediction head to detect different-scale objects, which is more sensitive to tiny objects using multi-scale features. Then we replace the original prediction heads with BiFormer Prediction Heads (BPH) to explore the prediction potential with an attention mechanism. Experiments showed that the mean average precision (mAP) of BPHYOLOv5 in the cigarette dataset reached 90.34%, which was 6.94% higher than that of YOLOv5-s prototype, 7.20% higher than that of YOLOv7, which are encouraging and competitive. Tests on the actual scenarios verified that the accuracy of smallscale target detection was significantly improved. © The Institution of engineering & Technology 2023.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

SegSdetr: Harnessing supervisions in SAM with detection label for efficient smoke detection

SegSdetr: Harnessing supervisions in SAM with detection labe...

引用

2023 Jiangsu Annual Conference on Automation, JACA 2023

作者： Li, Dan Lu, Xiaobo College of Software Engineering Southeast University Suzhou215123 China School of Automation Southeast University Nanjing210096 China Key Laboratory of Measurement and Control of Complex Systems of Engineering Ministry of Education Ministry of Education Nanjing210096 China

ISBN: (纸本)9781839539848

In recent years, forest fires often result in significant environmental and economic losses, spurred researchers to investigate early forest fire detection methods. Embedded devices with the object detection task can efficiently alarm early fire including smoke from a large field of view along with providing additional information such as location of the fire. But it's difficult to deploy large DNN in the chipsets in embedded devices. To satisfied the requirements for lightweight, this paper contributes to the existing research by proposing an efficient smoke detection model (SegSdetr) harnessing supervisions in Segment Anything Model (SAM) combined detection label. We lightweight the structure of model's backbone by solving linear redundancy observed in feature maps within models. And to compensate the drop accuracy from the lightweight, we have proposed a method harnessing the reference of SAM model. SegSdetr exhibits significantly superior precision and a 30% reduction in parameter efficiency compared to RT-DETR with the original backbone model. After incorporating our method, there is a further enhancement in precision. Additionally, our model outperforms the YOLOv8 family where a 3% increase in mAP50 with yolov8-n with the same Flops. © The Institution of engineering & Technology 2023.

关键词： Smoke

来源：评论

学校读者我要写书评

暂无评论

Instruct or Interact? Exploring and Eliciting LLMs' Capability in Code Snippet Adaptation Through Prompt engineering

Instruct or Interact? Exploring and Eliciting LLMs' Capabili...

引用

International Conference on software engineering (ICSE)

作者： Tanghaoran Zhang Yue Yu Xinjun Mao Shangwen Wang Kang Yang Yao Lu Zhang Zhang Yuxin Zhao College of Computer Science and Technology National University of Defense and Technology Changsha China Key Laboratory of Software Engineering for Complex System

ISBN: (数字)9798331505691

ISBN: (纸本)9798331505707

Code snippet adaptation is a fundamental activity in the software development process. Unlike code generation, code snippet adaptation is not a “free creation”, which requires developers to tailor a given code snippet in order to fit specific requirements and the code context. Recently, large language models (LLMs) have confirmed their effectiveness in the code generation task with promising results. However, their performance on code snippet adaptation, a reuse-oriented and context-dependent code change prediction task, is still unclear. To bridge this gap, we conduct an empirical study to investigate the performance and issues of LLMs on the adaptation task. We first evaluate the adaptation performances of three popular LLMs and compare them to the code generation task. Our result indicates that their adaptation ability is weaker than generation, with a nearly 15% decrease on pass@1 and more context-related errors. By manually inspecting 200 cases, we further investigate the causes of LLMs' sub-optimal performance, which can be classified into three categories, i.e., Unclear Requirement, Requirement Misalignment and Context Misapplication. Based on the above empirical research, we propose an interactive prompting approach to eliciting LLMs' ability on the adaptation task. Specifically, we enhance the prompt by enriching the context and decomposing the task, which alleviates context misapplication and improves requirement understanding. Besides, we enable LLMs' reflection by requiring them to interact with a human or a LLM counselor, compensating for unclear requirement. Our experimental result reveals that our approach greatly improve LLMs' adaptation performance. The best-performing Human-LLM interaction successfully solves 159 out of the 202 identified defects and improves the pass@1 and pass@5 by over 40% compared to the initial instruction-based prompt. Considering human efforts, we suggest multi-agent interaction as a trade-off, which can achieve comparable p

关键词： Bridges Codes Large language models Context awareness Reflection Prompt engineering software engineering software development management

来源：评论

学校读者我要写书评

暂无评论

Large Language Models are Few-Shot Summarizers: Multi-Intent Comment Generation via In-Context Learning

Large Language Models are Few-Shot Summarizers: Multi-Intent...

引用

International Conference on software engineering (ICSE)

作者： Mingyang Geng Shangwen Wang Dezun Dong Haotian Wang Ge Li Zhi Jin Xiaoguang Mao Xiangke Liao College of Computer Science National University of Defense Technology Changsha China Key Laboratory of Software Engineering for Complex Systems Key Lab of High Confidence Software Technology Peking University Beijing China

ISBN: (数字)9798400702174

ISBN: (纸本)9798350382143

Code comment generation aims at generating natural language descriptions for a code snippet to facilitate developers' program comprehension activities. Despite being studied for a long time, a bottleneck for existing approaches is that given a code snippet, they can only generate one comment while developers usually need to know information from diverse perspectives such as what is the functionality of this code snippet and how to use it. To tackle this limitation, this study empirically investigates the feasibility of utilizing large language models (LLMs) to generate comments that can fulfill developers' diverse intents. Our intuition is based on the facts that (1) the code and its pairwise comment are used during the pre-training process of LLMs to build the semantic connection between the natural language and programming language, and (2) comments in the real-world projects, which are collected for the pre-training, usually contain different developers' intents. We thus postulate that the LLMs can already understand the code from different perspectives after the pre-training. Indeed, experiments on two large-scale datasets demonstrate the rationale of our insights: by adopting the in-context learning paradigm and giving adequate prompts to the LLM (e.g., providing it with ten or more examples), the LLM can significantly outperform a state-of-the-art supervised learning approach on generating comments with multiple intents. Results also show that customized strategies for constructing the prompts and post-processing strategies for reranking the results can both boost the LLM's performances, which shed light on future research directions for using LLMs to achieve comment generation.

关键词： computer languages Codes Supervised learning Natural languages Semantics Task analysis software engineering

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：