检索结果-内蒙古大学图书馆

Fusion of infrared and visible images using nonsubsampled shearlet transform and block-based random image sampling

Revista Tecnica de la Facultad de Ingenieria Universidad del Zulia 2016年第7期39卷 337-347页

作者： Chen, Lili Guo, Hongjun Laboratory of Intelligent Information Processing Suzhou University Suzhou Anhui234000 China Key Laboratory of Intelligent Computing and Signal Processing MOE Anhui University Hefei Anhui230039 China

This paper presents a novel image fusion method based on nonsubsampled shearlet transform (NSST) and block-based random image sampling for infrared and visible images. In the method, the NSST is firstly performed on each of the source images to obtain their low frequency and high frequency subbands coefficients. Then the low frequency coefficients are fused by combing the block-based random image sampling and the sliding widow technique, while the high frequency coefficients are fused using classical absolute value maximum choosing rule. The fused image is reconstructed though performing the inverse NSST. The experiments are carried out on five pairs of infrared and visible images using five traditional fusion methods to verify the effectiveness and efficiency of the proposed method and by comparing the fused results visually and objectively it is demonstrated that the proposed fusion method is competitive or even superior to the other wellknown methods.

关键词： Image fusion

来源：评论

学校读者我要写书评

暂无评论

A CLASSIFICATION OF TWO-DISTANCE-TRANSITIVE CAYLEY GRAPHS OVER THE GENERALIZED QUATERNION GROUPS

arXiv

引用

arXiv 2025年

作者： Jin, Wei Li, Ping Shan Tan, Li School of Mathematics and Computational Science Key Laboratory of Intelligent Computing and Information Processing of Ministry of Education Xiangtan University Hunan Xiangtan411105 China School of Statistics and Data Science Jiangxi University of Finance and Economics Jiangxi Nanchang330013 China School of Mathematics and Statistics Key Laboratory of Intelligent Computing and Information Processing of Ministry of Education Xiangtan University Hunan Xiangtan411105 China

A non-complete graph is 2-distance-transitive if, for i = 1, 2 and for any two vertex pairs (u1, v1) and (u2, v2) with the same distance i in the graph, there exists an element of the graph automorphism group that maps (u1, v1) to (u2, v2). This is a generalization concept of the classical well-known distance-transitive graphs. In this paper, we completely determine the family of 2-distance-transitive Cayley graphs over the generalized quaternion groups. Copyright © 2025, The Authors. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Image enhancement method based on variable fuzzy sets

引用

Revista Tecnica de la Facultad de Ingenieria Universidad del Zulia 2016年第8期39卷 166-174页

作者： Chen, Lili Guo, Hongjun Pan, Zhenggao Laboratory of Intelligent Information Processing Suzhou University Suzhou Anhui234000 China Key Laboratory of Intelligent Computing and Signal Processing of MOE Anhui University Hefei Anhui230039 China

Image enhancement, as a means of digital image processing, often is fuzzy. Based on defects of excessive enhancement and insufficient detail enhancement existing in the traditional image enhancement method, this paper proposes a self-adaptive image enhancement method based on variable fuzzy sets, and such method can, according to the image's gray-scale properties, introduce the relative gray-scale level of the pixel as the fuzzy feature, and adopt the improved membership function to conduct the fuzzy enhancement, thus avoiding the loss of a great deal of gray-scale information after the enhancement. The paper also introduces the selection of optimal parameters, ensures the quality of the image enhanced and improves the feasibility and efficiency of this algorithm. This algorithm, on the basis of keeping the image's original brightness, enhances the image details at the same time. Experimental results show that such algorithm can obtain good visual effect and more obvious detail information.

关键词： Membership functions

来源：评论

学校读者我要写书评

暂无评论

A framework for Chinese event semantic constraint predicates and their generation using PEFT LLM and RAG

引用

International Journal of Machine Learning and Cybernetics 2025年

作者： Huang, Qiaojuan Wang, Shi He, Qing Cao, Cungen Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences Beijing100190 China University of Chinese Academy of Sciences Beijing100049 China

Event semantic analysis is a crucial area of research in natural language processing. It focuses on deeply understanding the semantics of events and their components. While existing semantic frameworks and event semantic labelling cover a wealth of semantic information, further research is needed to explore the fine-grained semantic relationships across participants in different events. To address this problem, we construct a head–tail event semantic constraint predicate framework that relates the complex semantics between different event components. However, considering the large-scale data, an automated method for generating constraint predicates is essential. Therefore, we combine parameter-efficient fine-tuning (PEFT) of large language models (LLMs) with retrieval-augmented generation (RAG) technologies. Specifically, this paper makes four significant contributions. Firstly, we construct a detailed framework of head–tail events semantic constraint predicates by combining manual definitions with the GPT-4o model. The framework includes 117 types of constraint predicates between head–tail events. Secondly, based on this framework, we annotate a dataset of head–tail events semantic constraint predicates and apply it to PEFT LLMs. Thirdly, we use PEFT LLMs prediction error data to construct an event knowledge base covering event entities, semantic roles, and entity types. Fourth, using the event knowledge base, the PEFT LLMs integrate the dual RAG techniques of keywords and vectors to improve the accuracy of generating the constraint predicates. Experimental results show that the method outperforms the baselines, achieving a precision of 97.18% and recall of 97.39%. This research fills the gap in semantic relationships between components across events and provides a novel approach for complex semantic generation tasks. © The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2025.

关键词： Domain Knowledge

来源：评论

学校读者我要写书评

暂无评论

DIVE: Inverting Conditional Diffusion Models for Discriminative Tasks

arXiv

引用

arXiv 2025年

作者： Li, Yinqi Chang, Hong Hou, Ruibing Shan, Shiguang Chen, Xilin Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences Beijing100190 China University of Chinese Academy of Sciences Beijing100049 China

Diffusion models have shown remarkable progress in various generative tasks such as image and video generation. This paper studies the problem of leveraging pretrained diffusion models for performing discriminative tasks. Specifically, we extend the discriminative capability of pretrained frozen generative diffusion models from the classification task [1], [2] to the more complex object detection task, by "inverting" a pretrained layout-to-image diffusion model. To this end, a gradient-based discrete optimization approach for replacing the heavy prediction enumeration process, and a prior distribution model for making more accurate use of the Bayes’ rule, are proposed respectively. Empirical results show that this method is on par with basic discriminative object detection baselines on COCO dataset. In addition, our method can greatly speed up the previous diffusion-based method [1], [2] for classification without sacrificing accuracy. Code and models are available at https://***/LiYinqi/DIVE. © 2025, CC BY-NC-ND.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Optimizing top-k retrieval:submodularity analysis and search strategies

引用

Frontiers of Computer Science 2016年第3期10卷 477-487页

作者： Chaofeng SHA Keqiang WANG Dell ZHANG Xiaoling WANG Aoying ZHOU School of Computer Science Shanghai Key Laboratory of Intelligent Information ProcessingFudan University Shanghai Key Laboratory of Trustworthy Computing East China Normal University Department of Computer Science and Information Systems BirkbeckUniversity of London

The key issue in top-fc retrieval,finding a set of fc documents(from a large document collection) that can best answer a user's query,is to strike the optimal balance between relevance and *** this paper,we study the top-fc retrieval problem in the framework of facility location analysis and prove the submodularity of that objective function which provides a theoretical approximation guarantee of factor 1--for the(best-first) greedy search ***,we propose a two-stage hybrid search strategy which first obtains a high-quality initial set of top-fc documents via greedy search,and then refines that result set iteratively via local *** on two large TREC benchmark datasets show that our two-stage hybrid search strategy approach can supersede the existing ones effectively and efficiently.

关键词： top-k retrieval diversification submodular function maximization

来源：评论

学校读者我要写书评

暂无评论

A sound and complete R-calculi with respect to contraction and minimal change

引用

Frontiers of Computer Science 2014年第2期8卷 184-191页

作者： Wei LI Yuefei SUI State Key Laboratory of Software Development Environment Beihang University Beijing 100191 China Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences Beijing 100190 China

AGM postulates are for belief revision （revision by a single belief）, and DP postulates are for iterated revision （revision by a finite sequence of beliefs）. R-calculus is given for R-configurations △｜Г, where △ is a set of atomic formulas or the negations of atomic formulas, and Г is a finite set of formulas. We shall give two R-calculi C and M （sets of de- duction rules） such that for any finite consistent sets Г, △of formulas in the propositional logic, there is a consistent set ⊙ Г C of formulas such that △IГ → △, ⊙ is provable and⊙ is a contraction of F by A or a minimal change of F by A; and prove that C and M are sound and complete with respect to the contraction and the minimal change, respectively.

关键词： belief revision R-calculus minimal change,contraction

来源：评论

学校读者我要写书评

暂无评论

A Survey of Commonsense Knowledge Acquisition

引用

Journal of Computer Science & Technology 2013年第4期28卷 689-719页

作者：臧良俊曹聪曹亚男吴昱明曹存根 Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences University of the Chinese Academy of Sciences The Second Lab Institute of Information Engineering Chinese Academy of Sciences

Collecting massive commonsense knowledge （CSK） for commonsense reasoning has been a long time standing challenge within artificial intelligence research. Numerous methods and systems for acquiring CSK have been developed to overcome the knowledge acquisition bottleneck. Although some specific commonsense reasoning tasks have been presented to allow researchers to measure and compare the performance of their CSK systems, we compare them at a higher level from the following aspects： CSK acquisition task （what CSK is acquired from where）, technique used （how can CSK be acquired）, and CSK evaluation methods （how to evaluate the acquired CSK）. In this survey, we first present a categorization of CSK acquisition systems and the great challenges in the field. Then, we review and compare the CSK acquisition systems in detail. Finally, we conclude the current progress in this field and explore some promising future research issues.

关键词： commonsense knowledge knowledge acquisition knowledge representation and reasoning

来源：评论

学校读者我要写书评

暂无评论

Improving mutation pathogenicity prediction of metal-binding sites in proteins with a panoramic attention mechanism

引用

Pattern Recognition 2025年

作者： Yuan Zhang Jiafeng Wu Qiuye Zhao Mingyuan Dong Junsheng Deng Xieping Gao Kai Hu Dapeng Xiong Key Laboratory of Intelligent Computing and Information Processing of Ministry of Education Xiangtan University Xiangtan 411105 China Department of Computational Biology Cornell University Ithaca NY 14853 USA Weill Institute for Cell and Molecular Biology Cornell University Ithaca NY 14853 USA Hunan Provincial Key Laboratory of Intelligent Computing and Language Information Processing Hunan Normal University Changsha 410081 China

Accurately identifying the pathogenicity of mutations at protein-metal binding sites is crucial for uncovering the structural and functional complexity of metalloproteins, and the molecular mechanisms behind numerous diseases. In this study, we present a novel deep learning framework, CASTLE, for the accurate pathogenicity prediction of mutations at metal-binding sites through an effective depiction of the message-passing in local environment surrounding both the metals and metal-binding sites. Specifically, CASTLE weaves multiple attention-driven units to construct comprehensive panoramic message-passing paths, enabling the capture of intricate structural patterns associated with the metal-binding conformations. In addition, CASTLE seamlessly integrates structural information at both residue and atomic levels, and further deeply fuses structural and sequence representations to ensure the incorporation of more comprehensive information. We demonstrate that CASTLE significantly outperforms other state-of-the-art methods across all datasets we evaluated, showcasing its robustness and generalization abilities. Our interpretability analysis illustrates CASTLE's capability in capturing meaningful representations from the metal coordinate-dependent environments. Moreover, the model optimization reaffirms the advantages of our model building strategy, which can effectively capture distinct binding patterns for different metal types. Overall, CASTLE provides a powerful deep learning tool that may offer valuable insights into the study of metalloprotein-related disease mechanisms and drug design.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Distilling Knowledge from Heterogeneous Architectures for Semantic Segmentation

arXiv

引用

arXiv 2025年

作者： Huang, Yanglin Hu, Kai Zhang, Yuan Chen, Zhineng Gao, Xieping Key Laboratory of Intelligent Computing and Information Processing Ministry of Education Xiangtan University China School of Computer Science Fudan University China Key Laboratory for Artificial Intelligence and International Communication Hunan Normal University China

Current knowledge distillation (KD) methods for semantic segmentation focus on guiding the student to imitate the teacher’s knowledge within homogeneous architectures. However, these methods overlook the diverse knowledge contained in architectures with different inductive biases, which is crucial for enabling the student to acquire a more precise and comprehensive understanding of the data during distillation. To this end, we propose for the first time a generic knowledge distillation method for semantic segmentation from a heterogeneous perspective, named HeteroAKD. Due to the substantial disparities between heterogeneous architectures, such as CNN and Transformer, directly transferring cross-architecture knowledge presents significant challenges. To eliminate the influence of architecture-specific information, the intermediate features of both the teacher and student are skillfully projected into an aligned logits space. Furthermore, to utilize diverse knowledge from heterogeneous architectures and deliver customized knowledge required by the student, a teacher-student knowledge mixing mechanism (KMM) and a teacher-student knowledge evaluation mechanism (KEM) are introduced. These mechanisms are performed by assessing the reliability and its discrepancy between heterogeneous teacher-student knowledge. Extensive experiments conducted on three main-stream benchmarks using various teacher-student pairs demonstrate that our HeteroAKD outperforms state-of-the-art KD methods in facilitating distillation between heterogeneous architectures. Copyright © 2025, The Authors. All rights reserved.

关键词： Students

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：