检索结果-内蒙古大学图书馆

CL-BioGAN: Biologically-Inspired Cross-Domain Continual Learning for Hyperspectral Anomaly Detection

IEEE Transactions on Geoscience and Remote Sensing 2025年 63卷

作者： Wang, Jianing Hua, Zheng Zhang, Wan Hao, Shengjia Yao, Yuqiong Gong, Maoguo Xidian University Key Laboratory of Collaborative Intelligence Systems Ministry of Education School of Computer Science and Technology Xi’an710071 China Xidian University Key Laboratory of Intelligent Perception and Image Understanding of Ministry of Education of China School of Artificial Intelligence Xi’an710071 China Xidian University Key Laboratory of Collaborative Intelligence Systems Ministry of Education Xi’an710071 China

Memory stability and learning flexibility in continual learning (CL) is a core challenge for cross-scene Hyperspectral Anomaly Detection (HAD) task. Biological neural networks can actively forget history knowledge that conflicts with the learning of new experiences by regulating learning-triggered synaptic expansion and synaptic convergence. Inspired by this phenomenon, we propose a novel Biologically-Inspired Continual Learning Generative Adversarial Network (CL-BioGAN) for augmenting continuous distribution fitting ability for cross-domain HAD task, where Continual Learning Bio-inspired Loss (CL-Bio Loss) and self-attention Generative Adversarial Network (BioGAN) are incorporated to realize forgetting history knowledge as well as involving replay strategy in the proposed BioGAN. Specifically, a novel Bio-Inspired Loss composed with an Active Forgetting Loss (AF Loss) and a CL loss is designed to realize parameters releasing and enhancing between new task and history tasks from a Bayesian perspective. Meanwhile, BioGAN loss with L2-Norm enhances self-attention (SA) to further balance the stability and flexibility for better fitting background distribution for open scenario HAD (OHAD) tasks. Experiment results underscore that the proposed CL-BioGAN can achieve more robust and satisfying accuracy for cross-domain HAD with fewer parameters and computation cost. This dual contribution not only elevates CL performance but also offers new insights into neural adaptation mechanisms in OHAD task. © 1980-2012 IEEE.

关键词： Active forgetting Continual learning GAN Hyperspectral anomaly detection

来源：评论

学校读者我要写书评

暂无评论

Meta knowledge assisted Evolutionary Neural Architecture Search

引用

IEEE Transactions on Circuits and Systems for Video Technology 2025年

作者： Li, Yangyang Liu, Guanlong Shang, Ronghua Jiao, Licheng Xidian University Key Laboratory of Intelligent Perception and Image Understanding of Ministry of Education Joint International Research Laboratory of Intelligent Perception and Computation International Research Center for Intelligent Perception and Computation Collaborative Innovation Center of Quantum Information of Shaanxi Province School of Artificial Intelligence Xi’an710071 China

Evolutionary computation (EC)-based neural architecture search (NAS) has achieved remarkable performance in the automatic design of neural architectures. However, the high computational cost associated with evaluating searched architectures poses a challenge for these methods, and a fixed form of learning rate (LR) schedule means greater information loss on diverse searched architectures. This paper introduces an efficient EC-based NAS method to solve these problems via an innovative meta-learning framework. Specifically, a meta-learning-rate (Meta-LR) scheme is used through pretraining to obtain a suitable LR schedule, which guides the training process with lower information loss when evaluating each individual. An adaptive surrogate model is designed through an adaptive threshold to select the potential architectures in a few epochs and then evaluate the potential architectures with complete epochs. Additionally, a periodic mutation operator is proposed to increase the diversity of the population, which enhances the generalizability and robustness. Experiments on CIFAR-10, CIFAR-100, and imageNet1K datasets demonstrate that the proposed method achieves high performance comparable to that of many state-of-the-art peer methods, with lower computational cost and greater robustness. © 1991-2012 IEEE.

关键词： evolutionary computation meta-learning Neural architecture search surrogate model

来源：评论

学校读者我要写书评

暂无评论

A Distributed Hybrid Quantum Convolutional Neural Network for Medical image Classification

arXiv

引用

arXiv 2025年

作者： Li, Yangyang Qi, Zhengya Li, Yuelin Yang, Haorui Shang, Ronghua Jiao, Licheng Key Laboratory of Intelligent Perception and Image Understanding of Ministry of Education Joint International Research Laboratory of Intelligent Perception and Computation International Research Center for Intelligent Perception and omputation Collaborative Innovation Center of Quantum Information of Shaanxi Province School of Artificial Intelligence Xidian University No. 2 Taibai South Road Yanta District Shaanxi Xi’an710071 China

Medical images are characterized by intricate and complex features, requiring interpretation by physicians with medical knowledge and experience. Classical neural networks can reduce the workload of physicians, but can only handle these complex features to a limited extent. Theoretically, quantum computing can explore a broader parameter space with fewer parameters, but it is currently limited by the constraints of quantum hardware. Considering these factors, we propose a distributed hybrid quantum convolutional neural network based on quantum circuit splitting. This model leverages the advantages of quantum computing to effectively capture the complex features of medical images, enabling efficient classification even in resource-constrained environments. Our model employs a quantum convolutional neural network (QCNN) to extract high-dimensional features from medical images, thereby enhancing the model’s expressive capability. By integrating distributed techniques based on quantum circuit splitting, the 8-qubit QCNN can be reconstructed using only 5 qubits. Experimental results demonstrate that our model achieves strong performance across 3 datasets for both binary and multiclass classification tasks. Furthermore, compared to recent technologies, our model achieves superior performance with fewer parameters, and experimental results validate the effectiveness of our model. Copyright © 2025, The Authors. All rights reserved.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Enhancing Cross-Dataset EEG Emotion Recognition: A Novel Approach with Emotional EEG Style Transfer Network

引用

IEEE Transactions on Affective Computing 2025年

作者： Zhou, Yijin Li, Fu Li, Yang Ji, Youshuo Zhang, Lijian Chen, Yuanfang Wang, Huaning Xidian University Key Laboratory of Intelligent Perception and Image Understanding of Ministry of Education School of Artificial Intelligence Xi'an710071 China Beijing Institute of Mechanical Equipment Beijing100000 China Fourth Military Medical University Department of Psychiatry at Xijing Hospital Xi'an China

Electroencephalogram (EEG)-based emotion recognition has achieved remarkable success in both subject-dependent and subject-independent scenarios. However, overcoming the challenges associated with reduced performance in EEG emotion recognition across devices, time, space, and subjects (i.e., cross-dataset) remains a significant obstacle for affective brain-computer interfaces (aBCIs). The key issue lies in the distributional mismatch between source and target domain EEG signals. To tackle the significant inter-domain differences in cross-dataset EEG emotion recognition, this paper introduces an innovative framework termed the Emotional EEG Style Transfer Network (E2STN), which aims to effectively capture the emotional content information from the source domain and the style features from the target domain, facilitating the reconstruction of stylized emotion EEG representations. These stylized EEG representations significantly enhance the discriminative prediction performance in cross-dataset EEG emotion recognition. Specifically, E2STN consists of three key modules: a Transfer Module for domain style transfer, a Transfer Evaluation Module for evaluating transfer quality, and a Discriminative Module for making discriminative predictions. Extensive experiments demonstrate that EE2STN achieves state-of-the-art performance in cross-dataset emotion EEG recognition. To the best of our knowledge, this is the first work to explicitly address cross-dataset emotion EEG recognition. The experimental results provide a valuable benchmark for future research in this area. The associated code is publicly available at https://***/zhouyijin1998/EESTN. © 2010-2012 IEEE.

关键词： Electroencephalography

来源：评论

学校读者我要写书评

暂无评论

Graph-based Adaptive Network With Spatial-Spectral Features for Hyperspectral Unmixing

引用

IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 2025年 18卷 12865-12881页

作者： Dong, Hua Zhang, Xiaohua Zhang, Jinhua Meng, Hongyun Jiao, Licheng Xidian University Key Laboratory of Intelligent Perception and Image Understanding of Ministry of Education School of Artificial Intelligence Xi'an 710126 China Xidian University School of Mathematics and Statistics Xi'an 710126 China

Hyperspectral unmixing aims to extract basic material (endmember) spectra and estimate their corresponding fractions (abundances) from observed pixels in hyperspectral images (HSIs). Recently, blind unmixing methods based on Autoencoders (AEs) have gained significant attention due to their capability to simultaneously obtain endmembers and abundances, as well as their strong performance. Also, many AE-based unmixing methods have explored the use of spatial information in HSIs to enhance unmixing performance. However, most of these methods fail to effectively integrate both local and global spatial-spectral features and lack adaptive selection and constraints for the extracted features. To address this issue, this paper proposes an adaptive spatial-spectral unmixing method based on Graph. In the method, HSIs are treated as data on manifold structures, with superpixels serving as graph nodes to construct a global graph-structured data. We then use Graph Attention Network (GAT) to learn superpixel-level global spatial-spectral feature representations from this graph-structured data. Since the spatial-spectral feature learned by GAT is at the superpixel level, it is impossible to unmix each pixel using only this feature. Thus, we integrate a Convolutional Neural Network (CNN) to learn local discriminative spatial-spectral features. Attention mechanisms are further employed to select the features most relevant to unmixing. Additionally, An Adaptive Abundance Constraint Module is proposed to dynamically enforce constraints on the extracted features. Experiments conducted on synthetic and real hyperspectral images demonstrate the efficiency and superior performance of our proposed method compared to several existing unmixing approaches. © 2008-2012 IEEE.

关键词： Abundance regularity constraints Autoencoder Graph attention network Hyperspectral unmixing

来源：评论

学校读者我要写书评

暂无评论

Robust Motion-Guided Frame Sampler with Interpretive Evaluation for Video Action Recognition

引用

IEEE Transactions on Mobile Computing 2025年

作者： Bai, Jing Zhang, Yuxiang Wang, Yiran Xiao, Zhu Xiong, Yong Jiao, Licheng Xidian University School of Artificial Intelligence Key Laboratory of Intelligent Perception and Image Understanding Ministry of Education Xi'an710071 China Hunan University College of Computer Science and Electronic Engineering Changsha410082 China Hunan Lianzhi Technology Co. Ltd Changsha410200 China

Due to the presence of redundancy and interference, frame sampling is a promising but challenging solution to mitigate the expensive computation of video action recognition. Although the motion prior has shown great potential for frame selection, existing motion-based strategies suffer from limitations in terms of robustness and interpretive evaluation. In this paper, we devise a robust frame sampling strategy called positive motion guided sampler (PMGSampler). It consists of two procedures, local motion capture and global motion statistics. At the local level, we propose two concepts about inter-frame motion amplitude and motion continuity, which helps to perceive the movement of subjects and identify abnormal events that may generate negative pseudo-motion information. Then, through a global analysis of the obtained local motions, the sampler becomes more sensitive to informative frames and robust to outliers. The proposed sampler can be applied to most existing models for improving recognition accuracy. We conduct extensive experiments on four widely-used benchmarks to demonstrate the superiority of our PMGSampler over other methods of the same type. In addition, to analyse how sampled frames influence action recognition, we present a visual interpretation method for video models, termed as spatio-temporal class activation map (STCAM). By introducing spatial and temporal branches, our STCAM is able to visualise the salience of spatio-temporal features. With the help of STCAM, we can further intuitively evaluate the performance of different sampling strategies. Code is available at https://***/zyx-cv/PMGSampler. © 2002-2012 IEEE.

关键词： Motion capture

来源：评论

学校读者我要写书评

暂无评论

Integrating Language-image Prior into EEG Decoding for Cross-Task Zero-Calibration RSVP-BCI

arXiv

引用

arXiv 2025年

作者： Li, Xujin Wei, Wei Qiu, Shuang Zhang, Xinyi Li, Fu He, Huiguang The Key Laboratory of Brain Cognition and Brain-inspired Intelligence Technology Institute of Automation Chinese Academy of Sciences Beijing100190 China Beijing100049 China The Key Laboratory of Intelligent Perception and Image Understanding of Ministry of Education The School of Artificial Intelligence Xidian University Xi’an710071 China

Rapid Serial Visual Presentation (RSVP)-based Brain-Computer Interface (BCI) is an effective technology used for information detection by detecting Event-Related Potentials (ERPs). The current RSVP decoding methods can perform well in decoding EEG signals within a single RSVP task, but their decoding performance significantly decreases when directly applied to different RSVP tasks without calibration data from the new tasks. This limits the rapid and efficient deployment of RSVP-BCI systems for detecting different categories of targets in various scenarios. To overcome this limitation, this study aims to enhance the cross-task zero-calibration RSVP decoding performance. First, we design three distinct RSVP tasks for target image retrieval and build an open-source dataset containing EEG signals and corresponding stimulus images. Then we propose an EEG with Language-image Prior fusion Transformer (ELIPformer) for cross-task zero-calibration RSVP decoding. Specifically, we propose a prompt encoder based on the language-image pre-trained model to extract language-image features from task-specific prompts and stimulus images as prior knowledge for enhancing EEG decoding. A cross bidirectional attention mechanism is also adopted to facilitate the effective feature fusion and alignment between the EEG and language-image features. Extensive experiments demonstrate that the proposed model achieves superior performance in cross-task zero-calibration RSVP decoding, which promotes the RSVP-BCI system from research to practical application. 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

Multi-Task Multi-Scale Feature Selection for Point Cloud Registration

引用

IEEE Transactions on Evolutionary Computation 2025年

作者： Wu, Yue Luo, Chuang Gong, Maoguo Ding, Hangqi Sheng, Jinlong Miao, Qiguang Li, Hao Ma, Wenping He, Hao Xidian University School of Computer Science and Technology Xi'an710071 China Xidian University School of Electronic Engineering Key Laboratory of Collaborative Intelligence Systems Ministry of Education Xi'an710071 China Inner Mongolia Normal University College of Mathematics Science Hohhot010028 China Xidian University School of Artificial Intelligence Key Laboratory of Intelligent Perception and Image Understanding Ministry of Education Xi'an710071 China

3D point cloud registration is a process of solving the geometric transformation between two point clouds. This process is an important issue in computer vision and pattern recognition. The registration methods based on geometric features are highly sensitive to the scale of feature extraction. Changes in scale can introduce inaccuracies in feature descriptions, thereby compromising the reliability of the registration results. To mitigate the impact of feature scale on the outcomes and the high-dimensional issue arising from features of different scales, we propose a method for multi-scale point cloud feature selection. We solve the high-dimensional problem of feature selection by designing a multi-task framework. By designing a mutual information dimensionality reduction method, we decomposed the high-dimensional feature selection task of different descriptors with multi-scale features into multiple related low-dimensional feature selection tasks. Then, by means of the knowledge transfer among these low-dimensional feature selection tasks, we sought the best feature subset to obtain more robust feature information. We evaluate the effectiveness of our method by conducting extensive experiments on various datasets. The experimental results show that the method outperforms other feature descriptors in terms of descriptive power and robustness and improves the effectiveness of point cloud registration. © 1997-2012 IEEE.

关键词： Feature Selection

来源：评论

学校读者我要写书评

暂无评论

Unsupervised Entity Alignment Based on Personalized Discriminative Rooted Tree

arXiv

引用

arXiv 2025年

作者： Yang, Yaming Wang, Zhe Guan, Ziyu Zhao, Wei Huang, Xinyan He, Xiaofei State Key Laboratory of Integrated Services Networks School of Computer Science and Technology Xidian University Xi’an710071 China Key Laboratory of Intelligent Perception and Image Understanding of Ministry of Education School of Artificial Intelligence Xidian University Xi’an710071 China State Key Laboratory of CAD&CG Zhejiang University Hangzhou China

Entity Alignment (EA) is to link potential equivalent entities across different knowledge graphs (KGs). Most existing EA methods are supervised as they require the supervision of seed alignments, i.e., manually specified aligned entity pairs. Very recently, several EA studies have made some attempts to get rid of seed alignments. Despite achieving preliminary progress, they still suffer two limitations: (1) The entity embeddings produced by their GNN-like encoders lack personalization since some of the aggregation subpaths are shared between different entities. (2) They cannot fully alleviate the distribution distortion issue between candidate KGs due to the absence of the supervised signal. In this work, we propose a novel unsupervised entity alignment approach called UNEA to address the above two issues. First, we parametrically sample a tree neighborhood rooted at each entity, and accordingly develop a tree attention aggregation mechanism to extract a personalized embedding for each entity. Second, we introduce an auxiliary task of maximizing the mutual information between the input and the output of the KG encoder, to regularize the model and prevent the distribution distortion. Extensive experiments show that our UNEA achieves a new state-of-the-art for the unsupervised EA task, and can even outperform many existing supervised EA baselines. Copyright © 2025, The Authors. All rights reserved.

关键词： Knowledge graph

来源：评论

学校读者我要写书评

暂无评论

Learning Inverse Laplacian Pyramid for Progressive Depth Completion

arXiv

引用

arXiv 2025年

作者： Wang, Kun Yan, Zhiqiang Fan, Junkai Li, Jun Yang, Jian PCA Lab 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 School of Computer Science and Engineering Nanjing University of Science and Technology Nanjing China

Depth completion endeavors to reconstruct a dense depth map from sparse depth measurements, leveraging the information provided by a corresponding color image. Existing approaches mostly hinge on single-scale propagation strategies that iteratively ameliorate initial coarse depth estimates through pixel-level message passing. Despite their commendable outcomes, these techniques are frequently hampered by computational inefficiencies and a limited grasp of scene context. To circumvent these challenges, we introduce LP-Net, an innovative framework that implements a multi-scale, progressive prediction paradigm based on Laplacian Pyramid decomposition. Diverging from propagation-based approaches, LP-Net initiates with a rudimentary, low-resolution depth prediction to encapsulate the global scene context, subsequently refining this through successive upsampling and the reinstatement of high-frequency details at incremental scales. We have developed two novel modules to bolster this strategy: 1) the Multi-path Feature Pyramid module, which segregates feature maps into discrete pathways, employing multi-scale transformations to amalgamate comprehensive spatial information, and 2) the Selective Depth Filtering module, which dynamically learns to apply both smoothness and sharpness filters to judiciously mitigate noise while accentuating intricate details. By integrating these advancements, LP-Net not only secures state-of-the-art (SOTA) performance across both outdoor and indoor benchmarks such as KITTI, NYUv2, and TOFDC, but also demonstrates superior computational efficiency. At the time of submission, LP-Net ranks 1st among all peer-reviewed methods on the official KITTI leaderboard. The source code will be made publicly accessible upon paper acceptance. © 2025, CC BY-NC-SA.

关键词： Laplace transforms

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：