检索结果-内蒙古大学图书馆

Shape Embedding and Knowledge Mining Network for Generalized Few-Shot Remote Sensing Segmentation

IEEE Geoscience and Remote Sensing Letters 2025年 22卷

作者： Qiu, Zifeng Liu, Hongyu Xiong, Hang Di, Chengliang Fang, Hao Cong, Runmin Beijing Jiaotong University Institute of Information Science Beijing100044 China The54th Research Institute of CETC Shijiazhuang050081 China China Communication System Co. Ltd. Hebei Branch Shijiazhuang050081 China Shandong University School of Control Science and Engineering Jinan250061 China Ministry of Education Key Laboratory of Machine Intelligence and System Control Jinan250061 China Beijing Key Laboratory of Advanced Information Science and Network Technology Beijing100044 China

In recent years, generalized few-shot segmentation (GFSS) has received widespread attention from scholars by virtue of its superiority in low-data regimes. Most of the existing research focuses on natural image processing, and few studies have been devoted to the practical but challenging topic of remote sensing image (RSI) understanding. In this paper, we propose a Shape Embedding and Knowledge Mining Network (SKNet) for generalized few-shot RSI segmentation. Specifically, the framework is divided into two key stages: (a) In the base class learning stage, shape representation embedding is introduced to enhance the network's ability to perceive remote sensing objects. Simultaneously, we introduce the self-reconstruction constraint to prevent new unseen classes from merging, thereby improving the representation uniqueness of these classes. (b) In the novel class learning stage, a base class knowledge mining mechanism is designed to update the prototypes of the novel class by using the prototype representation of the base class, so as to enhance the discrimination ability of the network. We validated our methods on the adapted version of OpenEarthMap and iSAID datasets. In comparison with existing GFSS methods, the proposed approach demonstrates an advancement. © 2004-2012 IEEE.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

Selective ensemble of RBFNNs based on improved negative correlation learning 13

Selective ensemble of RBFNNs based on improved negative corr...

引用

13th International Conference on machine learning and Cybernetics, ICMLC 2014

作者： Xing, Hongjie Liu, Lifei Li, Sen Key Laboratory of Machine Learning and Computational Intelligence College of Mathematics and Computer Science Hebei University Baoding Hebei Province071002 China

ISBN: (纸本)9783662456514

In this paper, a novel selective ensemble method based on the improved negative correlation learning is proposed. To make the proposed ensemble strategy more robust against noise, correntropy is utilized to substitute mean square error (MSE). Moreover, an L1-norm based regularization term of ensemble weights is incorporated into the objective function of the proposed ensemble strategy to fulfill the task of selective ensemble. The half-quadratic optimization technique and the surrogate function method are used to solve the optimization problem of the proposed ensemble strategy. Experimental results on two synthetic data sets and the five benchmark data sets demonstrate that the proposed method is superior to the single radial basis function neural network (RBFNN). © Springer-Verlag Berlin Heidelberg 2014.

关键词： Mean square error

来源：评论

学校读者我要写书评

暂无评论

A Novel Attention Model of Deep learning in Image Classification 11th

A Novel Attention Model of Deep Learning in Image Classifica...

引用

11th International Symposium on Parallel Architectures, Algorithms and Programming, PAAP 2020

作者： Hua, Qiang Chen, Liyou Li, Pan Zhao, Shipeng Li, Yan Key Laboratory of Machine Learning and Computational Intelligence of Hebei Province College of Mathematics and Information Science Hebei University Baoding071002 China

ISBN: (纸本)9789811600098

As the neural network becomes more and more complex, a large number of parameters are to be adjusted and more unrelated information will be generated, which is more time-consuming in model training and affects the model performance. The attention mechanic is often used to address this problem in the literature. However, the current attention models only consider the correlation within the pixel domain and channel domain individually, and the computational complexity is comparatively high. This paper proposes a new type of attention module called Pixel-wise And Channel-wise attention (PAC attention module for short) in which the hybrid correlation between the pixel and channel domain is also considered. Without increasing the number of parameters, this module can be used to calculate the correlation of the convolution feature map at any position in any layer of the convolutional neural network to achieve end-to-end training. In the experiments, the proposed PAC attention module is used in typical network structures in deep learning and has been compared to current models on some benchmark big datasets. © 2021, Springer Nature Singapore Pte Ltd.

关键词： Image classification

来源：评论

学校读者我要写书评

暂无评论

Correntropy based self-organizing map

Correntropy based self-organizing map

引用

2016 International Conference on machine learning and Cybernetics, ICMLC 2016

作者： Shang, Qing-Zhen Xing, Hong-Jie Key Laboratory of Machine Learning and Computational Intelligence College of Mathematics and Information Science Hebei University Baoding Hebei Province071002 China

ISBN: (纸本)9781509003891

Self-organizing map (SOM) is regarded as a type of feedfoward neural network. It has been successfully used for unsupervised learning. However, the objective function of the traditional SOM relies on the mean squared error (MSE) criterion, which makes the performance of SOM become poor in the presence of noise. In the paper, correntropy based measure is proposed to substitute MSE to enhance the anti-noise ability of SOM. Moreover, the half-quadratic optimization technique is utilized to deal with the corresponding optimization problem. Experimental results demonstrate that the proposed method achieves better anti-noise performance in comparison with the traditional SOM. © 2016 IEEE.

关键词： Self organizing maps

来源：评论

学校读者我要写书评

暂无评论

Modified AdaBoost based OCSVM ensemble for image retrieval

Modified AdaBoost based OCSVM ensemble for image retrieval

引用

2012 International Conference on machine learning and Cybernetics, ICMLC 2012

作者： Xing, Hong-Jie Wu, Jian-Guo Chen, Xue-Fang Key Laboratory of Machine Learning and Computational Intelligence College of Mathematics and Computer Science Hebei University Baoding 071002 Hebei Province China

ISBN: (纸本)9781467314855

For the traditional content-based image retrieval system, the number of irrelevant images for a given query image is significantly more than that of relevant images in an image repository. Therefore, the numbers of negative samples and positive samples are highly unbalanced, which makes the traditional binary classifiers ineffective. In this paper, our proposed modified AdaBoost-based one-class support vector machine (OCSVM) ensemble is utilized to deal with the aforesaid problem. In our proposed method, the weight update formula of training data for AdaBoost is modified to make AdaBoost fit for combining the results of OCSVMs even though OCSVM is regarded as a strong classifier. Compared with the other three related methods, our proposed approach exhibits better performance on the three benchmark image databases. © 2012 IEEE.

关键词： Adaptive boosting

来源：评论

学校读者我要写书评

暂无评论

Selective ensemble of support vector data descriptions for novelty detection

Selective ensemble of support vector data descriptions for n...

引用

9th International Symposium on Neural Networks, ISNN 2012

作者： Xing, Hong-Jie Chen, Xue-Fang Key Laboratory of Machine Learning and Computational Intelligence College of Mathematics and Computer Science Hebei University Baoding 071002 Hebei Province China

ISBN: (纸本)9783642313455

Since support vector data description (SVDD) is regarded as a strong classifier, the traditional ensemble methods are not fit for directly combining the results of several SVDDs. Moreover, as is well-known, when many trained classifiers are available, it is better to ensemble some of them rather than all. In this paper, a selective ensemble method based on correntropy is proposed to deal with the foresaid problems. The base classifier used in the proposed ensemble is SVDD. Experimental results on two synthetic data sets and five benchmark data sets demonstrate that the proposed method is superior to its related approaches. © 2012 Springer-Verlag.

关键词： Data description

来源：评论

学校读者我要写书评

暂无评论

Robust smooth one-class support vector machine 2

Robust smooth one-class support vector machine

引用

2nd International Conference on Information Technology and Electronic Commerce, ICITEC 2014

作者： Hu, Jin-Kou Xing, Hong-Jie Key Laboratory of Machine Learning and Computational Intelligence College of Mathematics and Information Science Hebei University Baoding Hebei Province071002 China

ISBN: (纸本)9781479952984

In this paper, a novel one-class classification approach, namely, robust smooth one-class support vector machine (RSOCSVM) is proposed. The proposed method can efficiently enhance the anti-noise ability of the traditional one-class support vector machine (OCSVM). Utilizing the smooth technique, RSOCSVM reformulates the quadratic programming problem of OCSVM as an unstrained optimization format. Moreover, half-quadratic minimization is used to solve the obtained unstrained optimization problem. Experimental results on two synthetic data sets and nine benchmark data sets demonstrate that the proposed method is superior to the traditional OCSVM. © 2014 IEEE.

关键词： Support vector machines

来源：评论

学校读者我要写书评

暂无评论

Linear discriminant analysis based on Zp-norm maximization 2

Linear discriminant analysis based on Zp-norm maximization

引用

2nd International Conference on Information Technology and Electronic Commerce, ICITEC 2014

作者： An, Lei-Lei Xing, Hong-Jie Key Laboratory of Machine Learning and Computational Intelligence College of Computer Science and Technology Hebei University Baoding Hebei Province071002 China

ISBN: (纸本)9781479952984

In this paper, linear discriminant analysis (LDA) based on Lp-norm (LDA-Lp) optimization method is proposed. The objective function utilizing the Lp-norm with arbitrary p value is studied. By maximizing the Lp-norm-based ratio between the between-class scatter and the within-class scatter, LDA-Lp can construct a set of local optimal projection vectors. Moreover, the optimal projection vectors can be obtained by the gradient ascent method. Experimental results on the two synthetic and fourteen benchmark datasets demonstrate that the better performance of LDA-Lp can be achieved by choosing the optimal value of p. © 2014 IEEE.

关键词： Benchmarking

来源：评论

学校读者我要写书评

暂无评论

Image classification by combining local and mid-level features 18

Image classification by combining local and mid-level featur...

引用

2nd International Conference on Innovation in Artificial intelligence, ICIAI 2018

作者： Lu, Yao Zhang, Hui Key Lab. of Machine Learning and Computational Intelligence College of Mathematics and Information Science Hebei University Hebei China

ISBN: (纸本)9781450363457

It is meaningful to study high performance image classification algorithms for massive image management and effective organization. Image feature representations directly affect the performance of classification algorithm, however, in the current popular sparse coding framework for image classification, the relationship between local features is neglected, and there is still considerable room left for further improvement. In this paper, mid-level features are extracted efficiently in the superpixel region, which embeds the structural information of local features. Then we combine the local features and mid-level features to achieve better discriminative ability under the sparse coding framework. Compared to using only local features, the new methods improve the classification performance about 3% on several popular benchmarks. © 2018 Association for Computing machinery.

关键词： Image classification

来源：评论

学校读者我要写书评

暂无评论

Fast reconstruction and optical-sectioning three-dimensional structured illumination microscopy (vol 6, 100757, 2025)

引用

INNOVATION 2025年第2期6卷 100757页

作者： Cao, Ruijie Li, Yaning Wang, Wenyi Fu, Yunzhe Bu, Xiaoyu Saimi, Dilizhatai Sun, Jing Ge, Xichuan Jiang, Shan Pei, Yuru Gao, Baoxiang Chen, Zhixing Li, Meiqi Xi, Peng Department of Biomedical Engineering College of Future Technology Peking University Beijing 100871 China National Biomedical Imaging Center College of Future Technology Peking University Beijing 100871 China China Academy of Space Technology Beijing Institute of Space Mechanics and Electricity Beijing 100094 China Airy Technologies Co. Ltd. Beijing 100081 China Key Laboratory of Analytical Science and Technology of Hebei Province College of Chemistry and Materials Science Hebei University Baoding 071002 China College of Future Technology Institute of Molecular Medicine National Biomedical Imaging Center Beijing Key Laboratory of Cardiometabolic Molecular Medicine Peking University Beijing 100871 China Institute of Biomedical Engineering Beijing Institute of Collaborative Innovation Beijing China Key Laboratory of Machine Perception (MOE) Department of Machine Intelligence Peking University Beijing 100871 China Peking-Tsinghua Center for Life Science Academy for Advanced Interdisciplinary Studies Peking University Beijing 100871 China School of Life Sciences Peking University Beijing 100871 China

Three-dimensional structured illumination microscopy (3DSIM) is a popular method for observing subcellular/cellular structures or animal/plant tissues with gentle phototoxicity and 3D super-resolution. However, its time-consuming reconstruction process poses challenges for high-throughput imaging and real-time observation. Moreover, traditional 3DSIM typically requires more than six z layers for successful reconstruction and is susceptible to defocused backgrounds. This poses a great gap between single-layer 2DSIM and 6-layer 3DSIM, and limits the observation of thicker samples. To address these limitations, we developed FO-3DSIM, a novel method that integrates spatial-domain reconstruction with optical-sectioning SIM. FO-3DSIM enhances reconstruction speed by up to 855.7 times with superior performance with limited z layers and under high defocused backgrounds. It retains the high-fidelity, low-photon reconstruction capabilities of our previously proposed Open-3DSIM. Utilizing fast reconstruction and optical sectioning, we achieved large field-of-view (FOV) 3D super-resolution imaging of mouse kidney actin, covering a region of 0.453 mm × 0.453 mm × 2.75 μm within 23 min of acquisition and 13 min of reconstruction. Near real-time performance was demonstrated in live actin imaging with FO-3DSIM. Our approach reduces photodamage through limited z layer reconstruction, allowing the observation of ER tubes with just three layers. We anticipate that FO-3DSIM will pave the way for near real-time, large FOV 6D imaging, encompassing xyz super-resolution, multi-color, long-term, and polarization imaging with less photodamage, removed defocused backgrounds, and reduced reconstruction time.

关键词： three-dimensional structure illumination optical sectioning reconstruction speed large field of view imaging real-time observation

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：