检索结果-内蒙古大学图书馆

Adaptive Multitype Contrastive Views Generation for remote sensing image Semantic segmentation

IEEE TRANSACTIONS ON GEOSCIENCE AND remote sensing 2025年 63卷

作者： Shi, Cheng Han, Peiwen Zhao, Minghua Fang, Li Miao, Qiguang Pun, Chi-Man Xian Univ Technol Sch Comp Sci & Engn Xian 710048 Shaanxi Peoples R China Chinese Acad Sci Quanzhou Inst Equipment Mfg Haixi Inst Quanzhou 362216 Fujian Peoples R China Xidian Univ Sch Comp Sci & Technol Xian 710071 Shaanxi Peoples R China

Self-supervised contrastive learning is a powerful pretraining framework for learning the invariant features from the different views of remote sensing images, therefore, the performance of contrastive learning heavily depends on the generation of views. Current view generation is primarily accomplished through different transformations, and the types and parameters of the transformations are required hand-crafted. Hence, the diversity and discriminability of generated views cannot be guaranteed. To address this, we propose a multitype views optimization method to optimize these transformations. We formulate contrastive learning as a min-max optimization problem, and transformation parameters are optimized by maximizing the contrastive loss. The optimized transformations encourage the negative sample pairs to be close and the positive sample pairs to be far apart. Different from the current adversarial view generation methods, our method can optimize both photometric transformations and geometric transformations. For remote sensing images, the geometric transformation is more critical for view generation, while the existing view optimization methods fail to achieve this. We consider the hue, saturation, brightness, contrast, and geometric rotation transformations in contrastive learning, and evaluate the optimized views on the downstream remote sensing images semantic segmentation task. Extensive experiments are carried out on the three remote sensing image segmentation datasets, including the ISPRS Potsdam dataset, the ISPRS Vaihingen dataset, and the LoveDA dataset. Results show that the learned views obtain high advantages compared to the hand-crafted views and other optimized views. The code associated with this article has been released and can be accessed at https://***/AAAA-CS/AMView.

关键词： remote sensing Contrastive learning Semantics Semantic segmentation Feature extraction Data augmentation Brightness Adaptation models Sensors Perturbation methods remote sensing image segmentation view generation

来源：评论

学校读者我要写书评

暂无评论

Advancing perturbation space expansion based on information fusion for semi-supervised remote sensing image semantic segmentation

引用

INFORMATION FUSION 2025年 117卷

作者： Zhou, Liang Duan, Keyi Dai, Jinkun Ye, Yuanxin Southwest Jiaotong Univ Fac Geosci & Engn Chengdu Sichuan Peoples R China State Prov Joint Engn Lab Spatial Informat Technol Chengdu Sichuan Peoples R China

Existing deep models have greatly enhanced the performance of semantic segmentation in remote sensing (RS) images, but they are often limited by the scarcity of labeled samples. Semi-supervised learning (SSL) can leverage a vast amount of unlabeled data for model training, effectively overcoming the reliance on labeled data. Nonetheless, due to the complexity and diversity of ground objects in RS images, existing perturbation methods designed for natural images exhibit significant limitations when extended to remote sensing imagery, failing to construct a broader perturbation space. Incorporating the unique characteristics of RS images, this work presents a novel consistency regularization semi-supervised framework, named FusionMatch, which innovatively integrates multi-modal near-infrared information (termed as NIRPerb) and applies differentiated pan-sharpening fusion techniques (termed as PSPerb) during the perturbation process, thereby significantly expanding the perturbation space. FusionMatch not only enriches the dataset with additional spectral dimensions but also improves the model's capacity to discern subtle variations and patterns. Moreover, FusionMatch can be seamlessly integrated with other consistency regularization methods to enhance the capabilities of segmentation models. Two binary classification datasets and two multi-class datasets are used to assess the segmentation performance of FusionMatch. The experimental results confirm the robustness of FusionMatch compared to other leading semi-supervised approaches.

关键词： remote sensing image segmentation Semi-supervised semantic segmentation Information fusion Perturbation space expansion

来源：评论

学校读者我要写书评

暂无评论

remote sensing image segmentation using active contours based on intercorrelation of nonsubsampled contourlet coefficients

引用

JOURNAL OF ELECTRONIC IMAGING 2016年第6期25卷

作者： Fang, Lingling Wang, Xianghai Sun, Yang Xu, Kainan Liaoning Normal Univ Dept Comp & Informat Technol Dalian 116029 Liaoning Provin Peoples R China

Considering that remote sensing images contain rich scale-dependent information and geographical detailed information, segmentation process must be carried out under the multiscale case. The vector-valued C-V active contour model is an effective image segmentation method, but the model cannot segment the non homogeneous remote sensing images well. The image processing methods based on nonsubsampled contourlet transform (NSCT) can fully use the detailed information of remote sensing images. The interscale distribution characteristics of NSCT coefficients at finer scale is first analyzed and then a statistical model of signal singularities combining the coefficient correlation between intrascale and interscale is proposed. Based on the above, the vector-valued C-V active contour model is then applied to the statistical characteristics for segmenting images. Consequently, the proposed method can preserve detailed information of images and other desirable properties of active contour model. Numerical examples indicate that the proposed method is very competitive with several state-of-the-art techniques. (C) 2016 SPIE and IS&T

关键词： remote sensing image segmentation nonsubsampled contourlet transform coefficient distribution vector-valued C-V model

来源：评论

学校读者我要写书评

暂无评论

remote sensing image segmentation based on region-split and graph cut

Remote sensing image segmentation based on region-split and ...

引用

International Conference on Advanced Materials and Computer Science

作者： Jiang Hua Wen Jing Guilin Uniers Elect Technol Comp Sch Guilin Peoples R China

ISBN: (纸本)9783037850978

According to the features of HVS, the algorithm uses region-split method to segment the remote sensing image into a large number of small regions. By integrating gray feature and spatial location of each region, NC is used to segment the image among regions from global view, by which the final segmented image can be generated. Experimental results show that comparing with the traditional NC, operating speed is significantly improved as getting close segmentation quality, and this is a kind of effective method of image segmentation.

关键词： remote sensing image segmentation region- split graph cut Normalized Cut

来源：评论

学校读者我要写书评

暂无评论

remote sensing image segmentation based on human visual system region-split and graph cut

Remote sensing image segmentation based on human visual syst...

引用

International Conference on Recent Trends in Materials and Mechanical Engineering (ICRTMME 2011)

作者： Jiang Hua Wen Jing Guilin Uniers Elect Technol Comp & Control Coll Guilin Peoples R China

ISBN: (纸本)9783037850992

Aiming at the problem of poor real-time ability of Normalized Cut (NC), this paper suggests a remote sensing image segmentation algorithm based on region-split and graph cut within human visual system (HVS). According to the features of HVS, the algorithm uses region-split method to segment the remote sensing image into a large number of small regions. By integrating gray feature and spatial location of each region, NC is used to segment the image among regions from global view, by which the final segmented image can be generated. Experimental results show that comparing with the traditional NC, operating speed is significantly improved as getting close segmentation quality, and this is a kind of effective method of image segmentation.

关键词： remote sensing image segmentation region- split graph cut Normalized Cut

来源：评论

学校读者我要写书评

暂无评论

remote sensing image segmentation based on a modified pulse coupled neural network

Remote sensing image segmentation based on a modified pulse ...

引用

Conference on Optical sensing and Imaging Technologies and Applications / International Symposium on Optoelectronic Technology and Application (OTA) / Annual Conference of the Chinese-Society-for-Optical-Engineering (CSOE)

作者： Li, Sun Hua, Nian Beijing Inst Space Mech & Elect Beijing 100094 Peoples R China

ISBN: (纸本)9781510623354

In order to adaptively adjust the model parameters and the global threshold for image segmentation, an improved pulse coupled neural network (PCNN) model based on human visual system (HVS) is proposed in this paper. Due to the property of HVS that human visual sensitivity to an image varies with different regions of the image where different regions correspond to different information rate area of the image, we analyze the characteristics of the improved model and its parameter optimization principle, and propose an improved segmentation algorithm. According to the gray scale of pixels, the algorithm adaptively realizes the division of the image information area. It not only preserves the excellent characteristics of PCNN for image segmentation, but also effectively preserves the gradation of image itself. The experiment results show that the algorithm proposed is efficient and has good segmentation effect, and has a wide application prospect in remote sensing image processing.

关键词： Pulse coupled neural network self-adaption parameter optimization remote sensing image segmentation remote sensing image processing

来源：评论

学校读者我要写书评

暂无评论

An improved optimum-path forest clustering algorithm for remote sensing image segmentation

引用

COMPUTERS & GEOSCIENCES 2018年 112卷 38-46页

作者： Chen, Siya Sun, Tieli Yang, Fengqin Sun, Hongguang Guan, Yu Northeast Normal Univ Sch Informat Sci & Technol Changchun 130024 Jilin Peoples R China Northeast Normal Univ Sch Geog Sci Changchun 130024 Jilin Peoples R China Northeast Normal Univ Key Lab Intelligent Informat Proc Jilin Univ Changchun 130024 Jilin Peoples R China

remote sensing image segmentation is a key technology for processing remote sensing images. The image segmentation results can be used for feature extraction, target identification and object description. Thus, image segmentation directly affects the subsequent processing results. This paper proposes a novel Optimum-Path Forest (OPF) clustering algorithm that can be used for remote sensing segmentation. The method utilizes the principle that the cluster centres are characterized based on their densities and the distances between the centres and samples with higher densities. A new OPF clustering algorithm probability density function is defined based on this principle and applied to remote sensing image segmentation. Experiments are conducted using five remote sensing land cover images. The experimental results illustrate that the proposed method can outperform the original OPF approach.

关键词： Optimum-path forest Clustering remote sensing image segmentation

来源：评论

学校读者我要写书评

暂无评论

Improved fast mean shift algorithm for remote sensing image segmentation

引用

IET image PROCESSING 2015年第5期9卷 389-394页

作者： Zhou, Jia-Xiang Li, Zhi-Wei Fan, Chong Cent South Univ Sch Geosci & Info Phys Changsha 410083 Hunan Peoples R China

image segmentation plays a crucial role in object-based remote sensing information extraction. This study improves the existing mean shift (MS) algorithm for segmenting high resolution remote sensing imagery by adopting two strategies. First, a pixel-based, fixed bandwidth and weighted MS algorithm is applied to cluster the image. In this process, the space bandwidth is selected according to the resolution of remote sensing images, and the range bandwidths of each band are calculated based on grey feature and the plug-in rule. Gaussian kernels are used for clustering. Second, a region-based MS algorithm is applied to globally merge modes which are obtained in the first step. The spatial and range bandwidths are adaptively adjusted based on the clustering result of the first step. Experimental results with two Quickbird images show that the improved algorithm is superior to the typical MS algorithm, producing high precision and requiring less operation time.

关键词： Quickbird images fixed bandwidth MS algorithm region-based MS algorithm image recognition Geophysics computing Computer vision and image processing techniques Data and information geophysical image processing remote sensing fast mean shift algorithm image segmentation pattern clustering space bandwidth Gaussian kernels remote sensing image segmentation plug-in rule grey feature Geophysical techniques and equipment pixel-based MS algorithm image clustering Instrumentation and techniques for geophysical, hydrospheric and lower atmosphere research acquisition, processing, storage and dissemination in geophysics weighted MS algorithm

来源：评论

学校读者我要写书评

暂无评论

remote sensing image segmentation Based on a Novel Gaussian Mixture Model and SURF Algorithm

引用

INTERNATIONAL JOURNAL OF SWARM INTELLIGENCE RESEARCH 2023年第2期14卷

作者： Yin, Shoulin Wang, Liguo Wang, Qunming Yang, Jinghui Jiang, Man Harbin Engn Univ Coll Informat & Commun Engn Harbin Peoples R China Dalian Minzu Univ Coll Informat & Commun Engn Dalian Peoples R China Tongji Univ Coll Surveying & Geoinformat Shanghai Peoples R China China Univ Geosci Sch Informat Engn Beijing Peoples R China Vocat Tech Coll Modern Serv Beijing Peoples R China

This paper proposes a novel remote sensing image segmentation method based on Gaussian mixture model and SURF algorithm. Firstly, Gaussian mixture model is used for remote sensing image segmentation. Then the surf matching algorithm is adopted for eliminating misidentified areas. The determinant of Hession matrix (DoH) is used to detect key points in the image. The non-maximum suppression method and interpolation operation are utilized to search and locate the extreme points. The maximum likelihood method is used to estimate model parameters. Some remote sensing images in THE DOTA data set are selected for experimental verification, and the results show that the new algorithm has obvious improvement in segmentation effect and efficiency. In the background complex image segmentation, the improved algorithm has more obvious advantages compared than state-of-the-art segmentation methods.

关键词： Gaussian Mixture Model Maximum Likelihood remote sensing image segmentation SURF Algorithm

来源：评论

学校读者我要写书评

暂无评论

remote sensing image segmentation based on region-split and graph cut

Remote sensing image segmentation based on region-split and ...

引用

2011 International Conference on Advanced Materials and Computer Science(ICAMCS 2011)

作者： JIANG Hua WEN Jing Computer school Guilin uniersity of electronic technology

According to the features of HVS,the algorithm uses region-split method to segment the remote sensing image into a large number of small regionsBy integrating gray feature and spatial location of each region,NC is used to segment the image among regions from global view,by which the final segmented image can be generatedExperimental results show that comparing with the traditional NC,operating speed is significantly improved as getting close segmentation quality,and this is a kind of effective method of image segmentation

关键词： remote sensing image segmentation region-split graph cut Normalized Cut

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：