检索结果-内蒙古大学图书馆

image processing, Computer Vision and Machine Learning (ICICML), International Conference on

作者： Tianyu Yang Weijie Xu Yu Hua College of Computer Science and Cyber Security (Oxford Brookes College) Chengdu University of Technology Chengdu China

With the frequent occurrence of geological disasters, landslide identification has become an important research problem. In recent years, with the application and research of deep learning in the fields of computer vision and remote sensing image analysis, deep learning has become a popular method for landslide identification research because of its excellent feature extraction and pattern recognition capabilities. In this paper, we add new bands reflecting vegetation and water of landslide factors to the original bands of remote sensing image samples by linear combination of bands, and conduct two sets of comparison experiments on Unet and Swin-Unet to verify that adding band features can help improve landslide identification accuracy. The experimental results show that the experimental group with additional band features has a numerical improvement in landslide identification accuracy, with F1-score as the evaluation index, the Unet experimental group has a 2.29% improvement and the Swin-Unet experimental group has a 1.78% improvement. The results of this paper have implications for the subsequent application of bands combination of landslide factor features in remote sensing landslide identification driven by deep learning methods.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Detecting Out-Of-Distribution Earth Observation images with Diffusion Models

Detecting Out-Of-Distribution Earth Observation Images with ...

引用

IEEE Computer Society Conference on Computer Vision and pattern recognition Workshops (CVPRW)

作者： Georges Le Bellier Nicolas Audebert Cnam CEDRIC EA4629 Paris France Univ. Gustave Eiffel ENSG IGN LASTIG Saint-Mandé France

ISBN: (数字)9798350365474

ISBN: (纸本)9798350365481

Earth Observation imagery can capture rare and unusual events, such as disasters and major landscape changes, whose visual appearance contrasts with the usual observations. Deep models trained on common remote sensing data will output drastically different features for these out-of-distribution samples, compared to those closer to their training dataset. Detecting them could therefore help anticipate changes in the observations, either geographical or environmental. In this work, we show that the reconstruction error of diffusion models can effectively serve as unsupervised out-of-distribution detectors for remote sensing images., using as a plausibility score. Moreover, we introduce ODEED, a novel reconstruction-based scorer using the probability-flow ODE of diffusion models. We validate it experimentally on SpaceNet 8 with various scenarios, such as classical OOD detection with geographical shift and near-OOD setups: pre/post-flood and non-flooded/flooded image recognition. We show that our ODEED scorer significantly outperforms other diffusion-based and discriminative baselines on the more challenging near-OOD scenarios of flood image detection, where OOD images are close to the distribution tail. We aim to pave the way towards better use of generative models for anomaly detection in remote sensing.

关键词： Training Visualization Tail Diffusion models Data models Satellite images pattern recognition

来源：评论

学校读者我要写书评

暂无评论

Robust deep alignment network with remote sensing knowledge graph for zero-shot and generalized zero-shot remote sensing image scene classification

引用

ISPRS JOURNAL OF PHOTOGRAMMETRY AND remote sensing 2021年 179卷 145-158页

作者： Li, Yansheng Kong, Deyu Zhang, Yongjun Tan, Yihua Chen, Ling Wuhan Univ Sch Remote Sensing & Informat Engn Wuhan 430079 Peoples R China Huazhong Univ Sci & Technol Sch Artificial Intelligence & Automat Wuhan 430074 Peoples R China Shenzhen Huazhong Univ Sci & Technol Res Inst Shenzhen 518000 Peoples R China Zhejiang Univ Coll Comp Sci & Technol Hangzhou 310027 Peoples R China

Although deep learning has revolutionized remote sensing (RS) image scene classification, current deep learningbased approaches highly depend on the massive supervision of predetermined scene categories and have disappointingly poor performance on new categories that go beyond predetermined scene categories. In reality, the classification task often has to be extended along with the emergence of new applications that inevitably involve new categories of RS image scenes, so how to make the deep learning model own the inference ability to recognize the RS image scenes from unseen categories, which do not overlap the predetermined scene categories in the training stage, becomes incredibly important. By fully exploiting the RS domain characteristics, this paper constructs a new remote sensing knowledge graph (RSKG) from scratch to support the inference recognition of unseen RS image scenes. To improve the semantic representation ability of RS-oriented scene categories, this paper proposes to generate a Semantic Representation of scene categories by representation learning of RSKG (SR-RSKG). To pursue robust cross-modal matching between visual features and semantic representations, this paper proposes a novel deep alignment network (DAN) with a series of well-designed optimization constraints, which can simultaneously address zero-shot and generalized zero-shot RS image scene classification. Extensive experiments on one merged RS image scene dataset, which is the integration of multiple publicly open datasets, show that the recommended SR-RSKG obviously outperforms the traditional knowledge types (e.g., natural language processing models and manually annotated attribute vectors), and our proposed DAN shows better performance compared with the state-of-the-art methods under both the zero-shot and generalized zero-shot RS image scene classification settings. The constructed RSKG will be made publicly available along with this paper (https://***/kdy2021/SR-RSKG).

关键词： Robust deep alignment network (DAN) remote sensing knowledge graph (RSKG) remote sensing image scene classification Zero-shot learning (ZSL) Generalized zero-shot learning (GZSL)

来源：评论

学校读者我要写书评

暂无评论

Diffusion posterior sampling for remote sensing image fusion

Diffusion posterior sampling for remote sensing image fusion

引用

7th International Conference on Vision, image and Signal processing (ICVISP 2023)

作者： C. Zhang Z. Wang J. Han College of science National University of Defense Technology Changsha People's Republic of China

remote sensing image fusion, i.e., fusing remote sensing images from different sensors or different time into a comprehensive image, can integrate image information for all kinds of image tasks, such as object detection, pattern recognition, and so on. In this paper, we focus on the image fusion of optical image and synthetic aperture radar (SAR) image, where the traditional methods, such as sparse representation and image decomposition, usually fail to improve the image quality and enhance object features' level. Inspired by the powerful generation model named diffusion denoising probabilistic model (DDPM), we propose a novel method based on diffusion posterior sampling for image fusion of optical image and SAR image. Starting from the variational model of image fusion and the total variation constraint, we approximate the posterior sampling process of image fusion by a closed-form analytic solution. After that, DDPM can be used to generate the fused image with high image quality and enhanced object features. The feasibility and the superiority of the proposed method is validated in numerical experiments.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Dual Stream Fusion Network for Multi-spectral High Resolution remote sensing image Segmentation 4th

Dual Stream Fusion Network for Multi-spectral High Resolutio...

引用

4th Chinese Conference on pattern recognition and Computer Vision (PRCV)

作者： Cao, Yong Shi, Yiwen Liu, Yiwei Huo, Chunlei Xiang, Shiming Pan, Chunhong Univ Chinese Acad Sci Sch Artificial Intelligence Beijing 100049 Peoples R China Chinese Acad Sci Inst Automat Natl Lab Pattern Recognit Beijing 100190 Peoples R China Beijing City Univ Intelligent Elect Mfg Res Ctr Beijing 100191 Peoples R China Beijing Univ Civil Engn & Architecture Beijing 102627 Peoples R China

ISBN: (纸本)9783030880071;9783030880064

Semantic segmentation is in-demand in High Resolution remote sensing (HRRS) image processing. Unlike natural images, HRRS images usually provide channels such as Near Infrared (NIR) in addition to RGB channels. However, in order to make use of the pre-trained model, the current semantic segmentation methods in remote sensing field usually only use the RGB channel and discard the information of other channels. In this paper, to make full use of the HRRS image information, a dual-stream fusion network is proposed to fuse the information of different channel combinations through a Feature Pyramid Network (FPN), then a Stage Pyramid Pooling (SPP) module is used to integrate the features of different scales and produce the final segmentation results. Experiments on the RSCUP competition dataset show that the proposed approach can effectively improve the segmentation performance.

关键词： Semantic segmentation remote sensing Stream fusion

来源：评论

学校读者我要写书评

暂无评论

Explainable scale distillation for hyperspectral image classification

引用

pattern recognition 2022年 122卷

作者： Shi, Cheng Fang, Li Lv, Zhiyong Zhao, Minghua Xian Univ Technol Sch Comp Sci & Engn Xian 710048 Shannxi Peoples R China Chinese Acad Sci Quanzhou Inst Equipment Mfg Haixi Inst Quanzhou 362216 Fujian Peoples R China

The land-covers within an observed remote sensing scene are usually of different scales;therefore, the ensemble of multi-scale information is a commonly used strategy to achieve more accurate scene inter-pretation;however, this process suffers from being time-consuming. In terms of this issue, this paper proposes a scale distillation network to explore the possibility that single-scale classification network can achieve the same (or even better) classification performance compared with multi-scale one. The pro-posed scale distillation network consists of a cumbersome multi-scale teacher network and a lightweight single-scale student network. The former is trained for multi-scale information learning, and the latter improves the classification accuracy by accepting the knowledge from the multi-scale teacher network and its true label. The experimental results show the advantages of scale distillation on hyperspectral image classification. The single-scale student network can even achieve higher evaluation accuracy than the multi-scale teacher network. In addition, a faithful explainable scale network is designed to visually explain the trained scale distillation network. The traditional deep neural network is a black-box and lacks interpretability. The explanation of the trained network can explore more hidden information from the predictions. We visually explain the prediction results of scale distillation network, and the results show that the explainable scale network can more precisely analyze the relationship between the learned scale features and the land-cover categories. Moreover, the possible application of the explainable scale network on classification is further discussed in this study. (c) 2021 Elsevier Ltd. All rights reserved.

关键词： Hyperspectral image classification Knowledge distillation Scale distillation Explainable scale network

来源：评论

学校读者我要写书评

暂无评论

FocalSR: Revisiting image super-resolution transformers with fourier-transform cross attention layers for remote sensing image enhancement

引用

Geomatica 2025年第1期77卷

作者： Ou, Botong Shao, Gang Yang, Baijian Fei, Songlin Computer and Information Technology Purdue University West LafayetteIN47907 United States Libraries and School of Information Studies Purdue University West LafayetteIN47907 United States Forestry and Natural Resources Purdue University West LafayetteIN47907 United States

Transformer architecture has attained noteworthy performance achievements in recent image super-resolution research. However, current transformer-based methods still expose limitations in fully harnessing domain-specific information within images, particularly when applied to broader-scale remote sensing images that contain diverse landscape objects on one scene. remote sensing images have relatively lower resolution compared to the common super-resolution training dataset and each landscape object covers a small area on the image. These natures of remote sensing images significantly reduced the attention pixels for image restoration in existing transformer-based methods. To address this challenge and enhance domain-specific multi-object image reconstruction, we introduce FocalSR, a Transformer model featuring FOurier-transform Cross Attention Layers for Super-Resolution. Drawing inspiration from state-of-the-art Transformer models like Hybrid Attention Transformer (HAT), FocalSR incorporates channel-focused and window-centric self-attention mechanisms. By integrating Fast Fourier Convolution into the cross-attention layer, FocalSR extends its capacity to capture image-wide information and intricate details in low-resolution images. Through unified task pretraining during model development, we validate the efficacy of these enhancements through extensive testing, resulting in substantial performance improvements. Notably, our experiments showcase FocalSR's superior performance in remote sensing datasets, demonstrating a notable 1 dB enhancement in the PSNR metric compared to other state-of-the-art methods. Additionally, significant improvements are observed in challenging scenarios such as pattern restoration and vegetation detail preservation, underscoring the transformative potential of FocalSR in advancing image processing and domain-specific vision tasks. © 2024 The Authors

关键词： Restoration

来源：评论

学校读者我要写书评

暂无评论

Leveraging Segment-Anything model for automated zero-shot road width extraction from aerial imagery

Leveraging Segment-Anything model for automated zero-shot ro...

引用

2023 International Conference on Digital image Computing: Techniques and Applications, DICTA 2023

作者： Xu, Nan Nice, Kerry Seneviratne, Sachith Stevenson, Mark Faculty of Engineering and It Department of Infrastructure Engineering Australia Melbourne School of Design Transport Health and Urban Systems Research Lab Australia The University of Melbourne Optimization and Pattern Recognition Group Faculty of Engineering and It ParkvilleVIC Australia

ISBN: (纸本)9798350382204

Segment-Anything model (SAM) is a foundation segmentation model published in April 2023. Trained on an unprecedented 11 million annotated images, the model can generate segmented masks bearing clear-cut contours by integrating user-provided prompts. It is zero-shot transferable, requiring no task-specific training. Nevertheless, its applicability for geographic vision tasks has not been fully evaluated. There is no automated prompt-feeding method incorporating with SAM that can work efficiently for purposeful batch processing as well. To fill these gaps, we developed a process that can be executed automatically from visual-prompts extraction to road width measurement, utilizing OpenStreetMap (OSM) and SAM. By examining the quality of segmentation in various image contexts, we evaluated the capacity and limitations of SAM working on aerial imagery. Through comparing measured widths to VicRoads records, we validated the specially designed width-measuring algorithm for high precision and accuracy. After this process, prompt-indicated zero-shot approach in solving basic geographic vision tasks is to be shaped synchronously on both theory and application ends. © 2023 IEEE.

关键词： remote sensing

来源：评论

学校读者我要写书评

暂无评论

Automatic wildfire monitoring system based on deep learning

引用

EUROPEAN JOURNAL OF remote sensing 2022年第1期55卷 551-567页

作者： Peng, Yingshu Wang, Yi Chinese Acad Sci Lushan Bot Garden Nanchang 332900 Jiangxi Peoples R China Nanjing Forestry Univ Coll Forestry Nanjing Peoples R China Jiangsu Wiscom Technol Co Ltd Smart City Res Inst Nanjing Peoples R China

Fire detection based on computer vision technology can avoid many flaws in conventional methods. However, existing methods fail to achieve a good trade-off in accuracy, model size, speed, and cost. This paper presents a high-performance forest fire recognition algorithm to solve the current problems in forest fire monitoring. Firstly, visual saliency areas in motion images are extracted to improve detection efficiency. Secondly, transfer learning techniques are employed to improve the generalization performance of the constructed deep learning classification model. Finally, fire detection is realized based on C++ deployment algorithms Compared with the existing forest fire detection methods, the proposed method has higher classification accuracy and speed, with a more comprehensive application range and lower cost. The performance of our method can meet the accuracy and speed requirements of real-time fire detection, and it can be deployed and practiced on multiple platforms.

关键词： Forest fire flame detection deep learning image processing model deployment

来源：评论

学校读者我要写书评

暂无评论

Overview of Research Progress of Digital image processing Technology

Overview of Research Progress of Digital Image Processing Te...

引用

2022 International Conference on Computing Innovation and Applied Physics, CONF-CIAP 2022

作者： Huang, Yitao Department of Electronic Science and Technology Tongji University Shanghai201804 China

Digital image processing technology has gone through rapid development and is extensively applied in daily life and production, with the rapid development of modern information technology. It plays an inestimable role in remote sensing, medicine, recognition and other fields. This paper briefly introduces the basic concept of digital image processing, summarizes and analyses the commonly used digital image processing technology and the latest scientific research achievements from four aspects, and puts forward the future development direction of digital image processing. In the future, it will pay more attention to artificial intelligence algorithms and achieve better processing results by optimizing the logical structure. By using the simplified image algorithm, the application scope of digital image processing will gradually expand, and will develop in the direction of miniaturization, intelligence, and convenience. © Published under licence by IOP Publishing Ltd.

关键词： Artificial intelligence

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：