检索结果-内蒙古大学图书馆

Spatial Variation Sequences for remote sensing Applications with Small Sample Sizes 1

11th Pacific-Rim Symposium on image and Video Technology (PSIVT)

作者： Jeune, Hayden Pechan, Niklas Reitsma, Sharn-Konet Kempa-Liehr, Andreas W. Univ Auckland Dept Engn Sci & Biomed Engn Auckland 1010 New Zealand

ISBN: (数字)9789819703760

ISBN: (纸本)9789819703753;9789819703760

Machine learning applications in remote sensing often require a labour-intensive feature engineering step, if only a small number of samples is available and transfer learning is not applicable. Here, we are introducing the concept of Spatial Variation Sequences, which allows to apply methodologies from automated time-series feature engineering to remote sensing applications of static images. The presented example application detects swimming pools from four-channel satellite images with an F-1-score of 0.95, by generating spatial variation sequences from a modified swimming pool index. The automated feature engineering approach reduced the dimensionality of the classification problem by 99.7%. A more traditional approach using transfer learning on pretrained Convolutional Neural Networks (CNN) was evaluated in parallel for comparison. The CNN approach boasted a higher performance of F-1-score of 0.98 but required the use of pre-trained weights. The comparable performance of the FE and CNN approach demonstrates that time-series feature extraction is a valuable alternative to traditional remote sensing methods in the presence of data scarcity or the need of significant dimensionality reduction.

关键词： pattern recognition Swimming pool detection Feature engineering Explainable machine learning

来源：评论

学校读者我要写书评

暂无评论

A lightweight license plate detection algorithm based on deep learning

引用

IET image processing 2024年第2期18卷 403-411页

作者： Zhu, Shuo Wang, Yu Wang, Zongyang Wuxi Univ Jiangsu Prov Engn Res Ctr Integrated Circuit Relia Wuxi 214105 Peoples R China Nanjing Univ Informat Sci & Technol Sch Elect & Informat Engn Nanjing Peoples R China Jiangsu Jicui Depth Sensing Technol Res Inst Co LT Wuxi Peoples R China

License plate detection is an important task in Intelligent Transportation Systems (ITS) and has a wide range of applications in vehicle management, traffic control, and public safety. In order to improve the accuracy and speed of mobile recognition, an improved lightweight YOLOv5s model is proposed for license plate detection. First, an improved Stemblock network is used to replace the original Focus layer in the network, which ensures strong feature expression capability and reduces a large number of parameters to lower the computational complexity;then, an improved lightweight network, ShuffleNetv2, is used to replace the backbone network of the YOLOv5s, which makes the model lighter and ensures the detection accuracy at the same time. Then, a feature enhancement module is designed to reduce the information loss caused by the rearrangement of the backbone network channels, which facilitates the information interaction in the feature fusion process;finally, the low-, medium- and high-level features in the Shufflenetv2 network structure are fused to form the final high-level output features. Experimental results on the CCPD dataset show that compared to other methods this paper obtains better performance and faster speed in the license plate detection task, in which the average precision mean value reaches 96.6%, and can achieve a detection speed of 43.86 frame/s, and the parameter volume is reduced to 5.07 M. 1. Improved Stemblock structure to reduce the number of parameters.2. Improve Shufflunetv2 as the backbone network and reduce the number of parameters.3. The attention mechanism is added to the network to improve the detection ***

关键词： image processing neural nets pattern recognition

来源：评论

学校读者我要写书评

暂无评论

Multi-Scale Mixed Pixel Network for remote sensing image Super-Resolution 7

Multi-Scale Mixed Pixel Network for Remote Sensing Image Sup...

引用

7th International Conference on pattern recognition and Artificial Intelligence, PRAI 2024

作者： Yang, He Ji, Xun Wang, Xu Chen, Shijie Dalian Maritime University School of Marine Electrical Engineering Liaoning Dalian China

ISBN: (纸本)9798350350890

Deep learning-based image super-resolution (SR) technology has gained extensive attention in the remote sensing community, which aims to reconstruct the abundant details of target images. However, the practical application of existing deep learning-based SR methods is often limited by the trade-off between network performance and computational load. This paper presents a novel multi-scale mixed pixel network (MSMP-Net) to explore the precise and cost-effective SR reconstruction of remote sensing images. Specifically, a dual-path feature capture mechanism (DPFCM) architecture is devised that incorporates diverse dilated convolution operations to comprehensively capture hierarchical features across varying receptive fields. In addition, a triple-path high-resolution rebuilder (TPHRR) structure is designed, which uses a set of dense branches, residual branches and pyramid branches to effectively promote the transfer and reuse of features. Extensive experiments reveal that our proposed MSMP-Net outperforms other state-of-the-art SR methods in reconstruction effects. © 2024 IEEE.

关键词： Cost effectiveness

来源：评论

学校读者我要写书评

暂无评论

Multi-agent hyperspectral and lidar features fusion for urban vegetation mapping

引用

EARTH SCIENCE INFORMATICS 2023年第1期16卷 165-173页

作者： Khoramak, Sahar Mahmoudi, Fatemeh Tabib Shahid Rajaee Teacher Training Univ Fac Civil Engn Dept Geomatics Tehran Iran Shahid Rajaee Teacher Training Univ Fac Civil Engn Dept Geomatics Tehran Iran

Urban vegetation recognition based on remote sensing data is highly affected by the complexities of urban areas due to the existence of various kinds of objects and their relations. Spectral similarities between tree canopies and types of grass lands, spatial adjacency between houses and tall trees, shadow and occluded areas make some difficulties for recognition of the plant species. In this research, the capabilities of multi-agent systems are utilized for feature fusion of hyperspectral imagery and lidar data for improving the vegetation recognition results in urban areas. The proposed algorithm has two main steps composed of generating a knowledge base containing spectral and height features extracted from input hyperspectral and Lidar data, respectively, and performing the hierarchical classification process to generate vegetation classification map based on parallel processing by object recognition agents. Evaluation of the capabilities of the proposed methodology is performed on hyperspectral and lidar DSM over Houston University and its surrounding areas. According to the obtained results, fusion of hyperspectral and Lidar DSM with the capabilities of multi-agent processing can improve the overall accuracy of vegetation recognition results for about 15.53% and 6.58% comparing with performing multi-agent and maximum likelihood classifier only on hyperspectral image, respectively.

关键词： Vegetation recognition Multi-agent System Feature Fusion Hyperspectral image Lidar DSM

来源：评论

学校读者我要写书评

暂无评论

An Automated Analysis Tool for the Classification of Sea Surface Temperature imagery

引用

pattern recognition AND image ANALYSIS 2022年第3期32卷 631-635页

作者： Reggiannini, M. Papini, O. Pieri, G. Natl Res Council Italy Inst Informat Sci & Technol I-56124 Pisa Italy

Sea observation through remote sensing technologies plays an essential role in understanding the health status of the marine coastal environment, its fauna species and their future behavior. Accurate knowledge of the marine habitat and the factors affecting faunal variations allows us to perform predictions and adopt proper decisions. This paper concerns the proposal of a classification system devoted to recognizing marine mesoscale events. These phenomena are studied and monitored by analyzing sea surface temperature imagery. Currently, the standard way to perform such analysis relies on experts manually visualizing, analyzing, and tagging large imagery datasets. Nowadays, the availability of remote sensing data has increased so much that it is desirable to replace the labor-intensive, time-consuming, and subjective manual interpretation with automated analysis tools. The results presented in this work have been obtained by applying the proposed approach to images captured over the southwestern region of the Iberian Peninsula.

关键词： image processing remote sensing mesoscale patterns sea surface temperature machine learning climate change

来源：评论

学校读者我要写书评

暂无评论

Supervised Spiking Neural Network for SAR image recognition: Analysis of Encoding Methods and Performance under Strong Noise Influence 12

Supervised Spiking Neural Network for SAR Image Recognition:...

引用

SPIE 12th International Symposium on Multispectral image processing and pattern recognition, MIPPR 2023

作者： Junyu, Wang Hao, Sun Tao, Tang Lin, Lei Kefeng, Ji National University of Defense Technology 109 Deya Road Kaifu District Hunan Changsha410003 China

ISBN: (数字)9781510675001

ISBN: (纸本)9781510674998

With the rapid development of artificial neural network (ANN), the field of synthetic aperture radar (SAR) target recognition has witnessed significant progress. However, due to the poor interpretability and ease of being affected by speckle noise, it brings challenges to ANN for SAR target recognition. Spiking Neural Network (SNN) has emerged as the third-generation neural network architecture and presents promising prospects for various applications. This study aims to explore the performance of SNN in SAR target recognition. In our experiments, we achieved comparable performance to conventional neural networks by utilizing directly trained SNN. This indicates the effectiveness of SNN in coping with SAR target recognition tasks. Moreover, we investigated the impact of different spiking encoders on SAR target recognition. Specifically, we compared the performance of SNN using the Poisson encoder and utilizing the first layer of the SNN as an encoder. This comparison provides valuable insights into the optimal coding strategy for SNN-based SAR target recognition. Additionally, we examined the robustness of SNNs in the presence of strong speckle noise. Our findings demonstrate that SNN can maintain good performance under the influence of strong speckle noise. The outcomes of this research shed light on the potential of SNN as a powerful tool for SAR target recognition. Future studies can focus on exploring SNN's applicability to SAR Interpretation. © 2024 SPIE.

关键词： Synthetic aperture radar

来源：评论

学校读者我要写书评

暂无评论

Multi-class Token-Guided End-to-End Weakly Supervised image Semantic Segmentation Method 7th

Multi-class Token-Guided End-to-End Weakly Supervised Image ...

引用

7th Chinese Conference on pattern recognition and Computer Vision

作者： Cao, Yifan He, Lijun Ma, Ting Li, Fan Xi An Jiao Tong Univ Xian Peoples R China

ISBN: (纸本)9789819784929;9789819784936

Weakly supervised image semantic segmentation has become the most popular method in recent years because of its low cost and has been widely used in medical image segmentation, automatic driving, remote sensing image analysis and other fields. However, the current weakly supervised semantic segmentation based on transformer has some problems, such as focusing on the whole, ignoring local details and confusing different categories. To solve these problems, we come up with a token-guided single stage weakly supervised image semantic segmentation algorithm. First of all, in order to solve the problem of insufficient attention to details, we proposed an optimization clipping method, which realized the selection of uncertain regions as much as possible and the fine marking of uncertain regions. Then, the single-class token to multiple class tokens method is purposed to obtain multiple class tokens for fine guidance. In particular, we designed a multiple class tokens guide method to complete the function of classifying uncertain regions and correctly activating them. The quantitative and qualitative results of the public dataset PASCAL VOC 2012 validate the effectiveness of the method.

关键词： Semantic segmentation Weakly supervision Multi-class tokens Regional guidance

来源：评论

学校读者我要写书评

暂无评论

High-Resolution SAR Imaging Characteristics for Multiple Scattering of Rotating Targets

引用

IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND remote sensing 2024年 17卷 9974-9988页

作者： Wang, Yixin Qiu, Xiaolan Wen, Xuejiao Chinese Acad Sci Aerosp Informat Res Inst Beijing 100094 Peoples R China Chinese Acad Sci Key Lab Technol Geospatial Informat Proc & Applica Beijing 100190 Peoples R China Univ Chinese Acad Sci Sch Elect Elect & Commun Engn Beijing 100049 Peoples R China Chinese Acad Sci Aerosp Informat Res Inst Natl key Lab Microwave Imaging Beijing 100190 Peoples R China Suzhou Aerosp Informat Res Inst Suzhou Key Lab MicrowaveImaging Proc & Applicat Te Suzhou 215123 Peoples R China Univ Sci & Technol China Hefei 230052 Peoples R China Suzhou Aerosp Informat Res Inst Lab Spatial Informat Intelligent Proc Syst Suzhou 215000 Peoples R China

The rotation of targets in high-resolution synthetic aperture radar (SAR) images results in complex and nonlinear micro-Doppler modulation of SAR returns, exhibiting distinct and diverse imaging characteristics under different conditions. This poses challenges to image interpretation, target detection and recognition. This article establishes signal models for single and multiple scattering of rotating targets in high-resolution spaceborne SAR and, in conjunction with the processing procedures of range Doppler algorithm (RDA), derives the azimuth bandwidth and the conditions for azimuth frequency aliasing caused by rotation. Quantitative analysis of the errors introduced by target rotation in various signal domains is performed, taking into account the stationary phase point offset, the rotation-induced range migration correction error, and the change of azimuth phase in detail. In addition, representative signatures of these influences in SAR images are simulated and interpreted and more accurate theoretical models of imaging results are derived. The processing and analysis of both simulation and real measured data validate the correctness of the theoretical derivation in this study and provide support for the interpretation and information extraction of high-resolution SAR rotating targets.

关键词： Scattering Azimuth Radar polarimetry Imaging History image resolution Spaceborne radar High-resolution SAR imaging characteristics multiple scattering target with rotational micro-motion

来源：评论

学校读者我要写书评

暂无评论

Robust Disaster Assessment from Aerial imagery Using Text-to-image Synthetic Data

Robust Disaster Assessment from Aerial Imagery Using Text-to...

引用

IEEE/CVF Conference on Computer Vision and pattern recognition (CVPR)

作者： Kalluri, Tarun Lee, Jihyeon Sohn, Kihyuk Singla, Sahil Chandraker, Manmohan Xu, Joseph Liu, Jeremiah Univ Calif San Diego La Jolla CA 92093 USA Google Res Mountain View CA USA

ISBN: (纸本)9798350365474

We present a simple and efficient method to leverage emerging text-to-image generative models in creating large-scale synthetic supervision for the task of damage assessment from aerial images. While significant recent advances have resulted in improved techniques for damage assessment using aerial or satellite imagery, they still suffer from poor robustness to domains where manual labeled data is unavailable, directly impacting post-disaster humanitarian assistance in such under-resourced geographies. Our contribution towards improving domain robustness in this scenario is two-fold. Firstly, we leverage the text-guided mask-based image editing capabilities of generative models and build an efficient and easily scalable pipeline to generate thousands of post-disaster images from low-resource domains. Secondly, we propose a simple two-stage training approach to train robust models while using manual supervision from different source domains along with the generated synthetic target domain data. We validate the strength of our proposed framework under cross-geography domain transfer setting from xBD and SKAI images in both single-source and multi-source settings, achieving significant improvements over a source-only baseline in each case.

关键词： disaster assessment domain adaptation domain generalization generative models remote sensing synthetic data

来源：评论

学校读者我要写书评

暂无评论

Improved Prototypical Networks for remote sensing Scene Classification

Improved Prototypical Networks for Remote Sensing Scene Clas...

引用

2023 International Conference on image, Signal processing, and pattern recognition, ISPP 2023

作者： Zhou, Yujin Zhang, Xiang Li, Jie Wang, Guoqing Xie, Yizhuang Beijing Key Laboratory of Embedded Real-Time Information Processing Technology Beijing Institute of Technology Beijing100081 China The Shanghai Aerospace Electronic Technology Institute Shanghai201109 China

ISBN: (纸本)9781510666351

In practical applications, remote sensing (RS) scene classification faces data shift problems, including novel class and data discrepancy problems. Due to these problems, it is difficult to obtain representative and discriminative features. Therefore, we propose improved prototypical networks (IPN) based on few-shot learning to solve data shift problems in RS scene classification. First, a novel and effective scheme is proposed, which is to introduce Vision Transformer (ViT) pre-trained on a large-scale dataset as the feature extractor of the prototypical networks. Based on the meta-task training framework, IPN can adapt well to RS scene classification tasks and obtain representative features. In addition, a novel loss function called self-distillation-based prototype loss is designed to obtain discriminative features by introducing inter-sample self-distillation and inter-layer self-distillation methods. Extensive experiments are conducted on several public RS scene datasets. Compared with the existing methods, the proposed method achieves an improvement of 4.18%-21.42%. Results demonstrate that the proposed method can effectively solve the data shift problems in RS scene classification. © 2023 SPIE.

关键词： Distillation

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：