检索结果-内蒙古大学图书馆

Deep Learning Based Tree Detection and Counting for remotely Sensed images 1

9th International Conference on pattern recognition and Machine Intelligence (PReMI)

作者： Vora, Nisarg Dave, Devam Shah, Kathan Birla Inst Technol & Sci Pilani Rajasthan India Pandit Deendayal Energy Univ Gandhinagar India

ISBN: (数字)9783031127007

ISBN: (纸本)9783031126994;9783031127007

Trees are an essential part of the environment and some trees are important economic crops whose count in a specific region is an important factor in the prediction of the yield of the product which they give. The number of trees in a particular region helps to monitor the growing situation of trees. In this proposed method, tree detection has been done using a deep learning based framework and the counting of these trees has been done using remote sensing high-resolution images for two regions in the state of Uttarakhand, India. The trees in our areas of study are congested, often leading to an overlap of crowns. Two multi-spectral images have been provided for the paper. The first image has four channels namely Red, Green, Blue (RGB) and Near-Infrared (NIR). For the first image provided, a variety of manually interpreted samples for the training as well as the optimization of the convolutional neural network (CNN) have been used. Thereafter, using the sliding window technique, the prediction of the labels of the samples in the image dataset has been carried out. The proposed model provides a weighted accuracy of over 98% during training and validation. Additionally, the text analyzes the results obtained in case the near-infrared band is removed from this image with four channels (i.e. in second image).

关键词： Tree Detection Convolutional Neural Networks remote sensing U-Net Architecture Deep Learning

来源：评论

学校读者我要写书评

暂无评论

Study on Hyperspectral remote sensing images of GF-5 De-Blurring Based on Sparse Representation 2023

Study on Hyperspectral Remote Sensing Images of GF-5 De-Blur...

引用

5th International Symposium on Signal processing Systems, SSPS 2023

作者： Wang, Hong School of Electronic Information Sichuan Modern Vocational College Chengdu China

ISBN: (纸本)9798400700040

Deblurring high resolution remote sensing image is a very important problem in remote sensing research. In this paper, we propose a new deblurring algorithm for high-resolution remote sensing images (HSI) based on sparse representation. The purpose of this study is to apply compressed sensing measurement and reconstruction technology to realize the processing of remote sensing image, and discuss the Under what circumstances can CS achieve better results in remote sensing image processing. The algorithm uses fast gradient projection algorithm to achieve deblurring and retain the important ground information of the original image. Experiments on remote sensing images obtained by GF-5 show that the algorithm can filter the blurring of remote sensing images well and improve the peak-to-noise ratio (PSNR) of images. The algorithm has better performance than other sparse representation algorithms. This paper explores the application of dictionary learning theory and sparse decomposition in remote sensing image processing. By further extending the algorithm proposed in this paper and adding new constraints, remote sensing image restoration, target recognition, deblurring, fusion and so on can be carried out. © 2023 ACM.

关键词： image reconstruction

来源：评论

学校读者我要写书评

暂无评论

Triple Robustness Augmentation Local Features for multi-source image registration

引用

ISPRS JOURNAL OF PHOTOGRAMMETRY AND remote sensing 2023年第1期199卷 1-14页

作者： Wang, Changwei Xu, Lele Xu, Rongtao Xu, Shibiao Meng, Weiliang Wang, Ruisheng Zhang, Xiaopeng Chinese Acad Sci Inst Automat MAIS Beijing Peoples R China Beijing Univ Posts & Telecommun Sch Artificial Intelligence Beijing Peoples R China Chinese Acad Sci Technol & Engn Ctr Space Utilizat Key Lab Space Utilizat Beijing Peoples R China Univ Chinese Acad Sci Sch Artificial Intelligence Beijing Peoples R China Univ Calgary Dept Geomat Engn Calgary AB Canada

With the proliferation of a wide variety of sensors, accurate multi-source image registration is crucial for many remote sensing image processing tasks. However, the registration of multi-source images faces the challenges of rotations, scales, and domain transformations caused by significant differences in shooting time, viewing angle, and sensor imaging modes. To cope with this problem, we propose a deep learning-based registration method named TRFeat, which aims to comprehensively improve the rotation, scale, and cross-domain robustness of local features. First, we introduce a special circular sampling convolutional layer to replace the standard square convolutional layer, in order to enhance the rotational robustness of local features. Second, we design a scale pyramid backbone network architecture to improve the robustness of the network to scale transformations. Third, we promote the use of hypercolumn domain alignment loss to extract cross-domain robust local descriptors for images from different sources. In addition, we develop a novel keypoint detection training framework based on iterative refinement supervision to obtain repeatable and reliable keypoints localization in multi-source images. Finally, we conduct thorough experiments on five multi-source datasets. Extensive experimental results validate that our TRFeat outperforms other state-of-the-art hand-crafted (e.g. RIFT) and deep learning-based methods (e.g. ASLFeat). Specifically, our TRFeat achieves an MMA@3 of 76.08% on the HPatches dataset and an RMSE of 3.38 on the Xiang dataset. The code is available at https://***/vignywang/TRFeat.

关键词： Multi-source image registration image matching Domain robust local features

来源：评论

学校读者我要写书评

暂无评论

Dense Center Prediction for Centerline Extraction Based on BP Network 12

Dense Center Prediction for Centerline Extraction Based on B...

引用

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

作者： Gui, Chuanqi Hubei Key Laboratory of Optical Information and Pattern Recognition Wuhan430205 China Hubei Engineering Research Center of Video Image and HD Projection Wuhan430205 China School of Electrical and Information Wuhan Institute of Technology Wuhan430205 China

ISBN: (数字)9781510675001

ISBN: (纸本)9781510674998

In the line structured light three-dimensional measurement system, high-precision laser stripe centerline extraction is the key to improving measurement accuracy. The laser line extraction technology based on neural network can automatically detect and extract laser lines with high accuracy and robustness. In this paper, a BP neural network containing one hidden layers is constructed, and the centerline obtained through feature extraction of the network can effectively overcome the influence of noise interference. Compared with the gray-gravity method and the Steger algorithm, the method in this paper has the advantage of high accuracy and can meet the requirements for extracting the center of complex light strips. © 2024 SPIE.

关键词： Extraction

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 7

Diffusion Posterior Sampling for Remote Sensing Image Fusion

引用

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

作者： Zhang, Chenlin Wang, Zelong Han, Jialing College of science National University of Defense Technology Changsha China

ISBN: (纸本)9781839539848

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. © The Institution of Engineering & Technology 2023.

关键词： image fusion

来源：评论

学校读者我要写书评

暂无评论

Research on Intelligent recognition System of Crop Growth Monitoring by UAV remote sensing image 6

Research on Intelligent Recognition System of Crop Growth Mo...

引用

2023 IEEE 6th International Conference on Information Systems and Computer Aided Education, ICISCAE 2023

作者： Xie, Wenhuan Heilongjiang Academy of Agricultural Sciences Institute of Agricultural Remote Sensing and Information Harbin China

ISBN: (纸本)9798350313444

This paper introduces a remote sensing load data storage system of UAV based on SOPC. This paper presents a method of data storage using multiple solid-state drives. The hard disk can be read and written in parallel. Use SDRAM as a buffer for data. In order to realize effective processing of crop data, a superblock file system is adopted. A number of image experiments with overlapping areas were carried out using unmanned aircraft. A new remote sensing image processing algorithm is proposed. The algorithm uses the spectral features, geometric features and texture features of multiple targets for classification. Relief-Pearson features are used to reduce the dimensionality of images, and redundant features with strong correlation and recognition performance are removed. Finally, according to the selected features, a multi-target recognition method is established and compared with the results of manual recognition. Based on the experimental data collected in China's corn production experiment and demonstration base, it is found that the new algorithm can simulate the real shadow well, and the overall accuracy is 0.9473. The F1 value reached 0.9184. © 2023 IEEE.

关键词： crop monitoring intelligent recognition remote sensing image UAV photography

来源：评论

学校读者我要写书评

暂无评论

Mixed Pixel Saturability Based Area Estimation Model on remote sensing image 6

Mixed Pixel Saturability Based Area Estimation Model on Remo...

引用

6th IEEE International Conference on pattern recognition and Artificial Intelligence, PRAI 2023

作者： Lu, Kedi Ma, Zhong Huo, Pengcheng He, Zhao Zhang, Haochen Zheng, Xin Xi'An Microelectronics Technology Institute Xi'an China

ISBN: (纸本)9798350325485

remote sensing technology has been widely applied in multiple fields with surface monitors, and area estimation is one of key problems. Currently, researchers usually compute the area directly according to the number of pixels, which is often affected by factors such as image resolution, resulting in large errors in the area estimation. Therefore, in order to alleviate the drawback and improve the accuracy of area estimation, we propose a Mixed Pixel Saturability based Wide Region Area Estimation Model on remote sensing image Data (MPS), which can be utilized to estimate areas in wide regions. The proposed model firstly declares a new definition named mixed pixels saturability which adequately considers the structure of remote sensing images in spatial, further, a weighting process method is introduced to eliminate the influence of outliers and improve the generalization capability of the area estimation model. Then, we conduct experiments on cropland remote sensing image datasets, the results show that the proposed area estimation model outperforms the existing models. It improves the accuracy of the wide region remote sensing image area estimation and provides the basis for remote sensing-based area monitoring. © 2023 IEEE.

关键词： image resolution

来源：评论

学校读者我要写书评

暂无评论

MACHINE LEARNING AND FEATURE EXTRACTION FOR INDUSTRIAL SMOKE PLUMES DETECTION FROM SENTINEL-2 imageS

MACHINE LEARNING AND FEATURE EXTRACTION FOR INDUSTRIAL SMOKE...

引用

IEEE International Geoscience and remote sensing Symposium (IGARSS)

作者： Poucin, Florentin Ouerghi, Elyes Lajouanie, Simon Rodrigues, Hugo de Almeida Facciolo, Gabriele de Franchis, Carlo Hessel, Charles Kayrros SAS Paris France Univ Paris Saclay CNRS Ctr Borelli ENS Paris Saclay Paris France

ISBN: (纸本)9798350320107

The detection of smoke plumes by satellite imagery is a comprehensive research topic that can be used to better monitor activity and emissions from the energy and industrial sectors. In this study, we propose a machine learning methodology based on the extraction of relevant features from Sentinel-2 images to perform industrial smoke plume detection. This computer vision problem is modeled as an image classification task based on the presence or absence of plumes from previously identified sources. A dataset of nearly 17,000 hand-labeled images of smoke plumes for activity classification has been compiled to train and evaluate our detection models. The final Gradient Boosting model only uses the 3 RGB bands of Sentinel-2 and after a post-processing step reaches an accuracy of 95%.

关键词： Industrial plumes Sentinel-2 images Machine learning Feature extraction pattern recognition

来源：评论

学校读者我要写书评

暂无评论

BFRNet: Bimodal Fusion and Rectification Network for remote sensing Semantic Segmentation 7th

BFRNet: Bimodal Fusion and Rectification Network for Remote ...

引用

7th Chinese Conference on pattern recognition and Computer Vision

作者： Weng, Qian Lin, Yifeng Pan, Zengying Lin, Jiawen Chen, Gengwei Chen, Mo Zhang, Yingying Fuzhou Univ Coll Comp & Data Sci Fuzhou Peoples R China Fuzhou Univ Fujian Key Lab Network Comp & Intelligent Informa Fuzhou Peoples R China

ISBN: (纸本)9789819784929;9789819784936

In the semantic segmentation of high-resolution remote sensing images, utilizing the normalized Digital Surface Model (nDSM) that provides height information as auxiliary data and fusing it with the visible image can improve the accuracy of segmentation. However, the better utilization of complementarity between different modal features has not been fully explored. In this work, we propose a new dual-branch and multi-stage Bimodal Fusion Rectification Network (BFRNet), which is end-to-end trainable. It consists of three modules: Channel and Spatial Fusion Rectification (CSFR) module, Edge Fusion Refinement (EFR) module, and Multiscale Feature Fusion (MSFF) module. The CSFR module integrates and rectifies multimodal features in both channel and spatial dimensions, achieving sufficient interaction and fusion between multimodal features. The EFR module obtains better multiscale edge features than single modality through feature fusion based on bimodal interactive edge attention and spatial gate, which helps to alleviate the edge loss of ground objects in single modality. The MSFF module is used to upsample and fuse multiscale features from EFR and CSFR to generate the final semantic segmentation results. The experimental results on the two public datasets, Vaihingen and Potsdam, provided by ISPRS, showcase the comparative advantage of the proposed method over other research methods.

关键词： Semantic Segmentation Modal Fusion Feature Rectification remote sensing

来源：评论

学校读者我要写书评

暂无评论

Analysis of land use/land cover changes and driving forces during the period 1992-2022: a case study of Jinan City, China

引用

AQUA-WATER INFRASTRUCTURE ECOSYSTEMS AND SOCIETY 2024年第3期73卷 584-607页

作者： Tan, Lingye Tiong, Robert L. K. Zhang, Ziyang Nanyang Technol Univ Sch Civil & Environm Engn Singapore 639798 Singapore

This research aims to quantify the spatial pattern of urban land use/land cover (LULC) change while considering environmental effects. This paper integrates historical Landsat imagery, a remote sensing image processing platform (ENVI), geographical information system (GIS), and socioeconomic data to determine the spatial-temporal urban LULC dynamics and the conversion of LULC in response to the rapid urbanization from 1992 to 2022. Principle component analysis and multiple linear regression are used to determine and model the relationship between the socioeconomic factors and the changes for identifying the driving forces. The results indicate that impervious surfaces have exponentially increased, expanding more than two times from 2,348 to 4,795 km(2), in contrast to bare lands, which drastically declined by 95%, from 1,888 to 87 km(2). Water bodies have always been relatively fewer, at approximately 100 km2. In addition, the majority of farmland in Jinan City is concentrated in the northern region with a steady area in the range of 2,100-2,900 km(2), while the majority of woodland located in the southern region declined from 3,774.52 km2 (37%) to 3,088.28 km(2) (30%). Economic development, population growth, and climate change are the primary factors that have an obvious impact on LULC changes.

关键词： driving factors land use/land cover dynamic change remote sensing spatial pattern transition

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：