scholarly journals Deep Hybrid Network for Land Cover Semantic Segmentation in High-Spatial Resolution Satellite Images

Information ◽  
2021 ◽  
Vol 12 (6) ◽  
pp. 230
Author(s):  
Sultan Daud Khan ◽  
Louai Alarabi ◽  
Saleh Basalamah
Keyword(s):  
High Resolution ◽  
Land Cover ◽  
Spatial Resolution ◽  
Satellite Images ◽  
High Spatial Resolution ◽  
Multiple Scales ◽  
Semantic Segmentation ◽  
Vital Role ◽  
Hybrid Network ◽  
High Resolution Satellite Images

Land cover semantic segmentation in high-spatial resolution satellite images plays a vital role in efficient management of land resources, smart agriculture, yield estimation and urban planning. With the recent advancement in remote sensing technologies, such as satellites, drones, UAVs, and airborne vehicles, a large number of high-resolution satellite images are readily available. However, these high-resolution satellite images are complex due to increased spatial resolution and data disruption caused by different factors involved in the acquisition process. Due to these challenges, an efficient land-cover semantic segmentation model is difficult to design and develop. In this paper, we develop a hybrid deep learning model that combines the benefits of two deep models, i.e., DenseNet and U-Net. This is carried out to obtain a pixel-wise classification of land cover. The contraction path of U-Net is replaced with DenseNet to extract features of multiple scales, while long-range connections of U-Net concatenate encoder and decoder paths are used to preserve low-level features. We evaluate the proposed hybrid network on a challenging, publicly available benchmark dataset. From the experimental results, we demonstrate that the proposed hybrid network exhibits a state-of-the-art performance and beats other existing models by a considerable margin.

scholarly journals Mapping Land Use from High Resolution Satellite Images by Exploiting the Spatial Arrangement of Land Cover Objects

2020 ◽  
Vol 12 (24) ◽  
pp. 4158
Author(s):  
Mengmeng Li ◽  
Alfred Stein
Keyword(s):  
Remote Sensing ◽  
Land Use ◽  
High Resolution ◽  
Land Cover ◽  
Spatial Resolution ◽  
Satellite Images ◽  
Spatial Arrangement ◽  
Remote Sensing Images ◽  
Land Use Types ◽  
High Resolution Satellite Images

Spatial information regarding the arrangement of land cover objects plays an important role in distinguishing the land use types at land parcel or local neighborhood levels. This study investigates the use of graph convolutional networks (GCNs) in order to characterize spatial arrangement features for land use classification from high resolution remote sensing images, with particular interest in comparing land use classifications between different graph-based methods and between different remote sensing images. We examine three kinds of graph-based methods, i.e., feature engineering, graph kernels, and GCNs. Based upon the extracted arrangement features and features regarding the spatial composition of land cover objects, we formulated ten land use classifications. We tested those on two different remote sensing images, which were acquired from GaoFen-2 (with a spatial resolution of 0.8 m) and ZiYuan-3 (of 2.5 m) satellites in 2020 on Fuzhou City, China. Our results showed that land use classifications that are based on the arrangement features derived from GCNs achieved the highest classification accuracy than using graph kernels and handcrafted graph features for both images. We also found that the contribution to separating land use types by arrangement features varies between GaoFen-2 and ZiYuan-3 images, due to the difference in the spatial resolution. This study offers a set of approaches for effectively mapping land use types from (very) high resolution satellite images.

scholarly journals AUTOMATED PANSHARPENING INFORMATION TECHNOLOGY OF SATELLITE IMAGES

2021 ◽  
pp. 123-132
Author(s):  
V. V. Hnatushenko ◽  
V. Yu. Kashtan
Keyword(s):  
Image Processing ◽  
Information Technology ◽  
High Resolution ◽  
Spatial Resolution ◽  
Satellite Images ◽  
High Spatial Resolution ◽  
Image Resolution ◽  
Resolution Image ◽  
Ndvi Index ◽  
High Resolution Satellite Images

Context. Nowadays, information technologies are widely used in digital image processing. The task of joint processing of satellite image obtained by different space systems that have different spatial differences is important. The already known pansharpening methods to improve the quality of the resulting image, there are new scientific problems associated with increasing the requirements for high-resolution image processing and the development of automated technology for processing the satellite data for further thematic analysis. Most spatial resolution techniques result in artifacts. Our work explores the major remote sensing data fusion techniques at pixel level and reviews the concept, principals, limitations and advantages for each technique with the program implementation of research. Objective. The goal of the work is analyze the effectiveness of the traditional pan-sharpening methods like the Brovey, the wavelet-transform, the GIHS, the HCT and the combined pansharpening method for satellite images of high-spatial resolution. Method. In this paper we propose the information technology for pansharpening high spatial resolution images with automation of choosing the best method based on the analysis of quantitative and qualitative evolutions. The method involves the scaling multispectral image to the size of the panchromatic image; using histogram equalization to adjust the primary images by aligning the integral areas of the sections with different brightness; conversion of primary images after the spectral correction on traditional pansharpening methods; analyze the effectiveness of the results obtained for conducts their detailed comparative visual and quantitative evaluation. The technology allows determining the best method of pansharpening by analyzing quantitative metrics: the NDVI index, the RMSE and the ERGAS. The NDVI index for the methods Brovey and HPF indicate color distortion in comparison with the reference data. This is due to the fact that the Brovey and HPF methods are based on the fusion of three channel images and do not include the information contained in the near infrared range. The RMSE and the ERGAS show the superiority of the combined HSVHCT-Wavelet method over the conventional and state-of-art image resolution enhancement techniques of high resolution satellite images. This is achieved, in particular, by preliminary processing of primary images, data processing localized spectral bases, optimized performance information, and the information contained in the infrared image. Results. The software implementing proposed method is developed. The experiments to study the properties of the proposed algorithm are conducted. Experimental evaluation performed on eight-primary satellite images of high spatial resolution obtained WorldView-2 satellite. The experimental results show that a synthesized high spatial resolution image with high information content is achieved with the complex use of fusion methods, which makes it possible to increase the spatial resolution of the original multichannel image without color distortions. Conclusions. The experiments confirmed the effectiveness of the proposed automated information technology for pansharpening high-resolution satellite images with the development of a graphical interface to obtain a new synthesized image. In addition, the proposed technology will effectively carry out further recognition and real-time monitoring infrastructure.

scholarly journals GRANULOMETRIC MAPS FROM HIGH RESOLUTION SATELLITE IMAGES

2011 ◽  
Vol 21 (1) ◽  
pp. 19 ◽  
Author(s):  
Catherine Mering ◽  
Franck Chopin
Keyword(s):  
High Resolution ◽  
Land Cover ◽  
Arid Regions ◽  
Satellite Images ◽  
Distribution Analysis ◽  
Granulometric Analysis ◽  
High Resolution Satellite Images ◽  
Thematic Mapping ◽  
Urban Settlements

A new method of land cover mapping from satellite images using granulometric analysis is presented here. Discontinuous landscapes such as steppian bushes of semi arid regions and recently growing urban settlements are especially concerned by this study. Spatial organisations of the land cover are quantified by means of the size distribution analysis of the land cover units extracted from high resolution remotely sensed images. A granulometric map is built by automatic classification of every pixel of the image according to the granulometric density inside a sliding neighbourhood. Granulometric mapping brings some advantages over traditional thematic mapping by remote sensing by focusing on fine spatial events and small changes in one peculiar category of the landscape.

scholarly journals Agricultural Greenhouses Detection in High-Resolution Satellite Images Based on Convolutional Neural Networks: Comparison of Faster R-CNN, YOLO v3 and SSD

Sensors ◽  
2020 ◽  
Vol 20 (17) ◽  
pp. 4938
Author(s):  
Min Li ◽  
Zhijie Zhang ◽  
Liping Lei ◽  
Xiaofan Wang ◽  
Xudong Guo
Keyword(s):  
High Resolution ◽  
Visual Inspection ◽  
Satellite Images ◽  
High Spatial Resolution ◽  
Fine Tuning ◽  
Single Shot ◽  
Modern Agriculture ◽  
High Resolution Satellite Images ◽  
Independent Test ◽  
Test Sets

Agricultural greenhouses (AGs) are an important facility for the development of modern agriculture. Accurately and effectively detecting AGs is a necessity for the strategic planning of modern agriculture. With the advent of deep learning algorithms, various convolutional neural network (CNN)-based models have been proposed for object detection with high spatial resolution images. In this paper, we conducted a comparative assessment of the three well-established CNN-based models, which are Faster R-CNN, You Look Only Once-v3 (YOLO v3), and Single Shot Multi-Box Detector (SSD) for detecting AGs. The transfer learning and fine-tuning approaches were implemented to train models. Accuracy and efficiency evaluation results show that YOLO v3 achieved the best performance according to the average precision (mAP), frames per second (FPS) metrics and visual inspection. The SSD demonstrated an advantage in detection speed with an FPS twice higher than Faster R-CNN, although their mAP is close on the test set. The trained models were also applied to two independent test sets, which proved that these models have a certain transability and the higher resolution images are significant for accuracy improvement. Our study suggests YOLO v3 with superiorities in both accuracy and computational efficiency can be applied to detect AGs using high-resolution satellite images operationally.

scholarly journals Semantic Segmentation-Based Building Footprint Extraction Using Very High-Resolution Satellite Images and Multi-Source GIS Data

2019 ◽  
Vol 11 (4) ◽  
pp. 403 ◽  
Author(s):  
Weijia Li ◽  
Conghui He ◽  
Jiarui Fang ◽  
Juepeng Zheng ◽  
Haohuan Fu ◽  
...  
Keyword(s):  
High Resolution ◽  
Satellite Images ◽  
Data Augmentation ◽  
Semantic Segmentation ◽  
Building Detection ◽  
Building Extraction ◽  
Three Solutions ◽  
High Resolution Satellite Images ◽  
Building Footprint ◽  
Map Data

Automatic extraction of building footprints from high-resolution satellite imagery has become an important and challenging research issue receiving greater attention. Many recent studies have explored different deep learning-based semantic segmentation methods for improving the accuracy of building extraction. Although they record substantial land cover and land use information (e.g., buildings, roads, water, etc.), public geographic information system (GIS) map datasets have rarely been utilized to improve building extraction results in existing studies. In this research, we propose a U-Net-based semantic segmentation method for the extraction of building footprints from high-resolution multispectral satellite images using the SpaceNet building dataset provided in the DeepGlobe Satellite Challenge of IEEE Conference on Computer Vision and Pattern Recognition 2018 (CVPR 2018). We explore the potential of multiple public GIS map datasets (OpenStreetMap, Google Maps, and MapWorld) through integration with the WorldView-3 satellite datasets in four cities (Las Vegas, Paris, Shanghai, and Khartoum). Several strategies are designed and combined with the U-Net–based semantic segmentation model, including data augmentation, post-processing, and integration of the GIS map data and satellite images. The proposed method achieves a total F1-score of 0.704, which is an improvement of 1.1% to 12.5% compared with the top three solutions in the SpaceNet Building Detection Competition and 3.0% to 9.2% compared with the standard U-Net–based method. Moreover, the effect of each proposed strategy and the possible reasons for the building footprint extraction results are analyzed substantially considering the actual situation of the four cities.

scholarly journals An Approach for the Pan Sharpening of Very High Resolution Satellite Images Using a CIELab Color Based Component Substitution Algorithm

2019 ◽  
Vol 9 (23) ◽  
pp. 5234 ◽  
Author(s):  
Rahimzadeganasl ◽  
Alganci ◽  
Goksel
Keyword(s):  
Spatial Resolution ◽  
Satellite Images ◽  
Spatial Information ◽  
High Spatial Resolution ◽  
Critical Role ◽  
Study Region ◽  
High Resolution Satellite Images ◽  
Very High ◽  
Component Substitution ◽  
Cielab Color

Recent very high spatial resolution (VHR) remote sensing satellites provide high spatial resolution panchromatic (Pan) images in addition to multispectral (MS) images. The pan sharpening process has a critical role in image processing tasks and geospatial information extraction from satellite images. In this research, CIELab color based component substitution Pan sharpening algorithm was proposed for Pan sharpening of the Pleiades VHR images. The proposed method was compared with the state-of-the-art Pan sharpening methods, such as IHS, EHLERS, NNDiffuse and GIHS. The selected study region included ten test sites, each of them representing complex landscapes with various land categories, to evaluate the performance of Pan sharpening methods in varying land surface characteristics. The spatial and spectral performance of the Pan sharpening methods were evaluated by eleven accuracy metrics and visual interpretation. The results of the evaluation indicated that proposed CIELab color-based method reached promising results and improved the spectral and spatial information preservation.

scholarly journals Image Fusion-Based Land Cover Change Detection Using Multi-Temporal High-Resolution Satellite Images

2017 ◽  
Vol 9 (8) ◽  
pp. 804 ◽  
Author(s):  
Biao Wang ◽  
Jaewan Choi ◽  
Seokeun Choi ◽  
Soungki Lee ◽  
Penghai Wu ◽  
...  
Keyword(s):  
High Resolution ◽  
Land Cover ◽  
Change Detection ◽  
Image Fusion ◽  
Land Cover Change ◽  
Satellite Images ◽  
High Resolution Satellite Images ◽  
Multi Temporal ◽  
Land Cover Change Detection

scholarly journals PROPOSAL OF CHANGE DETECTION IN LAND COVER FOR PIPELINE MONITORING FROM RAPIDEYE AND IKONOS IMAGES

2014 ◽  
Vol 32 (4) ◽  
pp. 655 ◽  
Author(s):  
Paulina Setti Riedel ◽  
Mara Lúcia Marques ◽  
Mateus Vidotti Ferreira ◽  
Marcelo Elias Delaneze
Keyword(s):  
High Resolution ◽  
Land Cover ◽  
Change Detection ◽  
Satellite Images ◽  
Bare Soil ◽  
Reference Limit ◽  
Object Based ◽  
High Resolution Satellite Images ◽  
Pipeline Monitoring ◽  
Structural Preservation

ABSTRACT. The goal of this study was to improve and evaluate the applicability of a methodological procedure of pipeline monitoring to reveal indicators of thirdparty activities that may interfere with the structural preservation of pipes and environmental damages. The procedure was developed from the technique of changedetection through object-based classification of land cover, using high resolution satellite images applied to a section of the Guararema-Mauá – São Paulo pipeline, Brazil. In the seven-month monitoring period performed with RapidEye imaging, an area of 2.024 km2 was identified as area of change, corresponding to 3.30% of thetotal area analyzed. For the monitoring performed with Ikonos imaging during a four-month period, changes were detected in an area of 0.187 km2, which correspondedto 1.92% of the total area analyzed. The main changes in land cover were from Bare Soil to Grassland, due to changes related to the different stages of agriculturalactivity and reforestation areas, as well as the natural regeneration of vegetation over the pipeline and solid waste landfill. The results of the change detection of landcover from object-based classification were close to the technique reference limit for areas with great complexity and diversity of space occupation.Keywords: structural preservation of pipes, object-based classification, high resolution satellite images. RESUMO. Este estudo teve por objetivo avaliar a aplicabilidade de um procedimento metodológico de monitoramento de faixas de dutos que revelem indicativos deatividades de terceiros que podem interferir na integridade estrutural dos dutos e provocar danos ambientais. O procedimento foi desenvolvido a partir da técnica dedetecção de mudanças na cobertura da terra pela classificação baseada no objeto, com utilização de imagens orbitais de alta resolução. Este procedimento foi empregadoem um trecho da faixa de dutos Guararema-Mauá – SP, no monitoramento realizado por meio de imagens RapidEye. Em um período de sete meses, foram identificados 2,024 km2 como área de mudança, que corresponde a 3,30% do total da área analisada. Para o monitoramento realizado a partir da imagem Ikonos, com período de quatro meses, foi identificada como mudança uma área de 0,187 km2, correspondendo a 1,92% do total da área analisada. As principais mudanças ocorridas foramentre Solo Exposto e Vegetaçao Rasteira, devido às alterações ocorridas nos estágios de cultivo agrícola e áreas de reflorestamento, como também, estão associadas às áreas de regeneração da vegetação da faixa de dutos e aterro sanitário. Os resultados da detecção de mudanças da cobertura da terra pela classificação baseada no objeto atingiram acertos próximos ao limite de para esta técnica, em áreas com grande complexidade e diversidade de ocupação do espaço.Palavras-chave: integridade estrutural dos dutos, classificação baseada no objeto, imagens orbitais de alta resolução.

Sign in / Sign up

Export Citation Format

Share Document