River Extraction From High-Resolution SAR Images Combining a Structural Feature Set and Mathematical Morphology

Sghaier, Moslem Ouled; Foucher, Samuel; Lepage, Richard

doi:10.1109/jstars.2016.2609804

Cited by 44 publications

(28 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…This theory was applied in this work at two levels: first, to combine a spatial-based and a radiometric-based sources of information resulting from the application of morphological profile and evidential C-means and second, to fuse mass function obtained from each image using the multidimensional evidential reasoning. The initial surface of the river or of the lake can be identified directly by applying the first two steps on the pre-flooding image, as we have already demonstrated in previous works [29,30]. For the sake of simplification, we will, in what follows, present the intermediate results obtained by applying each step of our algorithm to the two input images shown in Figure 3a,b corresponding to Richelieu River pre-and post flooding SAR images.…”

Section: Overviewmentioning

confidence: 98%

“…In previous works, we successfully applied the SFS texture measurement to extract homogeneous areas such as road networks from optical images [31,32], and water bodies from SAR images [29,30]. This descriptor has shown very promising results applied to optical images characterized by the presence of various artifacts arising from the use of high spatial resolution and SAR images affected by speckle noise.…”

Section: Texture Extraction Using a Structural Feature Set (Sfs)mentioning

confidence: 99%

See 1 more Smart Citation

Flood Extent Mapping from Time-Series SAR Images Based on Texture Analysis and Data Fusion

et al. 2018

Self Cite

View full text Add to dashboard Cite

Nowadays, satellite images are considered as one of the most relevant sources of information in the context of major disasters management. Their availability in extreme weather conditions and their ability to cover wide geographic areas make them an indispensable tool toward an effective disaster response. Among the various available sensors, Synthetic Aperture Radar (SAR) is distinguished in the context of flood management by its ability to penetrate cloud cover and its robustness to unfavourable weather conditions. This work aims at developing a new technique for flooded areas extraction from high resolution time-series SAR images. The proposed approach is mainly based on three steps: first, homogeneous regions characterizing water surfaces are extracted from each SAR image using a local texture descriptor. Then, mathematical morphology is applied to filter tiny artifacts and small homogeneous areas present in the image. And finally, spatial and radiometric information embedded in each pixel are extracted and are fused with the same pixel information but from another image to decide if the current pixel belongs to a flooded region. In order to assess the performance of the proposed algorithm, our methodology was applied to time-series images acquired before and during three different flooding events: (1) Richelieu River and lake Champlain floods, Quebec, Canada in 2011; (2) Evros River floods, Greece in 2014 and (3) Western and southwestern of Iran floods in 2016. Experiments show that our approach gives very promising results compared to existing techniques.

show abstract

Section: Overviewmentioning

confidence: 98%

Section: Texture Extraction Using a Structural Feature Set (Sfs)mentioning

confidence: 99%

Flood Extent Mapping from Time-Series SAR Images Based on Texture Analysis and Data Fusion

et al. 2018

Self Cite

View full text Add to dashboard Cite

show abstract

“…Being a time-saving method, deep learning can implicitly learn features and generate abstractions from inputs, mirroring what humans trained in specialized fields do, which does not require the end-user to design a proper set of features for a specific field, making its application relatively easy to road extraction. The deep neural network (DNN), a popular deep learning application (Sghaier, Foucher, & Lepage, 2017;Xu, Wang, Zhang, Li, & Zhang, 2017;Zhou, Wang, Xu, & Jin, 2016), exhibits particularly high efficiency in SAR image classification. Feature extraction processes have been integrated into DNN.…”

Section: Introductionmentioning

confidence: 99%

A new road extraction method using Sentinel-1 SAR images based on the deep fully convolutional neural network

Zhang

Kong

Zhang

et al. 2019

European Journal of Remote Sensing

View full text Add to dashboard Cite

There is a pressing need for an automatic road extraction method due to the continuous development of transportation networks. Free from the influence of weather, satellitemounted synthetic-aperture radar (SAR) opens the way for such a road detection application. This study introduces an automatic discrimination method based on a deep neural network (DNN) adjusted for roads from dual-polarimetric (VV and VH) Sentinel-1 SAR imagery. In this proposed method, the U-Net extended the convolutional neural network (CNN), is adjusted for road extraction from SAR images. To investigate the potential of using the U-Net based fully convolutional neural network (FCN) for road extraction, LeNet-5, is applied as a preliminary DNN model. Additionally, several training optimizations are introduced to improve accuracy. To assess the performance of the different polarization modes used in road extraction, both single-polarimetric and dual-polarimetric data were investigated. Moreover, four machine learning algorithms have been compared for accuracy and speed. As a result, the outcome evaluation of Precision, Recall, and F 1 obtained by FCN is better than the original CNN, and the training time has been significantly reduced. The DNN model (CNN and FCN) is superior to machine learning methods in accuracy and elapsed computation time.

show abstract

“…However, looking at the results of existing researches, they mostly ignore the estimation of waterinvasion extent and only regard water-body extent as inundation extent because it is difficult to describe waterinvasion extent which is related to the boundary of water, terrains, and landforms. us, incorporating the Mathematical Morphology [24], we proposed an inundation estimation method by combined space-filling method and RS data. e method divided inundation extent into water-body extent and water-invasion extent with high timeliness and accuracy.…”

Section: Space-filling Methodmentioning

confidence: 99%

Estimation Method of Inundation Extent Based on Mathematical Morphology

Lai

Safonov

2019

Mathematical Problems in Engineering

View full text Add to dashboard Cite

Estimation of inundation extent is critical for the flood control system. It helps decision-maker determine the scope of damage to save residents' lives and property. In this paper, we analyzed the drawbacks of two kinds of methods of inundation estimation, hydrodynamic method and space-filling method, and divided inundation extent into water-body extent and water-invasion extent for the purpose of enhancing estimation accuracy. Based on Mathematical Morphology, an inundation estimation method suitable for flood mapping around natural and artificial lakes was put forward by integrating remote sensing (RS) data and digital elevation model (DEM) data. is method consists of three aspects, including the subdivision of water-body extent, the deduction of water-invasion extent which is viewed as the expansion of water-body extent, and the smoothing the boundaries of waterinvasion extent. With subdividing DEM data horizontally into cross sections to build equiangular subsets, the expansion operation was implemented by structuring elements which were established based on set operations and scalar flexibilities. us, water-invasion extent was easily described by the structuring elements under given water volume, so that inundation extent was conveniently estimated by spatial data. e experiment results show that the method has the advantages of high accuracy and expressiveness and low time consumption. It had been used to analyze the inundation caused by the water conservancy project named Hanjiang-to-Weihe River Water Transfer Project in China. e method served as decision support for building an efficient emergency response mechanism and helped implement flood disaster forecast and simulation.

show abstract

River Extraction From High-Resolution SAR Images Combining a Structural Feature Set and Mathematical Morphology

Cited by 44 publications

References 37 publications

Flood Extent Mapping from Time-Series SAR Images Based on Texture Analysis and Data Fusion

Flood Extent Mapping from Time-Series SAR Images Based on Texture Analysis and Data Fusion

A new road extraction method using Sentinel-1 SAR images based on the deep fully convolutional neural network

Estimation Method of Inundation Extent Based on Mathematical Morphology

Contact Info

Product

Resources

About