Abstract:Landslides that take place in mountain cities tend to cause huge casualties and economic losses, and a precise survey of landslide areas is a critical task for disaster emergency. However, because of the complicated appearance of the nature, it is difficult to find a spatial regularity that only relates to landslides, thus landslides detection based on only spatial information or artificial features usually performs poorly. In this paper, an automated landslides detection approach that is aiming at mountain ci… Show more
“…Finding the extent of an existing landslide is difficult using this approach, as a landslide is better represented by a heterogeneous polygon (i.e., a collection of pixels). Detection of landslides activity using image correlation [55,56] and change detection [39,57] are also included in pixel-based methods, but they require a time-series of multi-temporal images.…”
Section: Pixel-based Methodsmentioning
confidence: 99%
“…Anantrasirichai et al [49] were able to use CNN for automatic detection of volcanic ground deformation from Sentinel-1 images. In another study, Chen et al [39] have used CNN to identify areas which have changed in a stack of bi-temporal images, and subsequently used spatio-temporary context analysis to identify landslides. Ghorbanzadeh et al [29] compared different machine learning methods along with CNN for landslide detection in the higher Himalayas.…”
Mapping landslides using automated methods is a challenging task, which is still largely done using human efforts. Today, the availability of high-resolution EO data products is increasing exponentially, and one of the targets is to exploit this data source for the rapid generation of landslide inventory. Conventional methods like pixel-based and object-based machine learning strategies have been studied extensively in the last decade. In addition, recent advances in CNN (convolutional neural network), a type of deep-learning method, has been widely successful in extracting information from images and have outperformed other conventional learning methods. In the last few years, there have been only a few attempts to adapt CNN for landslide mapping. In this study, we introduce a modified U-Net model for semantic segmentation of landslides at a regional scale from EO data using ResNet34 blocks for feature extraction. We also compare this with conventional pixel-based and object-based methods. The experiment was done in Douglas County, a study area selected in the south of Portland in Oregon, USA, and landslide inventory extracted from SLIDO (Statewide Landslide Information Database of Oregon) was considered as the ground truth. Landslide mapping is an imbalanced learning problem with very limited availability of training data. Our network was trained on a combination of focal Tversky loss and cross-entropy loss functions using augmented image tiles sampled from a selected training area. The deep-learning method was observed to have a better performance than the conventional methods with an MCC (Matthews correlation coefficient) score of 0.495 and a POD (probability of detection) rate of 0.72 .
“…Finding the extent of an existing landslide is difficult using this approach, as a landslide is better represented by a heterogeneous polygon (i.e., a collection of pixels). Detection of landslides activity using image correlation [55,56] and change detection [39,57] are also included in pixel-based methods, but they require a time-series of multi-temporal images.…”
Section: Pixel-based Methodsmentioning
confidence: 99%
“…Anantrasirichai et al [49] were able to use CNN for automatic detection of volcanic ground deformation from Sentinel-1 images. In another study, Chen et al [39] have used CNN to identify areas which have changed in a stack of bi-temporal images, and subsequently used spatio-temporary context analysis to identify landslides. Ghorbanzadeh et al [29] compared different machine learning methods along with CNN for landslide detection in the higher Himalayas.…”
Mapping landslides using automated methods is a challenging task, which is still largely done using human efforts. Today, the availability of high-resolution EO data products is increasing exponentially, and one of the targets is to exploit this data source for the rapid generation of landslide inventory. Conventional methods like pixel-based and object-based machine learning strategies have been studied extensively in the last decade. In addition, recent advances in CNN (convolutional neural network), a type of deep-learning method, has been widely successful in extracting information from images and have outperformed other conventional learning methods. In the last few years, there have been only a few attempts to adapt CNN for landslide mapping. In this study, we introduce a modified U-Net model for semantic segmentation of landslides at a regional scale from EO data using ResNet34 blocks for feature extraction. We also compare this with conventional pixel-based and object-based methods. The experiment was done in Douglas County, a study area selected in the south of Portland in Oregon, USA, and landslide inventory extracted from SLIDO (Statewide Landslide Information Database of Oregon) was considered as the ground truth. Landslide mapping is an imbalanced learning problem with very limited availability of training data. Our network was trained on a combination of focal Tversky loss and cross-entropy loss functions using augmented image tiles sampled from a selected training area. The deep-learning method was observed to have a better performance than the conventional methods with an MCC (Matthews correlation coefficient) score of 0.495 and a POD (probability of detection) rate of 0.72 .
“…Therefore, the conclusion made here could be biased. Other ways of expanding the current landslide catalogue can depend on automatic landslide detection methods based on remote sensing images (Nichol and Wong, 2005;Chen et al, 2018), internet new sources (as all landslides with a relevant impact on society will be reported on internet new sources), and automatic web data mining methods (Battistini et al, 2013;Goswami et al, 2018).…”
Section: Noahmentioning
confidence: 99%
“…One emerging area relies on modelling. Some studies have used modelled soil moisture data for landslide applications (Ponziani et al, 2012;Ciabatta et al, 2016;Zhao et al, 2019a, b). However, to our knowledge, there is a lack of existing studies using modelled soil moisture from state-of-theart land surface models (LSMs) for landslide studies, such as the Noah LSM (Ek et al, 2003) and the Community Land Model (CLM) (Oleson et al, 2010).…”
Abstract. This study assesses the usability of Weather Research and Forecasting (WRF)
model simulated soil moisture for landslide monitoring in the Emilia Romagna region, northern Italy, during the 10-year period between 2006 and 2015. In particular, three advanced land surface model (LSM) schemes (i.e. Noah, Noah-MP, and CLM4) integrated with the WRF are used to provide detailed multi-layer soil moisture information. Through the temporal evaluation with the single-point in situ soil moisture observations, Noah-MP is the only scheme that is able to simulate the large soil drying phenomenon close to the observations during the dry season, and it also has the highest correlation coefficient and the lowest RMSE at most soil layers. It is also demonstrated that a single soil moisture sensor located in a plain area has a high correlation with a significant proportion of the study area (even in the mountainous region 141 km away, based on the WRF-simulated spatial soil moisture information). The evaluation of the WRF rainfall estimation shows there is no distinct difference among the three LSMs, and their performances are in line with a published study for the central USA. Each simulated soil moisture product from the three LSM schemes is then used to build a landslide prediction model, and within each model, 17 different exceedance probability levels from 1 % to 50 % are adopted to determine the optimal threshold scenario (in total there are 612 scenarios). Slope degree information is also used to separate the study region into different groups. The threshold evaluation performance is based on the landslide forecasting accuracy using 45 selected rainfall events between 2014 and 2015. Contingency tables, statistical indicators, and receiver operating characteristic analysis for different threshold scenarios are explored. The results have shown that, for landslide monitoring, Noah-MP at the surface soil layer with 30 % exceedance probability provides the best landslide monitoring performance, with its hit rate at 0.769 and its false alarm rate at 0.289.
“…The method used in this study is Object-Based Image Analysis (OBIA) (Hay, Castilla, 2006) which was developed in order to do the automatic extraction of image features. OBIA had been used to extract regular features like buildings (Karna, Bhardwaj, 2014), irregular features like tree canopy (Gustafson et al, 2018) or landslides (Chen et al, 2018), and also in landuse land cover classification (Cai et al, 2019).…”
Abstract. Locating landslides and determining its extent is deemed an important task in estimating loss and damage and carry out mitigation works. As landslides are recurring phenomena in the research site, Siwalik Hills of western Nepal, freely available Sentinel-2 satellite images were considered to delineate landslides. The method employed in this process was Object-Based Image Analysis carried out in eCognition software using multiresolution segmentation algorithm. Parameters taken for segmentation were a scale of 20, the shape of 0.3, and compactness of 0.5. When a threshold value of < 0.35 in NDVI was used to distinguish landslides from image objects, some non-landslide objects were also selected. These false positives were removed successively using the threshold values on different bands, band ratios, slope information, hillshade and geometrical properties of image objects. There were altogether 264 landslides detected in the study area with size ranging from 300 m2 to 1675 m2 and landslide density of approximately 2 per km2. The accuracy, when compared to reference inventory, showed correctness and completeness measuring 80.28% and 66.27% respectively. These results showed semi-automatic landslide extraction was successful and Sentinel-2 can be used for similar tasks in other areas of Siwalik.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.