Artificial neural networks and geographic information systems for inland excess water classification

Leeuwen, Boudewijn van

doi:10.14232/phd.1520

Cited by 6 publications

(9 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the Carpathian Basin drought is one of the most severe environmental hazards (Bakonyi 2010;Pálfai and Herceg 2011;Bihari 2012;Gosic and Trajkovic 2013;WMO 2013) that occurs every 3-5 years. Excess water affects mostly lowland areas, causing damages to agriculture, soil structure, and inundated urban areas (Likens 2009;Rakonczai et al 2011;van Leeuwen 2012;Julian et al 2013;Shi et al 2013). Excess water occurs regularly in the Carpathian Basin, typically at the end of winter and in spring, but during summer as well every 2-4 years.…”

Section: Introductionmentioning

confidence: 99%

Climate Change Impacts on Environmental Hazards on the Great Hungarian Plain, Carpathian Basin

Mezősi

Bata

Meyer

et al. 2014

Int J Disaster Risk Sci

View full text Add to dashboard Cite

The potential impacts of climate change on the Great Hungarian Plain based on two regional climate models, REMO and ALADIN, were analyzed using indicators for environmental hazards. As the climate parameters (temperature, precipitation, and wind) will change in the two investigated periods (2021-2050 and 2071-2100), their influences on drought, wind erosion, and inland excess water hazards are modeled by simple predictive models. Drought hazards on arable lands will increasingly affect the productivity of agriculture compared to the reference period . The models predict an increase between 12.3 % (REMO) and 20 % (ALADIN) in the first period, and between 35.6 % (REMO) and 45.2 % (ALA-DIN) in the second period. The increase of wind erosion hazards is not as obvious (?15 % for the first period in the REMO model). Inland excess water hazards are expected to be slightly reduced (-4 to 0 %) by both model predictions in the two periods without showing a clear tendency on reduction. All three indicators together give a first regional picture of potential hazards of climate change. The predictive model and data combinations of the regional climate change models and the hazard assessment models provide insights into regional and subregional impacts of climate change and will be useful in planning and land management activities.

show abstract

Section: Introductionmentioning

confidence: 99%

Climate Change Impacts on Environmental Hazards on the Great Hungarian Plain, Carpathian Basin

Mezősi

Bata

Meyer

et al. 2014

Int J Disaster Risk Sci

View full text Add to dashboard Cite

show abstract

“…Finally, an ANN approach was examined in this study. ANNs mimic the functionality of biological neurons and they are extensively used in speech/face recognition, and also in landscape classification studies [33,34]. They possess a number of advantages that distinguish them from common algorithms, whereas they also hold some disadvantages.…”

Section: Supervised Machine Learning Algorithmsmentioning

confidence: 99%

The Hyper-Angular Cube Concept for Improving the Spatial and Acoustic Resolution of MBES Backscatter Angular Response Analysis

Alevizos

Greinert

2018

Geosciences

View full text Add to dashboard Cite

This study presents a novel approach, based on high-dimensionality hydro-acoustic data, for improving the performance of angular response analysis (ARA) on multibeam backscatter data in terms of acoustic class separation and spatial resolution. This approach is based on the hyper-angular cube (HAC) data structure which offers the possibility to extract one angular response from each cell of the cube. The HAC consists of a finite number of backscatter layers, each representing backscatter values corresponding to single-incidence angle ensonifications. The construction of the HAC layers can be achieved either by interpolating dense soundings from highly overlapping multibeam echo-sounder (MBES) surveys (interpolated HAC, iHAC) or by producing several backscatter mosaics, each being normalized at a different incidence angle (synthetic HAC, sHAC). The latter approach can be applied to multibeam data with standard overlap, thus minimizing the cost for data acquisition. The sHAC is as efficient as the iHAC produced by actual soundings, providing distinct angular responses for each seafloor type. The HAC data structure increases acoustic class separability between different acoustic features. Moreover, the results of angular response analysis are applied on a fine spatial scale (cell dimensions) offering more detailed acoustic maps of the seafloor. Considering that angular information is expressed through high-dimensional backscatter layers, we further applied three machine learning algorithms (random forest, support vector machine, and artificial neural network) and one pattern recognition method (sum of absolute differences) for supervised classification of the HAC, using a limited amount of ground truth data (one sample per seafloor type). Results from supervised classification were compared with results from an unsupervised method for inter-comparison of the supervised algorithms. It was found that all algorithms (regarding both the iHAC and the sHAC) produced very similar results with good agreement (>0.5 kappa) with the unsupervised classification. Only the artificial neural network required the total amount of ground truth data for producing comparable results with the remaining algorithms.

show abstract

“…Moreover, artificial neural network (ANN) analysis which can be applied to landslide susceptibility is a computational mechanism that can acquire, represent, and compute a mapping from one multivariate space of information to another, given a dataset representing that mapping (Garrett, 1994). For many years, artificial neural networks have been employed in a wide range of classification applications in Earth sciences (Van Leeuwen, 2012). Also, some studies have also applied artificial neural networks as a statistical analysis method for landslide susceptibility mapping (Ermini et al, 2005.…”

Section: Introductionmentioning

confidence: 99%

Integration of Gis and Advanced Remote Sensing Techniques for Landslide Hazard Assessment: A Case Study of Northwest Syria

Hammad

Leeuwen

Mucsi

2020

ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci.

View full text Add to dashboard Cite

Abstract. Landslide susceptibility and hazard mapping has been developed providing remarkable results through the integration of geographic information system (GIS) and remote sensing. In this regard, some approaches have considered the use of Sentinel-1 data and time-series interferometric synthetic aperture radar (InSAR) techniques, such as differential InSAR (D-InSAR) and persistent scatterers interferometric (PSI), for providing precise information about total amount and velocity of ground-surface deformations and landslides within a specific area during a specific time period which is important for disaster management’s planning process.In this paper, artificial neural network (ANN) was used as a statistical analysis method for landslide susceptibility mapping in Northwest Syria using multi-layer perceptron (MLP) neural network on a training dataset of one dependent variable (landslide or non-landslide) and nine independent variables (slope, aspect, curvature, land cover, NDVI, lithology, distance from faults, distance from road, distance from stream networks). The resulting map of landslide susceptibility was validated using area under curve (AUC) analysis using a testing dataset which showed 90.28% of AUC value. Then, landslide susceptibility map was reclassified into high-moderate-low classes and integrated with intensity map of mean velocity of ground-surface deformations during the time period form (16 October 2018) until (21 March 2019) by using a landslide hazard matrix in a GIS environment in order to get landslide hazard map of the study area for that time period. The result shows that around 44.4%, 52.9% and 2.5% of total study area was classified as a high, moderate and low hazard zone of landslide, respectively.

show abstract

Artificial neural networks and geographic information systems for inland excess water classification

Cited by 6 publications

References 39 publications

Climate Change Impacts on Environmental Hazards on the Great Hungarian Plain, Carpathian Basin

Climate Change Impacts on Environmental Hazards on the Great Hungarian Plain, Carpathian Basin

The Hyper-Angular Cube Concept for Improving the Spatial and Acoustic Resolution of MBES Backscatter Angular Response Analysis

Integration of Gis and Advanced Remote Sensing Techniques for Landslide Hazard Assessment: A Case Study of Northwest Syria

Contact Info

Product

Resources

About