Special Issue "Application of Artificial Neural Networks in Geoinformatics"

A special issue of Applied Sciences (ISSN 2076-3417).

Deadline for manuscript submissions: closed (31 August 2017)

Special Issue Editor

Guest Editor
Prof. Dr. Saro Lee

Geological Research Division, Korea Institute of Geoscience and Mineral Resources (KIGAM), 124 Gwahang-no, Yuseong-gu, Daejeon 305-350, Korea; Korea University of Science and Technology, 217 Gajeong-ro, Yuseong-gu, Daejeon 305-350, Korea
E-Mail
Interests: GIS; data mining; spatial analysis; natural hazard; geoinformatics

Special Issue Information

Dear Colleagues,

Over the last few decades, artificial neural networks, such as in data mining and machine learning technology, are being successfully applied across a wide range of science and engineering areas. In addition, according to the development of computer and space technologies, geoinformatics, as science and technology dealing with spatial information, are growing rapidly. Thus, recently, artificial neural networks have been widely applied in geoinformatics and have produced valuable results in geoscience, environment, natural hazards, and natural resources areas.

This Special Issue of the journal Applied Sciences, “Application of Artificial Neural Networks in Geoinformatics”, aims to attract novel contributions covering a wide range of applications in artificial neural networks in geoinformatics.

Our topics of interest include, but are not limited to:

  • Application of Artificial Neural Networks combined with Geographic Information System (GIS)
  • Application of Artificial Neural Networks in Remote Sensing
  • Application of Artificial Neural Networks in Global Positioning System (GPS)
  • Spatial Analysis based on Artificial Neural Networks
  • Geocomputation using Artificial Neural Networks
  • Spatial Prediction using Artificial Neural Networks
  • Processing of Geoinformation using Artificial Neural Networks
  • Application of Artificial Neural Networks on Geosciences, Environments, Natural Hazard, Natural Resources and Plnning

Comparison and Validation of Artificial Neural Networks with other Machine Learning models

Prof. Dr. Saro Lee
Guest Editor

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All papers will be peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Applied Sciences is an international peer-reviewed open access monthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 1400 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Keywords

  • Data mining
  • Machine Learning
  • Artificial Neural Networks
  • Spatial Database
  • Geoinformatics
  • Geographic Information System (GIS)
  • Remote Sensing
  • Global Positioning System (GPS)
  • Spatial Analysis

Published Papers (15 papers)

View options order results:
result details:
Displaying articles 1-15
Export citation of selected articles as:

Editorial

Jump to: Research

Open AccessEditorial Editorial for Special Issue: “Application of Artificial Neural Networks in Geoinformatics”
Appl. Sci. 2018, 8(1), 55; doi:10.3390/app8010055
Received: 27 December 2017 / Revised: 27 December 2017 / Accepted: 28 December 2017 / Published: 2 January 2018
PDF Full-text (148 KB) | HTML Full-text | XML Full-text
Abstract
Recently, a need has arisen for prediction techniques that can address a variety of problems by combining methods from the rapidly developing field of machine learning with geoinformation technologies such as GIS, remote sensing, and GPS.[…] Full article
(This article belongs to the Special Issue Application of Artificial Neural Networks in Geoinformatics)

Research

Jump to: Editorial

Open AccessArticle Classification of Forest Vertical Structure in South Korea from Aerial Orthophoto and Lidar Data Using an Artificial Neural Network
Appl. Sci. 2017, 7(10), 1046; doi:10.3390/app7101046
Received: 31 August 2017 / Accepted: 9 October 2017 / Published: 12 October 2017
Cited by 2 | PDF Full-text (7988 KB) | HTML Full-text | XML Full-text
Abstract
Every vegetation colony has its own vertical structure. Forest vertical structure is considered as an important indicator of a forest’s diversity and vitality. The vertical structure of a forest has typically been investigated by field survey, which is the traditional method of forest
[...] Read more.
Every vegetation colony has its own vertical structure. Forest vertical structure is considered as an important indicator of a forest’s diversity and vitality. The vertical structure of a forest has typically been investigated by field survey, which is the traditional method of forest inventory. However, this method is very time- and cost-consuming due to poor accessibility. Remote sensing data such as satellite imagery, aerial photography, and lidar data can be a viable alternative to the traditional field-based forestry survey. In this study, we classified forest vertical structures from red-green-blue (RGB) aerial orthophotos and lidar data using an artificial neural network (ANN), which is a powerful machine learning technique. The test site was Gongju province in South Korea, which contains single-, double-, and triple-layered forest structures. The performance of the proposed method was evaluated by comparing the results with field survey data. The overall accuracy achieved was about 70%. It means that the proposed approach can classify the forest vertical structures from the aerial orthophotos and lidar data. Full article
(This article belongs to the Special Issue Application of Artificial Neural Networks in Geoinformatics)
Figures

Figure 1

Open AccessArticle Shallow Landslide Susceptibility Modeling Using the Data Mining Models Artificial Neural Network and Boosted Tree
Appl. Sci. 2017, 7(10), 1000; doi:10.3390/app7101000
Received: 18 July 2017 / Revised: 14 September 2017 / Accepted: 21 September 2017 / Published: 28 September 2017
Cited by 1 | PDF Full-text (11211 KB) | HTML Full-text | XML Full-text
Abstract
The main purpose of this paper is to present some potential applications of sophisticated data mining techniques, such as artificial neural network (ANN) and boosted tree (BT), for landslide susceptibility modeling in the Yongin area, Korea. Initially, landslide inventory was detected from visual
[...] Read more.
The main purpose of this paper is to present some potential applications of sophisticated data mining techniques, such as artificial neural network (ANN) and boosted tree (BT), for landslide susceptibility modeling in the Yongin area, Korea. Initially, landslide inventory was detected from visual interpretation using digital aerial photographic maps with a high resolution of 50 cm taken before and after the occurrence of landslides. The debris flows were randomly divided into two groups: training and validation sets with a 50:50 proportion. Additionally, 18 environmental factors related to landslide occurrence were derived from the topography, soil, and forest maps. Subsequently, the data mining techniques were applied to identify the influence of environmental factors on landslide occurrence of the training set and assess landslide susceptibility. Finally, the landslide susceptibility indexes from ANN and BT were compared with a validation set using a receiver operating characteristics curve. The slope gradient, topographic wetness index, and timber age appear to be important factors in landslide occurrence from both models. The validation result of ANN and BT showed 82.25% and 90.79%, which had reasonably good performance. The study shows the benefit of selecting optimal data mining techniques in landslide susceptibility modeling. This approach could be used as a guideline for choosing environmental factors on landslide occurrence and add influencing factors into landslide monitoring systems. Furthermore, this method can rank landslide susceptibility in urban areas, thus providing helpful information when selecting a landslide monitoring site and planning land-use. Full article
(This article belongs to the Special Issue Application of Artificial Neural Networks in Geoinformatics)
Figures

Open AccessArticle Application of Deep Networks to Oil Spill Detection Using Polarimetric Synthetic Aperture Radar Images
Appl. Sci. 2017, 7(10), 968; doi:10.3390/app7100968
Received: 29 July 2017 / Revised: 10 September 2017 / Accepted: 15 September 2017 / Published: 21 September 2017
Cited by 2 | PDF Full-text (3179 KB) | HTML Full-text | XML Full-text
Abstract
Polarimetric synthetic aperture radar (SAR) remote sensing provides an outstanding tool in oil spill detection and classification, for its advantages in distinguishing mineral oil and biogenic lookalikes. Various features can be extracted from polarimetric SAR data. The large number and correlated nature of
[...] Read more.
Polarimetric synthetic aperture radar (SAR) remote sensing provides an outstanding tool in oil spill detection and classification, for its advantages in distinguishing mineral oil and biogenic lookalikes. Various features can be extracted from polarimetric SAR data. The large number and correlated nature of polarimetric SAR features make the selection and optimization of these features impact on the performance of oil spill classification algorithms. In this paper, deep learning algorithms such as the stacked autoencoder (SAE) and deep belief network (DBN) are applied to optimize the polarimetric feature sets and reduce the feature dimension through layer-wise unsupervised pre-training. An experiment was conducted on RADARSAT-2 quad-polarimetric SAR image acquired during the Norwegian oil-on-water exercise of 2011, in which verified mineral, emulsions, and biogenic slicks were analyzed. The results show that oil spill classification achieved by deep networks outperformed both support vector machine (SVM) and traditional artificial neural networks (ANN) with similar parameter settings, especially when the number of training data samples is limited. Full article
(This article belongs to the Special Issue Application of Artificial Neural Networks in Geoinformatics)
Figures

Figure 1

Open AccessArticle Application of Artificial Neural Networks to Ship Detection from X-Band Kompsat-5 Imagery
Appl. Sci. 2017, 7(9), 961; doi:10.3390/app7090961
Received: 31 July 2017 / Revised: 10 September 2017 / Accepted: 18 September 2017 / Published: 20 September 2017
Cited by 1 | PDF Full-text (5024 KB) | HTML Full-text | XML Full-text
Abstract
For ship detection, X-band synthetic aperture radar (SAR) imagery provides very useful data, in that ship targets look much brighter than surrounding sea clutter due to the corner-reflection effect. However, there are many phenomena which bring out false detection in the SAR image,
[...] Read more.
For ship detection, X-band synthetic aperture radar (SAR) imagery provides very useful data, in that ship targets look much brighter than surrounding sea clutter due to the corner-reflection effect. However, there are many phenomena which bring out false detection in the SAR image, such as noise of background, ghost phenomena, side-lobe effects and so on. Therefore, when ship-detection algorithms are carried out, we should consider these effects and mitigate them to acquire a better result. In this paper, we propose an efficient method to detect ship targets from X-band Kompsat-5 SAR imagery using the artificial neural network (ANN). The method produces the ship-probability map using ANN, and then detects ships from the ship-probability map by using a threshold value. For the purpose of getting an improved ship detection, we strived to produce optimal input layers used for ANN. In order to reduce phenomena related to the false detections, the non-local (NL)-means filter and median filter were utilized. The NL-means filter effectively reduced noise on SAR imagery without smoothing edges of the objects, and the median filter was used to remove ship targets in SAR imagery. Through the filtering approaches, we generated two input layers from a Kompsat-5 SAR image, and created a ship-probability map via ANN from the two input layers. When the threshold value of 0.67 was imposed on the ship-probability map, the result of ship detection from the ship-probability map was a 93.9% recall, 98.7% precision and 6.1% false alarm rate. Therefore, the proposed method was successfully applied to the ship detection from the Kompsat-5 SAR image. Full article
(This article belongs to the Special Issue Application of Artificial Neural Networks in Geoinformatics)
Figures

Figure 1

Open AccessArticle Characterization of Surface Ozone Behavior at Different Regimes
Appl. Sci. 2017, 7(9), 944; doi:10.3390/app7090944
Received: 25 July 2017 / Revised: 31 August 2017 / Accepted: 12 September 2017 / Published: 14 September 2017
Cited by 1 | PDF Full-text (1925 KB) | HTML Full-text | XML Full-text | Supplementary Files
Abstract
Previous studies showed that the influence of meteorological variables and concentrations of other air pollutants on O3 concentrations changes at different O3 concentration levels. In this study, threshold models with artificial neural networks (ANNs) were applied to characterize the O3
[...] Read more.
Previous studies showed that the influence of meteorological variables and concentrations of other air pollutants on O3 concentrations changes at different O3 concentration levels. In this study, threshold models with artificial neural networks (ANNs) were applied to characterize the O3 behavior at an urban site (Porto, Portugal), describing the effect of environmental and meteorological variables on O3 concentrations. ANN characteristics, and the threshold variable and value, were defined by genetic algorithms (GAs). The considered predictors were hourly average concentrations of NO, NO2, and O3, and meteorological variables (temperature, relative humidity, and wind speed) measured from January 2012 to December 2013. Seven simulations were performed and the achieved models considered wind speed (at 4.9 m·s−1), temperature (at 17.5 °C) and NO2 (at 26.6 μg·m−3) as the variables that determine the change of O3 behavior. All the achieved models presented a similar fitting performance: R2 = 0.71–0.72, RMSE = 14.5–14.7 μg·m−3, and the index of agreement of the second order of 0.91. The combined effect of these variables on O3 concentration was also analyzed. This statistical model was shown to be a powerful tool for interpreting O3 behavior, which is useful for defining policy strategies for human health protection concerning this air pollutant. Full article
(This article belongs to the Special Issue Application of Artificial Neural Networks in Geoinformatics)
Figures

Figure 1

Open AccessArticle Analysis of the Pyroclastic Flow Deposits of Mount Sinabung and Merapi Using Landsat Imagery and the Artificial Neural Networks Approach
Appl. Sci. 2017, 7(9), 935; doi:10.3390/app7090935
Received: 25 July 2017 / Revised: 7 September 2017 / Accepted: 8 September 2017 / Published: 11 September 2017
Cited by 1 | PDF Full-text (12161 KB) | HTML Full-text | XML Full-text
Abstract
Volcanic eruptions cause pyroclastic flows, which can destroy plantations and settlements. We used image data from Landsat 7 Bands 7, 4 and 2 and Landsat 8 Bands 7, 5 and 3 to observe and analyze the distribution of pyroclastic flow deposits for two
[...] Read more.
Volcanic eruptions cause pyroclastic flows, which can destroy plantations and settlements. We used image data from Landsat 7 Bands 7, 4 and 2 and Landsat 8 Bands 7, 5 and 3 to observe and analyze the distribution of pyroclastic flow deposits for two volcanos, Mount Sinabung and Merapi, over a period of 10 years (2001–2017). The satellite data are used in conjunction with an artificial neural network method to produce maps of pyroclastic precipitation for Landsat 7 and 8, then we calculated the pyroclastic precipitation area using an artificial neural network method after dividing the images into four classes based on color. Red, green, blue and yellow were used to indicate pyroclastic deposits, vegetation and forest, water and cloud, and farmland, respectively. The area affected by a volcanic eruption was deduced from the neural network processing, including calculating the area of pyroclastic deposits. The main differences between the pyroclastic flow deposits of Mount Sinabung and Mount Merapi are: the sediment deposits of the pyroclastic flows of Mount Sinabung tend to widen, whereas those of Merapi elongated; the direction of pyroclastic flow differed; and the area affected by an eruption was greater for Mount Merapi than Mount Sinabung because the VEI (Volcanic Explosivity Index) during the last 10 years of Mount Merapi was larger than Mount Sinabung. Full article
(This article belongs to the Special Issue Application of Artificial Neural Networks in Geoinformatics)
Figures

Figure 1

Open AccessArticle Real-Time Transportation Mode Identification Using Artificial Neural Networks Enhanced with Mode Availability Layers: A Case Study in Dubai
Appl. Sci. 2017, 7(9), 923; doi:10.3390/app7090923
Received: 30 July 2017 / Revised: 31 August 2017 / Accepted: 6 September 2017 / Published: 8 September 2017
Cited by 1 | PDF Full-text (6307 KB) | HTML Full-text | XML Full-text
Abstract
Traditionally, departments of transportation (DOTs) have dispatched probe vehicles with dedicated vehicles and drivers for monitoring traffic conditions. Emerging assisted GPS (AGPS) and accelerometer-equipped smartphones offer new sources of raw data that arise from voluntarily-traveling smartphone users provided that their modes of transportation
[...] Read more.
Traditionally, departments of transportation (DOTs) have dispatched probe vehicles with dedicated vehicles and drivers for monitoring traffic conditions. Emerging assisted GPS (AGPS) and accelerometer-equipped smartphones offer new sources of raw data that arise from voluntarily-traveling smartphone users provided that their modes of transportation can correctly be identified. By introducing additional raster map layers that indicate the availability of each mode, it is possible to enhance the accuracy of mode detection results. Even in its simplest form, an artificial neural network (ANN) excels at pattern recognition with a relatively short processing timeframe once it is properly trained, which is suitable for real-time mode identification purposes. Dubai is one of the major cities in the Middle East and offers unique environments, such as a high density of extremely high-rise buildings that may introduce multi-path errors with GPS signals. This paper develops real-time mode identification ANNs enhanced with proposed mode availability geographic information system (GIS) layers, firstly for a universal mode detection and, secondly for an auto mode detection for the particular intelligent transportation system (ITS) application of traffic monitoring, and compares the results with existing approaches. It is found that ANN-based real-time mode identification, enhanced by mode availability GIS layers, significantly outperforms the existing methods. Full article
(This article belongs to the Special Issue Application of Artificial Neural Networks in Geoinformatics)
Figures

Figure 1

Open AccessFeature PaperArticle Habitat Potential Mapping of Marten (Martes flavigula) and Leopard Cat (Prionailurus bengalensis) in South Korea Using Artificial Neural Network Machine Learning
Appl. Sci. 2017, 7(9), 912; doi:10.3390/app7090912
Received: 15 July 2017 / Revised: 18 August 2017 / Accepted: 31 August 2017 / Published: 5 September 2017
Cited by 1 | PDF Full-text (4656 KB) | HTML Full-text | XML Full-text
Abstract
This study developed habitat potential maps for the marten (Martes flavigula) and leopard cat (Prionailurus bengalensis) in South Korea. Both species are registered on the Red List of the International Union for Conservation of Nature, which means that they
[...] Read more.
This study developed habitat potential maps for the marten (Martes flavigula) and leopard cat (Prionailurus bengalensis) in South Korea. Both species are registered on the Red List of the International Union for Conservation of Nature, which means that they need to be managed properly. Various factors influencing the habitat distributions of the marten and leopard were identified to create habitat potential maps, including elevation, slope, timber type and age, land cover, and distances from a forest stand, road, or drainage. A spatial database for each species was constructed by preprocessing Geographic Information System (GIS) data, and the spatial relationship between the distribution of leopard cats and environmental factors was analyzed using an artificial neural network (ANN) model. This process used half of the existing habitat location data for the marten and leopard cat for training. Habitat potential maps were then created considering the relationships. Using the remaining half of the habitat location data for each species, the model was validated. The results of the model were relatively successful, predicting approximately 85% for the marten and approximately 87% for the leopard cat. Therefore, the habitat potential maps can be used for monitoring the habitats of both species and managing these habitats effectively. Full article
(This article belongs to the Special Issue Application of Artificial Neural Networks in Geoinformatics)
Figures

Figure 1

Open AccessArticle Impacts of Sample Design for Validation Data on the Accuracy of Feedforward Neural Network Classification
Appl. Sci. 2017, 7(9), 888; doi:10.3390/app7090888
Received: 20 July 2017 / Revised: 11 August 2017 / Accepted: 21 August 2017 / Published: 30 August 2017
Cited by 1 | PDF Full-text (684 KB) | HTML Full-text | XML Full-text
Abstract
Validation data are often used to evaluate the performance of a trained neural network and used in the selection of a network deemed optimal for the task at-hand. Optimality is commonly assessed with a measure, such as overall classification accuracy. The latter is
[...] Read more.
Validation data are often used to evaluate the performance of a trained neural network and used in the selection of a network deemed optimal for the task at-hand. Optimality is commonly assessed with a measure, such as overall classification accuracy. The latter is often calculated directly from a confusion matrix showing the counts of cases in the validation set with particular labelling properties. The sample design used to form the validation set can, however, influence the estimated magnitude of the accuracy. Commonly, the validation set is formed with a stratified sample to give balanced classes, but also via random sampling, which reflects class abundance. It is suggested that if the ultimate aim is to accurately classify a dataset in which the classes do vary in abundance, a validation set formed via random, rather than stratified, sampling is preferred. This is illustrated with the classification of simulated and remotely-sensed datasets. With both datasets, statistically significant differences in the accuracy with which the data could be classified arose from the use of validation sets formed via random and stratified sampling (z = 2.7 and 1.9 for the simulated and real datasets respectively, for both p < 0.05%). The accuracy of the classifications that used a stratified sample in validation were smaller, a result of cases of an abundant class being commissioned into a rarer class. Simple means to address the issue are suggested. Full article
(This article belongs to the Special Issue Application of Artificial Neural Networks in Geoinformatics)
Figures

Figure 1

Open AccessArticle Road Safety Risk Evaluation Using GIS-Based Data Envelopment Analysis—Artificial Neural Networks Approach
Appl. Sci. 2017, 7(9), 886; doi:10.3390/app7090886
Received: 31 July 2017 / Revised: 20 August 2017 / Accepted: 22 August 2017 / Published: 29 August 2017
Cited by 1 | PDF Full-text (23512 KB) | HTML Full-text | XML Full-text
Abstract
Identification of the most significant factors for evaluating road risk level is an important question in road safety research, predominantly for decision-making processes. However, model selection for this specific purpose is the most relevant focus in current research. In this paper, we proposed
[...] Read more.
Identification of the most significant factors for evaluating road risk level is an important question in road safety research, predominantly for decision-making processes. However, model selection for this specific purpose is the most relevant focus in current research. In this paper, we proposed a new methodological approach for road safety risk evaluation, which is a two-stage framework consisting of data envelopment analysis (DEA) in combination with artificial neural networks (ANNs). In the first phase, the risk level of the road segments under study was calculated by applying DEA, and high-risk segments were identified. Then, the ANNs technique was adopted in the second phase, which appears to be a valuable analytical tool for risk prediction. The practical application of DEA-ANN approach within the Geographical Information System (GIS) environment will be an efficient approach for road safety risk analysis. Full article
(This article belongs to the Special Issue Application of Artificial Neural Networks in Geoinformatics)
Figures

Figure 1

Open AccessArticle A New Damage Assessment Method by Means of Neural Network and Multi-Sensor Satellite Data
Appl. Sci. 2017, 7(8), 781; doi:10.3390/app7080781
Received: 29 June 2017 / Revised: 24 July 2017 / Accepted: 26 July 2017 / Published: 1 August 2017
Cited by 1 | PDF Full-text (4250 KB) | HTML Full-text | XML Full-text
Abstract
Artificial Neural Network (ANN) is a valuable and well-established inversion technique for the estimation of geophysical parameters from satellite images. After training, ANNs are able to generate very fast products for several types of applications. Satellite remote sensing is an efficient way to
[...] Read more.
Artificial Neural Network (ANN) is a valuable and well-established inversion technique for the estimation of geophysical parameters from satellite images. After training, ANNs are able to generate very fast products for several types of applications. Satellite remote sensing is an efficient way to detect and map strong earthquake damage for contributing to post-disaster activities during emergency phases. This work aims at presenting an application of the ANN inversion technique addressed to the evaluation of building collapse ratio (CR), defined as the number of collapsed buildings with respect to the total number of buildings in a city block, by employing optical and SAR satellite data. This is done in order to directly relate changes in images with damage that has occurred during strong earthquakes. Furthermore, once they have been trained, neural networks can be used rapidly at application stage. The goal was to obtain a general tool suitable for re-use in different scenarios. An ANN has been implemented in order to emulate a regression model and to estimate the CR as a continuous function. The adopted ANN has been trained using some features obtained from optical and Synthetic Aperture Radar (SAR) images, as inputs, and the corresponding values of collapse ratio obtained from the survey of the 2010 M7 Haiti Earthquake, i.e., as target output. As regards the optical data, we selected three change parameters: the Normalized Difference Index (NDI), the Kullback–Leibler divergence (KLD), and Mutual Information (MI). Concerning the SAR images, the Intensity Correlation Difference (ICD) and the KLD parameters have been considered. Exploiting an object-oriented approach, a segmentation of the study area into several regions has been performed. In particular, damage maps have been generated by considering a set of polygons (in which satellite parameters have been calculated) extracted from the open source Open Street Map (OSM) geo-database. The trained ANN has been proposed for the M6.0 Amatrice earthquake that occurred on 24 August 2016, in central Italy, by using the features extracted from Sentinel-2 and COSMO-SkyMed images as input. The results show that the ANN is able to retrieve a building collapse ratio with good accuracy. In particular, the fusion approach modelled the collapse ratio characterized by high values of CR (more than 0.5) over the historical center that agrees with observed damages. Since the technique is independent from different typologies of input data (i.e., for radiometric or spatial resolution characteristics), the study demonstrated the strength of the proposed approach for estimating damaged areas and its importance in near real time monitoring activities, owing to its fast application. Full article
(This article belongs to the Special Issue Application of Artificial Neural Networks in Geoinformatics)
Figures

Open AccessArticle Optimized Neural Architecture for Automatic Landslide Detection from High‐Resolution Airborne Laser Scanning Data
Appl. Sci. 2017, 7(7), 730; doi:10.3390/app7070730
Received: 29 June 2017 / Revised: 11 July 2017 / Accepted: 13 July 2017 / Published: 16 July 2017
Cited by 3 | PDF Full-text (10456 KB) | HTML Full-text | XML Full-text
Abstract
An accurate inventory map is a prerequisite for the analysis of landslide susceptibility, hazard, and risk. Field survey, optical remote sensing, and synthetic aperture radar techniques are traditional techniques for landslide detection in tropical regions. However, such techniques are time consuming and costly.
[...] Read more.
An accurate inventory map is a prerequisite for the analysis of landslide susceptibility, hazard, and risk. Field survey, optical remote sensing, and synthetic aperture radar techniques are traditional techniques for landslide detection in tropical regions. However, such techniques are time consuming and costly. In addition, the dense vegetation of tropical forests complicates the generation of an accurate landslide inventory map for these regions. Given its ability to penetrate vegetation cover, high-resolution airborne light detection and ranging (LiDAR) has been used to generate accurate landslide maps. This study proposes the use of recurrent neural networks (RNN) and multi-layer perceptron neural networks (MLP-NN) in landscape detection. These efficient neural architectures require little or no prior knowledge compared with traditional classification methods. The proposed methods were tested in the Cameron Highlands, Malaysia. Segmentation parameters and feature selection were respectively optimized using a supervised approach and correlation-based feature selection. The hyper-parameters of network architecture were defined based on a systematic grid search. The accuracies of the RNN and MLP-NN models in the analysis area were 83.33% and 78.38%, respectively. The accuracies of the RNN and MLP-NN models in the test area were 81.11%, and 74.56%, respectively. These results indicated that the proposed models with optimized hyper-parameters produced the most accurate classification results. LiDAR-derived data, orthophotos, and textural features significantly affected the classification results. Therefore, the results indicated that the proposed methods have the potential to produce accurate and appropriate landslide inventory in tropical regions such as Malaysia. Full article
(This article belongs to the Special Issue Application of Artificial Neural Networks in Geoinformatics)
Figures

Open AccessArticle Data Mining Approaches for Landslide Susceptibility Mapping in Umyeonsan, Seoul, South Korea
Appl. Sci. 2017, 7(7), 683; doi:10.3390/app7070683
Received: 8 June 2017 / Revised: 26 June 2017 / Accepted: 27 June 2017 / Published: 2 July 2017
Cited by 2 | PDF Full-text (8502 KB) | HTML Full-text | XML Full-text
Abstract
The application of data mining models has become increasingly popular in recent years in assessments of a variety of natural hazards such as landslides and floods. Data mining techniques are useful for understanding the relationships between events and their influencing variables. Because landslides
[...] Read more.
The application of data mining models has become increasingly popular in recent years in assessments of a variety of natural hazards such as landslides and floods. Data mining techniques are useful for understanding the relationships between events and their influencing variables. Because landslides are influenced by a combination of factors including geomorphological and meteorological factors, data mining techniques are helpful in elucidating the mechanisms by which these complex factors affect landslide events. In this study, spatial data mining approaches based on data on landslide locations in the geographic information system environment were investigated. The topographical factors of slope, aspect, curvature, topographic wetness index, stream power index, slope length factor, standardized height, valley depth, and downslope distance gradient were determined using topographical maps. Additional soil and forest variables using information obtained from national soil and forest maps were also investigated. A total of 17 variables affecting the frequency of landslide occurrence were selected to construct a spatial database, and support vector machine (SVM) and artificial neural network (ANN) models were applied to predict landslide susceptibility from the selected factors. In the SVM model, linear, polynomial, radial base function, and sigmoid kernels were applied in sequence; the model yielded 72.41%, 72.83%, 77.17% and 72.79% accuracy, respectively. The ANN model yielded a validity accuracy of 78.41%. The results of this study are useful in guiding effective strategies for the prevention and management of landslides in urban areas. Full article
(This article belongs to the Special Issue Application of Artificial Neural Networks in Geoinformatics)
Figures

Open AccessArticle Severity Prediction of Traffic Accidents with Recurrent Neural Networks
Appl. Sci. 2017, 7(6), 476; doi:10.3390/app7060476
Received: 14 March 2017 / Revised: 27 April 2017 / Accepted: 28 April 2017 / Published: 8 June 2017
Cited by 3 | PDF Full-text (2652 KB) | HTML Full-text | XML Full-text
Abstract
In this paper, a deep learning model using a Recurrent Neural Network (RNN) was developed and employed to predict the injury severity of traffic accidents based on 1130 accident records that have occurred on the North-South Expressway (NSE), Malaysia over a six-year period
[...] Read more.
In this paper, a deep learning model using a Recurrent Neural Network (RNN) was developed and employed to predict the injury severity of traffic accidents based on 1130 accident records that have occurred on the North-South Expressway (NSE), Malaysia over a six-year period from 2009 to 2015. Compared to traditional Neural Networks (NNs), the RNN method is more effective for sequential data, and is expected to capture temporal correlations among the traffic accident records. Several network architectures and configurations were tested through a systematic grid search to determine an optimal network for predicting the injury severity of traffic accidents. The selected network architecture comprised of a Long-Short Term Memory (LSTM) layer, two fully-connected (dense) layers and a Softmax layer. Next, to avoid over-fitting, the dropout technique with a probability of 0.3 was applied. Further, the network was trained with a Stochastic Gradient Descent (SGD) algorithm (learning rate = 0.01) in the Tensorflow framework. A sensitivity analysis of the RNN model was further conducted to determine these factors’ impact on injury severity outcomes. Also, the proposed RNN model was compared with Multilayer Perceptron (MLP) and Bayesian Logistic Regression (BLR) models to understand its advantages and limitations. The results of the comparative analyses showed that the RNN model outperformed the MLP and BLR models. The validation accuracy of the RNN model was 71.77%, whereas the MLP and BLR models achieved 65.48% and 58.30% respectively. The findings of this study indicate that the RNN model, in deep learning frameworks, can be a promising tool for predicting the injury severity of traffic accidents. Full article
(This article belongs to the Special Issue Application of Artificial Neural Networks in Geoinformatics)
Figures

Figure 1

Back to Top