Neural networks-based simulation of land cover scenarios in Doon valley, India

Maithani, Sandeep

doi:10.1080/10106049.2014.927535

Cited by 15 publications

(9 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Applied to LC datasets, the aim is to use these automated, statistical methods to identify and analyze spatial patterns that have resulted from underlying processes of LCC over time. ML methods employed in LC simulations and assessments have previously included neural networks (NNs) [13], decision trees (DTs), and support vector machines (SVMs) [14].…”

Section: Introductionmentioning

confidence: 99%

Analyzing the Effects of Temporal Resolution and Classification Confidence for Modeling Land Cover Change with Long Short-Term Memory Networks

Duynhoven

Dragićević

2019

Remote Sensing

View full text Add to dashboard Cite

Land cover change (LCC) is typically characterized by infrequent changes over space and time. Data-driven methods such as deep learning (DL) approaches have proven effective in many domains for predictive and classification tasks. When applied to geospatial data, sequential DL methods such as long short-term memory (LSTM) have yielded promising results in remote sensing and GIScience studies. However, the characteristics of geospatial datasets selected for use with these methods have demonstrated important implications on method performance. The number of data layers available, the rate of LCC, and inherent errors resulting from classification procedures are expected to influence model performance. Yet, it is unknown how these can affect compatibility with the LSTM method. As such, the main objective of this study is to explore the capacity of LSTM to forecast patterns that have emerged from LCC dynamics given varying temporal resolutions, persistent land cover classes, and auxiliary data layers pertaining to classification confidence. Stacked LSTM modeling approaches are applied to 17-year MODIS land cover datasets focused on the province of British Columbia, Canada. This geospatial data is reclassified to four major land cover (LC) classes during pre-processing procedures. The evaluation considers the dataset at variable temporal resolutions to demonstrate the significance of geospatial data characteristics on LSTM method performance in several scenarios. Results indicate that LSTM can be utilized for forecasting LCC patterns when there are few limitations on temporal intervals of the datasets provided. Likewise, this study demonstrates improved performance measures when there are classes that do not change. Furthermore, providing classification confidence data as ancillary input also demonstrated improved results when the number of timesteps or temporal resolution is limited. This study contributes to future applications of DL and LSTM methods for forecasting LCC.

show abstract

Section: Introductionmentioning

confidence: 99%

Analyzing the Effects of Temporal Resolution and Classification Confidence for Modeling Land Cover Change with Long Short-Term Memory Networks

Duynhoven

Dragićević

2019

Remote Sensing

View full text Add to dashboard Cite

show abstract

“…Several studies have revealed that the integrated MultiLayer Perceptron-Markov chain analysis (MLP-MCA) method is a robust tool to quantify and model the spatio-temporal LULC changes that combines remote sensing and GIS efficiently (Ahmed & Ahmed, 2012;Mishra et al, 2014;Ozturk, 2015). In the MLP-MCA hybrid model, the MLP neural network is trained by supervized backpropagation (BP) algorithm with efficient generalization capability for each LULC transition and simulation (Maithani, 2015), while the MCA model determines transition probability areas to predict likely LULC changes in the future (Dadhich & Hanaoka, 2011).…”

Section: Introductionmentioning

confidence: 99%

Spatio-temporal urban growth dynamics of Lagos Metropolitan Region of Nigeria based on Hybrid methods for LULC modeling and prediction

Wang

Maduako

2018

European Journal of Remote Sensing

106

View full text Add to dashboard Cite

An accurate information on the amount and location of Land use and land cover (LULC) changes is necessary to develop and implement a sustainable-urban planning.This research investigates the potential of an integrated Multi-Layer Perceptron and Markov Chain Analysis (MLP-MCA) to map and accurately predict the future LULC change scenarios in Lagos Metropolitan Region of Nigeria. Multi-temporal LULC datasets derived from remotely sensed Landsat images from 1984, 2000 and 2015 were used for modeling, validation and prediction. Predicted LULC changes for 2030 and 2050 were performed based on the LULC map of 2015 using MLP-MCA method. The result reveals a significant expansion of built-up areas during the whole study period. Analysis of LULC distribution in Lagos metropolitan region shows that about 50% of urban land expansion happened beyond the administrative boundary of Lagos State during the period of 2000-2015. It is predicted that more than 75% of future urban growth will occur across the border of Lagos State, in the neighbouring Ogun State by 2050. These results imply that a strong and consistent collaboration between different states is crucial to establish an effective regional planning framework and ensure a proper planned growth of the metropolitan region.ARTICLE HISTORY

show abstract

“…Each model has advantages and disadvantages [27,30]. Thus, various hybrid models have also been adapted to simulate urban growth [19,22,25,[31][32][33][34][35][36]. Determining which hybrid model provides the best results is difficult because each study reaches unique conclusions.…”

mentioning

confidence: 99%

Urban Growth Simulation of Atakum (Samsun, Turkey) Using Cellular Automata-Markov Chain and Multi-Layer Perceptron-Markov Chain Models

Öztürk

2015

Remote Sensing

155

View full text Add to dashboard Cite

Abstract:In this study, urban growth of the Atakum District in Samsun, Turkey, was simulated by Cellular Automata-Markov Chain (CA-MC) and Multi-layer Perceptron-Markov Chain (MLP-MC) hybrid models in a geographical information system (GIS) environment. Historical land use/land cover (LU/LC) data were extracted from 1989, 2000 and 2013 Landsat TM/ETM+/OLI images. Using the LU/LC data for the years 1989 and 2000, the urban growth for 2013 was simulated using the CA-MC and MLP-MC models. The simulation results were compared with the 2013 LU/LC data to assess the validity of the simulation. The MLP-MC method provided the best results according to the validation based on the kappa index of agreement. Based on this result, the urban growth for the year 2025 was simulated using MLP-MC. The simulation estimated an urban growth rate of 35.2% between 2013 and 2025, an increase in the area of artificial surfaces from 1681.9 ha to 2274.3 ha and the destruction of 511.7 ha of agricultural land and 4.4 ha of forest. The results of this study demonstrate that the urban growth models provide a better understanding of the current patterns and temporal dynamics and can predict future changes according to past and current dynamics. The results also show that simulations are most accurate when using a model that best conforms to the changes in the given study area.

show abstract

Neural networks-based simulation of land cover scenarios in Doon valley, India

Cited by 15 publications

References 42 publications

Analyzing the Effects of Temporal Resolution and Classification Confidence for Modeling Land Cover Change with Long Short-Term Memory Networks

Analyzing the Effects of Temporal Resolution and Classification Confidence for Modeling Land Cover Change with Long Short-Term Memory Networks

Spatio-temporal urban growth dynamics of Lagos Metropolitan Region of Nigeria based on Hybrid methods for LULC modeling and prediction

Urban Growth Simulation of Atakum (Samsun, Turkey) Using Cellular Automata-Markov Chain and Multi-Layer Perceptron-Markov Chain Models

Contact Info

Product

Resources

About