Selection of Optimal Object Features in Object-Based Image Analysis Using Filter-Based Algorithms

With the increase in spatial resolution of recent sensors, object-based image analysis (OBIA) has gained importance for producing detailed land use maps. One of the main advantages of OBIA is that a variety of spectral, spatial and textural features can be extracted for the segmented image objects that are later utilized in classification. However, using a large number of features not only increases the required computational time, but also requires a large number of ground samples, which is unavailable in most cases. For these reasons, feature selection (FS) has become an important research topic for OBIA based classification studies. In this study, three filter-based FS algorithms namely, Chi square, information gain and ReliefF were applied to determine the most effective object features that ensure high separability among landscape features. For this purpose, importance degree (i.e. ranks) of 110 input object features were firstly estimated by the algorithms, and correlation-based merit function was then applied to determine optimum feature subset size. Multi-resolution segmentation algorithm was applied for segmenting a WorldView-2 image. Support vector machine, random forest and nearest neighbour classifiers were all utilized to classify segmented image objects using the selected object features. Results revealed that the FS algorithms were effective for selecting the most relevant features. Also, the classifiers produced the highest performances with 24 out of 110 features selected by the information gain (IG) algorithm. Particularly, the support vector machine classifier produced the highest overall accuracy (92.00%) with 24 selected features determined by the IG algorithm. A significant improvement of about 4% was achieved by applying FS procedures that was found statistically significant in terms of Wilcoxon signed-ranks test.     

Investigating the flood damages in Lower Indus Basin since 2000: Spatiotemporal analyses of the major flood events

While historically significant for ancient civilizations, the Indus basin is also known for its floods and complex anthropogenic management history. Resulting from years of modifications by the pre-British era Mughal rulers followed by the post-partition division of river waters among the two neighbors, India and Pakistan, Pakistan faces severe management and financial challenges of water management. This study investigates the intricacies arising from this complicated management doctrine for the lower Indus basin. A detailed remote sensing-based analysis of the significant floods to hit the lower Indus basin since 2000 has been provided. Flood years were identified, and Moderate Resolution Imaging Spectroradiometer (MODIS) data for the years 2003, 2005, 2006, 2010, 2011, 2012, 2015, and 2016 were used to map their spatiotemporal extents. Almost all the flood water accumulated in the north is released in one river channel of the lower Indus basin. Further, the challenges were exacerbated due to the excessive rainfall in 2011 and 2012 in southeastern Sindh. A trend analysis of rainfall data shows an increase in the southern basin in the last 21 years, particularly toward the central plains and Sindh Province. The floodwater accumulated in the lower basin for as many as?~?425 days on average, stretching to?~?800 days of stagnancy in some places. The water stagnation period has been the highest in the river floodplain, highly populated and cultivated. The analyses of the current study suggest that the riverine channel has been better managed after the 2010 floods; however, the monsoon’s shift in 2011 and 2012 led to widespread disaster in low-lying regions of Sindh Province.

     

Empirical Estimation of Uniaxial Compressive Strength of Rock: Database of Simple,Multiple, and Artificial Intelligence-Based Regressions

Geotechnical and Geological Engineering - Empirical relationships for estimating Uniaxial Compressive Strength (UCS) of rock from other rock properties are numerous in literature. This is because...     

Prediction of future situation of land use/cover change and modeling sensitivity to pollution in Edku Lake,Egypt based on geospatial analyses

GeoJournal - Assessing land use/cover (LULC) change in coastal lakes is an essential process for sustainable development. Edku Lake is one of the most important lakes in the northern part of Nile...     

Land suitability analysis for maize production in Indonesia using satellite remote sensing and GIS-based multicriteria decision support system

GeoJournal - Maize is one of the potential crops can help in regional food production with self-sufficiency of foods in the drought prone areas of East Java in Indonesia. The purpose of this...     

The use of vertical electrical sounding resistivity method for the location of low salinity groundwater for irrigation in Chaj and Rachna Doabs

A geoelectrical resistivity survey using vertical electrical sounding (VES) was conducted at Chaj Doab (land between rivers Jhelum and Chenab, Pakistan) and Rachna Doab (land between rivers Chenab and Ravi, Pakistan), with the objective of investigating groundwater conditions. A total of 90 sites were selected with 43 sites in Chaj and 47 sites in Rachna Doabs. The resistivity meter (ABEM Terrameter SAS 4000, Sweden) was used to collect the VES data by employing a Schlumberger electrode configuration, with half current electrode spacings (AB/2) ranging from 2 to 180 m and the potential electrode (MN) from 1 to 40 m. The field data were interpreted using the Interpex IX1D computer software and the resistivity versus depth models for each location was estimated. The outputs of subsurface layers with resistivities and thickness presented in contour maps and 3-D views by using SURFER software were created. A total of 102 groundwater samples from nearby hydrowells at different depths were collected to develop a correlation between the aquifer resistivity of VES and the electrical conductivity (EC) of the groundwater and to confirm the resulted geophysical resistivity models. From the correlation developed, it was observed that the groundwater salinity in the aquifer may be considered low and so safe for irrigation if resistivity >45 Ω m, and marginally fit for irrigation having resistivity between 25 and 45 Ω m. The study area has resistivities from 3.9 to 2,222 Ω m at the top of the unsaturated layer, between 1.21 and 171 Ω m, in the shallow aquifers, and 0.14–152 Ω m in the deep aquifers of the study area. The results indicate that the quality of groundwater is better near the rivers and in the shallow layers compared to the deep layers.     

A Neuro-Fuzzy modeling for prediction of solar cycles 24 and 25

The paper presents a Neuro-Fuzzy model to predict the features of the forthcoming sunspot cycles 24 and 25. The sunspot time series were analyzed with the proposed model. It is optimized based on Backpropagation scheme and applied to the yearly smoothed sunspot numbers. The appropriate number of network inputs for the sunspots data series is obtained based on sequential forward search for the Neuro-Fuzzy model. According to the model prediction the maximum amplitudes of the cycles 24 and 25 will occur in the year 2013 and year 2022 with peaks of 101±8 and 90.7±8, respectively. The correlation and error analysis are discussed to ensure the performance of the proposed Neuro-Fuzzy approach as a predictor for sunspot time series. The correlation coefficient between Neuro-Fuzzy model forecasted sunspot number values with the actual ones is 0.96.     

A Hybrid Model for Runoff Prediction Using Variational Mode Decomposition and Artificial Neural Network

Water Resources - Hydrological runoff prediction in a reliable and precise manner contributes significantly to the optimal management of hydropower resources. Considering the importance of runoff...     

Estimating water balance components in irrigated agriculture using a combined approach of soil moisture and energy balance monitoring,and numerical modelling

Information on water balance components such as evapotranspiration and groundwater recharge are crucial for water management. Due to differences in physical conditions, but also due to limited budgets, there is not one universal best practice, but a wide range of different methods with specific advantages and disadvantages. In this study, we propose an approach to quantify actual evapotranspiration, groundwater recharge and water inflow, i.e. precipitation and irrigation, that considers the specific conditions of irrigated agriculture in warm, arid environments. This approach does not require direct measurements of precipitation or irrigation quantities and is therefore suitable for sites with an uncertain data basis. For this purpose, we combine soil moisture and energy balance monitoring, remote sensing data analysis and numerical modelling using Hydrus. Energy balance data and routine weather data serve to estimate ET₀. Surface reflectance data from satellite images (Sentinel-2) are used to derive leaf area indices, which help to partition ET₀ into energy limited evaporation and transpiration. Subsequently, first approximations of water inflow are derived based on observed soil moisture changes. These inflow estimates are used in a series of forward simulations that produce initial estimates of drainage and ET_act, which in turn help improve the estimate of water inflow. Finally, the improved inflow estimates are incorporated into the model and then a parameter optimization is performed using the observed soil moisture as the reference figure. Forward simulations with calibrated soil parameters result in final estimates for ET_act and groundwater recharge. The presented method is applied to an agricultural test site with a crop rotation of cotton and wheat in Punjab, Pakistan. The final model results, with an RMSE of 2.2% in volumetric water content, suggest a cumulative ET_act and groundwater recharge of 769 and 297 mm over a period of 281 days, respectively. The total estimated water inflow accounts for 946 mm, of which 77% originates from irrigation.