Showing papers in "Journal of The Indian Society of Remote Sensing in 2018"
TL;DR: In this article, the spatial prediction of rainfall-induced landslides at the Pauri Gahwal area, Uttarakhand, India has been done using Aggregating One-Dependence Estimators (AODE) classifier which has not been applied earlier for landslide problems.
Abstract: In this study, the spatial prediction of rainfall-induced landslides at the Pauri Gahwal area, Uttarakhand, India has been done using Aggregating One-Dependence Estimators (AODE) classifier which has not been applied earlier for landslide problems. Historical landslide locations have been collated with a set of influencing factors for landslide spatial analysis. The performance of the AODE model has been assessed using statistical analyzing methods and receiver operating characteristic curve technique. The predictive capability of the AODE model has also been compared with other popular landslide models namely Support Vector Machines (SVM), Radial Basis Function Neural Network (ANN-RBF), Logistic Regression (LR), and Naive Bayes (NB). The result of analysis illustrates that the AODE model has highest predictability, followed by the SVM model, the ANN-RBF model, the LR model, and the NB model, respectively. Thus AODE is a promising method for the development of better landslide susceptibility map for proper landslide hazard management.
62 citations
TL;DR: In this article, the CERES-Maize model was calibrated and evaluated with the field experiment data and after calibration and evaluation, this model was used to forecast maize yield.
Abstract: Real time, accurate and reliable estimation of maize yield is valuable to policy makers in decision making. The current study was planned for yield estimation of spring maize using remote sensing and crop modeling. In crop modeling, the CERES-Maize model was calibrated and evaluated with the field experiment data and after calibration and evaluation, this model was used to forecast maize yield. A Field survey of 64 farm was also conducted in Faisalabad to collect data on initial field conditions and crop management data. These data were used to forecast maize yield using crop model at farmers’ field. While in remote sensing, peak season Landsat 8 images were classified for landcover classification using machine learning algorithm. After classification, time series normalized difference vegetation index (NDVI) and land surface temperature (LST) of the surveyed 64 farms were calculated. Principle component analysis were run to correlate the indicators with maize yield. The selected LSTs and NDVIs were used to develop yield forecasting equations using least absolute shrinkage and selection operator (LASSO) regression. Calibrated and evaluated results of CERES-Maize showed the mean absolute % error (MAPE) of 0.35–6.71% for all recorded variables. In remote sensing all machine learning algorithms showed the accuracy greater the 90%, however support vector machine (SVM-radial basis) showed the higher accuracy of 97%, that was used for classification of maize area. The accuracy of area estimated through SVM-radial basis was 91%, when validated with crop reporting service. Yield forecasting results of crop model were precise with RMSE of 255 kg ha−1, while remote sensing showed the RMSE of 397 kg ha−1. Overall strength of relationship between estimated and actual grain yields were good with R2 of 0.94 in both techniques. For regional yield forecasting remote sensing could be used due greater advantages of less input dataset and if focus is to assess specific stress, and interaction of plant genetics to soil and environmental conditions than crop model is very useful tool.
43 citations
TL;DR: D drone-based imagery in combination with 3D scene reconstruction algorithms provide flexible and effective tools to map and monitor landslide apart from accurately assessing the dimensions of the landslides.
Abstract: Dimension estimation of landslides is a major challenge while preparing the landslide inventory map, for which very high-resolution satellite data/aerial photography is required, which is very expensive. An alternative is the application of drones for such mapping. This study presents the utility of drone/unmanned aerial vehicle (UAV) for mapping and 3D reconstruction of two landslides near IIT Mandi, Himachal Pradesh. In this study, we used DJI Phantom 3 Advanced drone to collect high-resolution images of landslides. Features in the images were automatically detected, described, and matched among photographs using scale invariant feature transform (SIFT) technique. The 3D position and orientation of the cameras and the XYZ location of each feature in the photographs was estimated using bundle block adjustment. This resulted in sparse 3D point cloud, which was densified using Clustering View for Multi-View Stereo (CMVS) algorithm. Finally, surface reconstruction was done using Poisson Surface Reconstruction method, which was visualised in open source software CloudCompare. The 3D model, generated from drone images, was validated using field measurements of some objects, and 3D surface created from total station. Various quantities i.e. width (length), height and perimeter were measured in the 3D drone model and verified with total station data. The differences among all the measured quantities for both the landslides are less than 5% keeping the measurements of total station as reference. The 3D reconstructed from the sets of photographs is very accurate giving the measurements up to cm level and even small objects could be easily identified. In addition, digital elevation model (DEM) of sub meter resolution could be easily generated and used for various applications. Hence drone-based imagery in combination with 3D scene reconstruction algorithms provide flexible and effective tools to map and monitor landslide apart from accurately assessing the dimensions of the landslides.
41 citations
TL;DR: A new automatic method is proposed for road extraction by integrating the SVM and Level Set methods, where the estimated probability of classification by SVM is used as input in Level Set Method.
Abstract: Currently, methods of extracting spatial information from satellite images are mainly based on visual interpretations and drawing the consequences by human factor, which is both costly and time consuming. A large volume of data collected by satellite sensors, and significant improvement in spatial and spectral resolution of these images require the development of new methods for optimal use of these data in order to produce rapid economic and updating road maps. In this study, a new automatic method is proposed for road extraction by integrating the SVM and Level Set methods. The estimated probability of classification by SVM is used as input in Level Set Method. The average of completeness, correctness, and quality was 84.19, 88.69 and 76.06% respectively indicate high performance of proposed method for road extraction from Google Earth images.
36 citations
TL;DR: In this paper, a remote sensing-based approach of predicting sugarcane yield, at district level, using Vegetation Condition Index (VCI), under the FASAL programme of the Ministry of Agriculture & Farmers' Welfare 13-years' historical database (2003-2015) of NDVI was used to derive the VCI NDVI products (MOD-13A2) of MODIS instrument on board Terra satellite at 16-day interval from first fortnight of June to second fortnight of October (peak growing period).
Abstract: Early yield assessment at local, regional and national scales is a major requirement for various users such as agriculture planners, policy makers, crop insurance companies and researchers This current study explored a remote sensing-based approach of predicting sugarcane yield, at district level, using Vegetation Condition Index (VCI), under the FASAL programme of the Ministry of Agriculture & Farmers’ Welfare 13-years’ historical database (2003–2015) of NDVI was used to derive the VCI NDVI products (MOD-13A2) of MODIS instrument on board Terra satellite at 16-day interval from first fortnight of June to second fortnight of October (peak growing period) were used to calculate the VCI Stepwise regression technique was used to develop empirical models between VCI and historical yield of sugarcane over 52 major sugarcane-growing districts in five states of India For all the districts, the empirical models were found to be statistically significant A large number of statistical parameters were computed to evaluate the performance of VCI-based models in predicting district-level sugarcane yield Though there was variation in model performance in different states, overall, the study showed the usefulness of VCI, which can be used as an input for operational sugarcane yield forecasting
32 citations
TL;DR: In this article, the authors investigated the airborne LiDAR and UAV-based digital elevation model (DEM) comparisons in terms of correlation and vertical accuracy, and performed statistical analyses to calculate the correlations and accuracies of DEMS.
Abstract: A digital elevation model (DEM) is a very important product that represents the topography digitally. It is an essential requirement of many engineering applications. From past to present, the methodology of DEM generation process is changed with respect to technology. Today, the laser scanner and aerial imagery are two widely used technologies to get DEM. Especially, the computer vision aided the use of unmanned aerial vehicles (UAV) opened new horizons in this regard. This study investigates the airborne LiDAR and UAV based DEM comparisons in terms of correlation and vertical accuracy. For this purpose four different LiDAR data are provided. Moreover, a photogrammetric flight is carried out with UAV and images of the study area are captured after field surveys. Then, five different DEMs are generated from five different point clouds. Finally, the statistical analyses are performed to calculate the correlations and accuracies of DEMS. According to the analysis, the UAV based models are as accurate as LiDAR based models along with some other advantages.
30 citations
TL;DR: In this paper, the authors used historical satellite images to determine the long-term shoreline changes along the coast of Chilika lagoon and analyzed the shoreline erosion and accretion based on statistical approach.
Abstract: Shoreline is the dynamic interfaces of both terrestrial and marine environment, which constantly affected by natural coastal processes includes wave, tide, littoral drift and cyclonic storms as well as coastal development. Wave induced littoral drift and fluvial discharge causing the gradual inlet migration and has the concurrent impact on shoreline of Chilika lagoon. This study is to determine the long-term shoreline changes along the coast of Chilika lagoon. Historical satellite images were used to analyse the shoreline erosion and accretion based on statistical approach. The satellite data from 1975 to 2015 were processed by using ERDAS Imagine and the shorelines are extracted. The shoreline oscillation was analysed at an interval of 100 m along the coast of Chilika lagoon using DSAS software. Most commonly used statistical methods such as end point rate and linear regression rate are used. The shoreline change analysis for entire coast of the lagoon since 40 years (1975–2015) indicates that 62% is of accretion, 25% is under stable coast and erosion is 13%. The result reveals that the lagoon coast shows high accretion of 9.12 m/year at updrift side of the lagoon inlet whereas the downdrift side shows high erosion of − 10.73 m/year due to the wave induced northeasterly longshore sediment transport round the year and riverine discharge. This study would provide the potential erosion and accretion area at Chilika lagoon coast and would help in adaptive shoreline management plan.
28 citations
TL;DR: The experimental results illustrate that the proposed fusion algorithm outperforms Curvelet transform and other traditional fusion algorithms in whole both in visual effect and objective evaluation indexes.
Abstract: In order to fuse two registered multi-spectral (MS) image and panchromatic (PAN) image in the same scene, a new remote sensing image fusion algorithm based on Principal Component Analysis (PCA) and Curvelet transform is proposed. The first principle component PC1 of MS image is extracted via PCA transform, at the same time, we perform the Morphology-Hat transform on the PAN image, and segment the transformed PAN image by the PCNN segmentation algorithm. Perform the Curvelet transform on the component PC1 of MS image and the PAN image after Morphology-Hat transform, and use different fusion rule to fuse different scale layers coefficients (coarse, detail and fine scale layer). For obtaining the fused image, we use the inverse Curvelet transform and inverse PCA transform to obtain the fused image. The experimental results illustrate that the proposed fusion algorithm outperforms Curvelet transform and other traditional fusion algorithms in whole such as intensity–hue–saturation, PCA, Brovey and Weighted Average both in visual effect and objective evaluation indexes (standard deviation, mean, information entropy, correlation coefficient, spectral distortions and deviation index).
28 citations
TL;DR: In this paper, the authors present a study on the burnt area assessment caused by the fire episode of April 2016 to the forest vegetation, which was found to be 3774.14 km2, representing 15.28% of the total forest area of the state.
Abstract: The hills of Uttarakhand witness forest fire every year during the summer season and the number of these fire events is reported to have increased due to increased anthropogenic disturbances as well as changes in climate. These fires cause significant damage to the natural resources which can be mapped and monitored using satellite images by virtue of its synoptic coverage of the landscape and near real time monitoring. This study presents burnt area assessment caused by the fire episode of April 2016 to the forest vegetation. Digital classification of satellite images was done to extract the burnt area which was found to be 3774.14 km2, representing 15.28% of the total forest area of the state. It also gives an account of cumulative progression of forest fire in Uttarakhand using satellite images of three dates viz. 23rd, 27th May and 2nd June, 2016. Results were analyzed at district, administrative and forest division level using overlay analysis. Separate area statistics were given for different categories of biological richness, forest types and protected areas affected by forest fire. The burnt area assessment can be used in mitigation planning to prevent drastic ecological impacts of the forest fire on the landscape.
27 citations
TL;DR: The proposed hybrid deep learning framework approach for accurate mapping of debris covered glaciers was observed to outperform the current state-of-art machine learning algorithms such as artificial neural network, support vector machine, and random forest.
Abstract: The main aim of this study is to propose a novel hybrid deep learning framework approach for accurate mapping of debris covered glaciers. The framework comprises of integration of several CNNs architecture, in which different combinations of Landsat 8 multispectral bands (including thermal band), topographic and texture parameters are passed as input for feature extraction. The output of an ensemble of these CNNs is hybrid with random forest model for classification. The major pillars of the framework include: (1) technique for implementing topographic and atmospheric corrections (preprocessing), (2) the proposed hybrid of ensemble of CNNs and random forest classifier, and (3) procedures to determine whether a pixel predicted as snow is a cloud edge/shadow (post-processing). The proposed approach was implemented on the multispectral Landsat 8 OLI (operational land imager)/TIRS (thermal infrared sensor) data and Shuttle Radar Topography Mission Digital Elevation Model for the part of the region situated in Alaknanda basin, Uttarakhand, Himalaya. The proposed framework was observed to outperform (accuracy 96.79%) the current state-of-art machine learning algorithms such as artificial neural network, support vector machine, and random forest. Accuracy assessment was performed by means of several statistics measures (precision, accuracy, recall, and specificity).
26 citations
TL;DR: This work demonstrates the use and potential of LARS in agriculture, particularly small holder open field agriculture, as well as comparison of performance parameters of KMeans spectral–spatial and ELM spectral-spatial classification methods.
Abstract: UAVs are fast emerging as a remote sensing platform to complement satellite based remote sensing. Agriculture and ecology is one of the important applications of UAV remote sensing, also known as low altitude remote sensing (LARS). This work demonstrates the use and potential of LARS in agriculture, particularly small holder open field agriculture. Two UAVs are used for remote sensing. The first UAV is a fixed wing aircraft with a high spatial resolution visible spectrum also known as RGB camera as a payload. The second UAV is a quadrotor UAV with an RGB camera interfaced to an on-board single board computer as the payload. LARS was carried out to acquire aerial high spatial resolution RGB images of different farms. Spectral–spatial classification of high spatial resolution RGB images for detection, delineation and counting of tree crowns in the image is presented. Supervised classification is carried out using extreme learning machine (ELM), a single hidden layer feed forward network neural network classifier. ELM was modelled for RGB values as input feature vectors and binary (tree and non-tree pixels) output class. Due to similarities in spectral intensities, some of the non-tree pixels were classified as tree pixels and in order to remove them, spatial classification was performed on the image. Spatial classification was carried out using thresholded geometrical property filtering techniques. Threshold values chosen for carrying out spatial classification were analysed to obtain optimal values. Finally in the delineation and counting, the connected tree crowns were segmented using Watershed algorithm performed on the image after marking individual tree crowns using Distance Transform method. Five representative UAV images captured at different altitudes with different crowns of banana plant, mango trees and coconut trees were used to demonstrate the performance of the proposed method. The performance was compared with the traditional KMeans spectral–spatial method of clustering. Results and comparison of performance parameters of KMeans spectral–spatial and ELM spectral–spatial classification methods are presented. Results indicate that ELM performed better than KMeans.
TL;DR: In this paper, eight spatial interpolation methods are used to interpolate precipitation and temperature over several integration periods in a local scale, and the performance is assessed using independent validation based on four measurements: the root mean squared error, the mean squared relative error, coefficient of determination (r2), and the coefficient of efficiency.
Abstract: Eight spatial interpolation methods are used to interpolate precipitation and temperature over several integration periods in a local scale. The methods used are inverse distance weighting (IDW), Thiessen polygons (TP), trend surface analysis, local polynomial interpolation, thin plate spline, and three Kriging methods: ordinary, universal, and simple (OK, UK, and SK). Daily observations from 17 stations in the Seyhan Basin, Turkey, between 1987 and 1994 are used. A variety of parameters and models are used in each method to interpolate surfaces for several integration periods, namely, daily, monthly and annual total precipitation; monthly and annual average precipitation; and daily, monthly and annual average temperature. The performance is assessed using independent validation based on four measurements: the root mean squared error, the mean squared relative error, the coefficient of determination (r2), and the coefficient of efficiency. Based on these validation measurements, the method with smallest errors for most of the integration periods concerning both precipitation and temperature is IDW with a power of 3, whereas TP has the highest errors. The Gaussian model is found superior than other models with less errors in the three Kriging methods for interpolating precipitation, but no specific model is better than another for modeling temperature. UK with elevation as the external drift and SK with the mean as an additional parameter show no superiority over OK. For precipitation, annual average and monthly totals are found to be the worst and best modeled integration periods respectively, with the monthly average the best for temperature.
TL;DR: In this article, the authors developed an optimum scale parameter selector (OSPS) tool for objective determination of multiple optimal scales in an image by multi-resolution segmentation using eCognition software.
Abstract: Image segmentation to create representative objects by region growing image segmentation techniques such as multi resolution segmentation (MRS) is mostly done through interactive selection of scale parameters and is still a subject of great research interest in object-based image analysis. In this study, we developed an optimum scale parameter selector (OSPS) tool for objective determination of multiple optimal scales in an image by MRS using eCognition software. The ready to use OSPS tool consists of three modules and determines optimum scales in an image by combining intrasegment variance and intersegment spatial autocorrelation. The tool was tested using WorldView-2 and Resourcesat-2 LISS-IV Mx images having different spectral and spatial resolutions in two areas to find optimal objects for ground features such as water bodies, trees, buildings, road, agricultural fields and landslides. Quality of the objects created for these features using scale parameters obtained from the OSPS tool was evaluated quantitatively using segmentation goodness metrics. Results show that OSPS tool is able determine optimum scale parameters for creation of representative objects from high resolution satellite images by MRS method.
TL;DR: A two-stage method to extract planar ground surfaces and global properties of road, that is, topology and smoothness and its radiometric response to laser beam of MLS are used in the second stage of the proposed method.
Abstract: The existing roadway infrastructures are mostly archived with two-dimensional (2D) drawings that lack the possibility for three-dimensional (3D) interpretation and advanced 3D analysis. The mobile LiDAR system (MLS) is gaining popularity in 3D mapping applications along various types of road corridors. MLS achieves the highest data quality and completeness among the traditional roadway data collection methods. The rural roads in different countries especially in India form a substantial portion of the road network. Therefore the proper maintenance and road safety analysis of rural roads are recommended activity, which could be addressed using detailed 3D road surface information. The absence of raised curb at road boundary, and presence of complexity, heterogeneity and occlusions along the rural roadway settings restrict the use of existing studies for road surface extraction using MLS point cloud data. Therefore considering the above requirement, this research paper proposes a two-stage method. The first stage extract planar ground surfaces which are further used to filter road surface in the second stage. Global properties of road, that is, topology and smoothness and its radiometric response to laser beam of MLS are used in the second stage. MLS point cloud data of rural roadway were used to test the proposed method. The road surface points were accurately extracted without being affected by the absence of raised curb and hanging objects over the road surface, that is, tree canopies and overhead power lines. The quantitative assessment of the proposed method was performed in terms of correctness, completeness and quality, which were 96.3, 94.2, and 90.9%, respectively.
TL;DR: A new method is proposed to simplify point cloud data by removing the least important points and updating the normal vectors and importance values progressively until user-specified reduction ratio is reached.
Abstract: With the development of modern 3D measurement technologies, it becomes easy to capture dense point cloud datasets. To settle the problem of pruning the redundant points and fast reconstruction, simplification for point cloud is a necessary step during the processing. In this paper, a new method is proposed to simplify point cloud data. The kernel procedure of the method is to evaluate the importance of points based on local entropy of normal angle. After the estimation of normal vectors, the importance evaluation of points is derived based on normal angles and the theory of information entropy. The simplification proceeds and finishes by removing the least important points and updating the normal vectors and importance values progressively until user-specified reduction ratio is reached. To evaluate the accuracy of the simplification results quantitatively, an indicator is determined by calculating the mean entropy of the simplified point cloud. Furthermore, the performance of the proposed approach is illustrated with two sets of validation experiments where other three classical simplification methods are employed for contrast. The results show that the proposed method performs much better than other three methods for point cloud simplification.
TL;DR: A ship detection algorithm that combines CFAR and CNN is proposed based on the CFAR global detection algorithm and image recognition with the CNN model that is more suitable for application to ship detection systems.
Abstract: To maintain national socio-economic development and maritime rights and interests, it is necessary to obtain the space location information of various ships. Therefore, it is important to detect the locations of ships accurately and rapidly. At present, ship detection is mainly carried out by combining satellite remote sensing imaging with constant false alarm rate (CFAR) detection. However, with the rapid development of satellite remote sensing technology, remote sensing data have gradually begun to show the characteristics of “big data”; additionally, the accuracy and speed of ship detection can be improved by analysing big data, such as by deep learning. Thus, a ship detection algorithm that combines CFAR and CNN is proposed based on the CFAR global detection algorithm and image recognition with the CNN model. Compared with the multi-level CFAR algorithm that is based on multithreading, the algorithm in this paper is more suitable for application to ship detection systems.
TL;DR: In this article, the authors validate the performance of three types of rainfall estimation algorithms viz-hydro estimation (HE), IMSRA (IMSRA) and integrated multi-satellite retrieval for GPM (IMERG) of Global Precipitation Mission (GPM) satellites.
Abstract: India Meteorological department (IMD) used INSAT-3D Metrological Satellite Imager data to drive two type rainfall estimation products viz-Hydro Estimate (HE) and INSAT Multi-Spectral Rainfall Algorithm (IMSRA) on half hourly rainfall rate and daily accumulated rainfall in millimeter (mm). Integrated Multi-Satellite Retrieval for GPM (IMERG) product is being derived by NASA and JAXA by using Global Precipitation Mission (GPM) satellites data. IMSRA and GPM (IMERG) are gridded data at 10 km spatial resolution and HE is available at pixel level (4 km at Nadir). IMD provides gridded rainfall data at 0.25° × 0.25° resolution which is based on wide coverage of 6955 actual observation. In present study, validation of INSAT-3D based Hydro Estimator (HE), INSAT Multi-Spectral Rainfall Algorithm (IMSRA) and Integrated Multi-Satellite Retrieval for GPM (IMERG) of Global Precipitation Mission (GPM) satellites are carried out with IMD gridded data set for heavy rainfall event during winter monsoon, over peninsular India (November–December 2015). In validation, Nash–Sutcliffe efficiencies (NSE), RMSE, Correlation, Skilled scores are calculated at grid level for heavy and very heavy rain categories and the values of NSE of HE (− 32.36, − 3.12), GPM (− 68.67, − 2.39) and IMSRA (− 0.02, 0.28) on 16th November 2015 and HE (− 13.65, − 1.69), GPM (− 43.79, − 2.94) and IMSRA (− 1.08, − 1.60) on 2nd Dec 2015, for heavy and very heavy rainfall. On both days, HE is showing better rainfall estimate compare to GPM for Heavy rainfall and GPM showing better estimation for very heavy rainfall events. In all the cases IMSRA is underestimating, if daily rain fall exceeded 75 mm.
TL;DR: The study explores the semantic modelling capabilities of the CityGML for the transformation of native 3D virtual city models to one satisfying capabilities like semantic information and support towards interoperability.
Abstract: Virtual 3D city models are increasingly being used to model the realms of the real world for utilization in a number of applications related to environmental simulations including, urban planning, mapping the energy characteristics of buildings, noise mapping, flood modelling, etc. Apart from geometric and appearance/textural information, these applications have a requirement for complex urban semantics. Currently, a number of 3D standards are available in CAD, BIM and GIS related domains for the storage, visualization and transfer of 3D geospatial datasets. Initially, the 3D data models (such as COLLADA, VRML, X3D, etc.) were purely graphical/geometrical in nature and mainly used for visualization purposes. With the inclusion of thematic modules in OGC CityGML, the integration of geometry and semantics in a single data model paved the way for better sharing of virtual 3D city models. In spite of the availability of a wide range of 3D data standards, there are certain differences with respect to geometry, topology, semantics, LODs, etc., which complicates the integration of 3D geodata from heterogeneous sources. This paper serves to highlights the need for the innovative solutions with respect to the urban environmental related simulations primarily based on the use of virtual 3D city models. Four use cases are studied in this context namely, (1) urban solar potential estimation using CityGML models, (2) simulation of traffic noise level mapped on building walls from the urban road segments, (3) CityGML based 3D data models interoperability, and (4) 3D indoor logistics and subsurface utilities. However, for modelling majority of use cases, CityGML does not provide explicit thematic representations but provides support for extending the CityGML schema using Application Domain Extensions. In a nutshell, the study explores the semantic modelling capabilities of the CityGML for the transformation of native 3D virtual city models to one satisfying capabilities like semantic information and support towards interoperability.
TL;DR: Three algorithms for the mapping of snow cover region over north Indian Himalayas (India) are compared and it is highlighted that topographically derived classifiers achieved better accuracy in mapping as compared to simple classifiers.
Abstract: In recent years, the significant increase in research on spatial information is observed. Classification or clustering is one of the well-known methods in spatial data analysis. Traditionally, classifiers are generally based on per-pixel approaches and are not utilizing the spatial information within pixel, called mixels which is an important source of information to image classification. There are two foremost reasons behind the existence of mixels: (a) coarse or low spatial resolution of sensor and (b) topographic effects that recorded on optical satellite imagery due to differential terrain illuminations over rugged areas such as Himalayas. In the present study, different classification algorithms have been implemented to drive the impact of topography on them. Among various available, three algorithms for the mapping of snow cover region over north Indian Himalayas (India) are compared: (a) maximum likelihood classification (MLC) as supervised classifier; (b) k-mean clustering as unsupervised classifier; and (c) linear spectral mixing model (LSMM) as soft classifier. These algorithms have been implemented on AWiFS multispectral data, and analysis was carried out. The classification accuracy is estimated by the error matrices, and LSMM achieved higher accuracy (84.5–88.5%) as compared to MLC (81–84%) and k-mean (74–81%). The results highlight that topographically derived classifiers achieved better accuracy in mapping as compared to simple classifiers. The study has many applications in snow hydrology, glaciology and climatology of mountain topography.
TL;DR: In this article, the authors used regression kriging to predict soil organic carbon (SOOC) using satellite images and digital elevation model for Lalo khala subwatershed (a part of Solani watershed) located in western Uttar Pradesh, India.
Abstract: Geo-spatial mapping of soil organic carbon using regression kriging was performed for Lalo khala sub-watershed (a part of Solani watershed) located in western Uttar Pradesh, India. Soil organic carbon was predicted using eight predictor variables derived from the advanced space borne thermal emission and reflection radiometer satellite images and digital elevation model. The soil organic carbon was determined in 248 soil samples collected randomly within a 300 m2 grid overlaid on the study area. Out of the eight predictor variables used in simple regression, the normalized difference vegetation index has the maximum correlation with the soil organic carbon (0.64) followed by vegetation temperature condition index (0.60), brightness index (− 0.60), greenness index (0.57) and wetness index (0.51). Standardized principle components of the predictor variables were used in the prediction model so as to address the multicollinearity problem. The regression kriging predicted SOC value ranged from 0.19 to 1.93% with a mean value of 0.64 and standard deviation of 0.29. The SOC values were higher in upper piedmont with moderate forest followed by Siwalik hills while low values were found in the upper alluvial plains. The RMSE of the predicted SOC map was only 0.196 indicating the closeness of predicted values to the observed values. Regression kriging predicted SOC map can be used for spatial agriculture planning and consider as an ideal input for spatially distributed models. The higher efforts for its preparation are justified when quality, spatial distribution and accuracy are considered.
TL;DR: In this paper, a geospatial analytical hierarchy process (AHP) model based on expert judgements has been applied in this study to map the potential avalanche susceptible zones in Siachen region of the Western Indian Himalaya.
Abstract: The main objective of this study was to mapping the potential avalanche susceptible zones in Siachen region of the Western Indian Himalaya. A geospatial analytical hierarchy process (AHP) model based on expert judgements has been applied in this study. Five most important terrain based avalanche occurrence parameters including slope, aspect, curvature, elevation and ground cover are employed in the present model. The ASTER GDEM and Landsat 8 OLI imagery are used to generate the avalanche occurrence parameters. A pairwise comparison matrix is computed to estimate the weight values for input terrain parameters. These weight values are then assigned to each respective avalanche occurrence parameter and employed in a geospatial AHP model to generate an avalanche susceptibility map. Finally, an avalanche inventory has been utilized to validate the results. The avalanche susceptibility map has been compared with the avalanche inventory map by calculating the area under the receiver operating characteristics curve (ROC-AUC) technique.
TL;DR: In this article, the authors assessed the reservoir sedimentation using Landsat 8 OLI satellite data combined with ancillary data, which was used to produce the water spread area of the reservoir and calculated the sedimentation rate.
Abstract: Reservoir sedimentation is the gradual accumulation of incoming sediments from upstream catchment leading to the reduction in useful storage capacity of the reservoir
Quantifying the reservoir sedimentation rate is essential for better water resources management Conventional techniques such as hydrographic survey have limitations including time-consuming, cumbersome and costly On the contrary, the availability of high resolution (both spatial and temporal) in public domain overcomes all these constraints This study assessed Jayakwadi reservoir sedimentation using Landsat 8 OLI satellite data combined with ancillary data Multi-date remotely sensed data were used to produce the water spread area of the reservoir, which was applied to compute the sedimentation rate The revised live storage capacity of the reservoir between maximum and minimum levels observed under the period of analysis (2015–2017) was assessed utilizing the trapezoidal formula The revised live storage capacity is assessed as 1942258 against the designed capacity of 2170935 Mm3 at full reservoir level The total loss of reservoir capacity due to the sediment deposition during the period of 41 years (1975–2017) was estimated as 228677 Mm3 (1053%) which provided the average sedimentation rate of 558 Mm3 year1 As this technique also provides the capacity of the reservoir at the different elevation on the date of the satellite pass, the revised elevation–capacity curve was also developed The sedimentation analysis usually provides the volume of sediment deposited and rate of the deposition However, the interest of the reservoir authorities and water resources planner’s lies in sub-watershed-wise sediment yield, and the critical sub-watersheds upstream reservoir requires conservation, etc Therefore, in the present study, Soil and Water Assessment Tool (SWAT) was used for the estimation of sediment yield of the reservoir The average annual sediment yield obtained from the SWAT model using 36 years of data (1979–2014) was 13144 Mm3 year−1 with the density of the soil (loamy and clay) of 144 ton m−3 The findings revealed that the rate of sedimentation obtained from the remote sensing-based methods is in agreement with the results of the hydrographic survey
TL;DR: In this article, the authors employed integration of weighted index overlay analysis (WIOA) and geographical information system (GIS) techniques to assess the groundwater potential zones in Krishna river basin, India and the validation of the result with existing groundwater levels.
Abstract: The groundwater occurrence and movement within the flow systems are governed by many natural factors like topography, geology, geomorphology, lineament structures, soil, drainage network and land use land cover (LULC). Due to complex natural geological/hydro-geological regime a systematic planning is needed for groundwater exploitation. It is even more important to characterize the aquifer system and delineate groundwater potential zones in different geological terrain. The study employed integration of weighted index overlay analysis (WIOA) and geographical information system (GIS) techniques to assess the groundwater potential zones in Krishna river basin, India and the validation of the result with existing groundwater levels. Different thematic layers such as geology, geomorphology, soil, slope, LULC, drainage density, lineament density and annual rainfall distribution were integrated with WIOA using spatial analyst tools in Arc-GIS 10.1. These thematic layers were prepared using Geological survey of India maps, European Digital Archive of Soil Maps, Bhuvan (Indian-Geo platform of ISRO, NRSC) and 30 m global land cover data. Drainage, watershed delineation and slope were prepared from the Shuttle Radar Topography Mission digital elevation model of 30 m resolution data. WIOA is being carried out for deriving the normalized score for the suitability classification. Weight factor is assigned for every thematic layer and their individual feature classes considering their significant importance in groundwater occurrence. The final map of the study area is categorized into five classes very good, good, moderate, poor and very poor groundwater potential zones. The result describes the groundwater potential zones at regional scale which are in good agreement with observed ground water condition at field level. Thus, the results derived can be very much useful in planning and management of groundwater resources in a regional scale.
TL;DR: In this article, the authors used the analytical hierarchy process (AHP) approach to map landslide in Kelantan river basin, Peninsular Malaysia using Landsat-8 and phased array type L-band synthetic aperture radar-2 (PALSAR-2) datasets.
Abstract: Integration of satellite remote sensing data and GIS techniques is an applicable approach for landslide mapping and assessment in highly vegetated regions with a tropical climate. In recent years, there have been many severe flooding and landslide events with significant damage to livestock, agricultural crop, homes, and businesses in the Kelantan river basin, Peninsular Malaysia. In this investigation, Landsat-8 and phased array type L-band synthetic aperture radar-2 (PALSAR-2) datasets and analytical hierarchy process (AHP) approach were used to map landslide in Kelantan river basin, Peninsular Malaysia. Landslides were determined by tracking changes in vegetation pixel data using Landsat-8 images that acquired before and after flooding. The PALSAR-2 data were used for comprehensive analysis of major geological structures and detailed characterizations of lineaments in the state of Kelantan. AHP approach was used for landslide susceptibility mapping. Several factors such as slope, aspect, soil, lithology, normalized difference vegetation index, land cover, distance to drainage, precipitation, distance to fault, and distance to the road were extracted from remotely sensed data and fieldwork to apply AHP approach. The excessive rainfall during the flood episode is a paramount factor for numerous landslide occurrences at various magnitudes, therefore, rainfall analysis was carried out based on daily precipitation before and during flood episode in the Kelantan state. The main triggering factors for landslides are mainly due to the extreme precipitation rate during the flooding period, apart from the favorable environmental factors such as removal of vegetation within slope areas, and also landscape development near slopes. Two main outputs of this study were landslide inventory occurrences map during 2014 flooding episode and landslide susceptibility map for entire Kelantan state. Modeled/predicted landslides with a susceptible map generated prior and post-flood episode, confirmed that intense rainfall throughout Kelantan has contributed to produce numerous landslides with various sizes. It is concluded that precipitation is the most influential factor for landslide event. According to the landslide susceptibility map, 65% of the river basin of Kelantan is found to be under the category of low landslide susceptibility zone, while 35% class in a high-altitude segment of the south and south-western part of the Kelantan state located within high susceptibility zone. Further actions and caution need to be remarked by the local related authority of the Kelantan state in very high susceptibility zone to avoid further wealth and people loss in the future. Geo-hazard mitigation programs must be conducted in the landslide recurrence regions for reducing natural catastrophes leading to loss of financial investments and death in the Kelantan river basin. This investigation indicates that integration of Landsat-8 and PALSAR-2 remotely sensed data and GIS techniques is an applicable tool for Landslide mapping and assessment in tropical environments.
TL;DR: Wang et al. as mentioned in this paper used a new type of remote sensing data to locate poverty-stricken areas based on nighttime light, taking Chongqing Municipality as a sample, and constructed a multidimensional poverty index (MPI) system, guided by a well-known and widely used conceptual framework of sustainable livelihood.
Abstract: Poverty has emerged as one of the chronic dilemmas facing the development of human society during the twenty first century. Accurately identifying regions of poverty could lead to more effective poverty-alleviation programs. This study used a new type of remote-sensing data, NPP-VIIRS, to locate poverty-stricken areas based on nighttime light, taking Chongqing Municipality as a sample, and constructed a multidimensional poverty index (MPI) system, guided by a well-known and widely used conceptual framework of sustainable livelihood. A regression model was constructed and results were correlated with those using the average nighttime light index. The model was then tested on Shaanxi Province, and average relative error of the estimated MPI was only 11.12%. These results showed that multidimensional poverty had a high spatial concentration effect at the regional scale. We then applied the index nationwide, at the county scale, analyzing 2852 counties, which we divided into seven classifications, based on the MPI: extremely low, low, relatively low, medium, relatively high, high, and extremely high. Eight hundred forty-eight counties in 26 provinces were identified as multidimensionally poor. Among these, 254 were absolutely poor counties and 543 were relatively poor counties; 195 of these are not on the list of poverty-stricken counties as identified by income levels alone. By improving the accuracy of targeting, this method of identifying multidimensional poverty areas could help the Chinese government improve the effectiveness of poverty reduction strategies, and it could also be used as a reference for other countries or regions that seek to target poor areas that suffer multidimensional deprivation.
TL;DR: In this paper, a study aimed to identify the exact interior sources of dust in Khuzestan Province (Iran), using a hybrid method of remote sensing, GIS and sedimentology.
Abstract: There is an urgent need for zoning the dust sources, as the first step to control dust storms. Hence, this study aimed to identify the exact interior sources of dust in Khuzestan Province (Iran), using a hybrid method of remote sensing, GIS and sedimentology. To this end, the spatial data of pedology, landuse, climate, slope and sedimentology were used as the constraint layers and vegetation, land surface temperature and soil moisture were used as the major layers. The major layers were extracted by performing the necessary computational process on the image of the Landsat 8 satellite. Constraint layers were used to eliminate the regions without dust production potential. In the next step, the major layers were weighted applying the paired comparison and fuzzy analytic hierarchy process methods. Then, the final integration of the layers took place by multiplying each major layer in the respective weight, and the map of the dust sources in the region was prepared. To validate the results, field trips were done for 180 points of the sources which indicate the high accuracy of the identified regions. The results revealed that 9% of the area in Khuzestan Plain equal to 350,000 ha is regarded as the source of dust production. Moreover, according to the results, it can be said that satellite images, especially those with efficient resolutions such as Landsat 8 products, are suitable basis data for extracting indicators (temperature, humidity and vegetation) of the dust sources.
TL;DR: In this article, a study was conducted to understand the dynamics of shifting cultivation through the use of Landsat time-series data from 1999 to 2016 in Champhai district of Mizoram.
Abstract: The dynamics of crop-fallow rotation cycles of shifting cultivation has been poorly understood in northeastern part of the country although it is one of the major land use systems in the hilly states of this region. The present study was conducted to understand the dynamics of shifting cultivation through the use of Landsat time-series data from 1999 to 2016 in Champhai district of Mizoram. We mapped the current jhum fields and abandoned areas of each imagery of the study period and performed a post classification comparison to assess the crop-fallow rotation cycle/jhum cycle. The chrono-sequential change of slash and burn area over the past 17 years showed a decreasing trend with the greater part of the shifting cultivation area being occupied by 2nd year crop fields, covering 48.81% of total jhum land. On average, 114.46 km2 area were annually slashed for current jhum, out of which 33.41% continued with current jhum 2nd year cropping and only 3.27% of jhumias continued with 3rd year cropping. The shifting cultivation patches were mostly confined to moderately steep slopes (15°–30°). East facing aspect was mostly preferred and North facing aspect was least preferred. During the study period, 10 years jhum cycle covered the maximum area followed by 9 years and 11 years jhum cycle. The end result of this study proved that the prevalent jhum cycle in Champhai district is 8–11 years with a fallow period of 6–9 years.
TL;DR: In this paper, a novel object based approach has been proposed for automatic and robust detections as well as extraction of the building in high spatial resolution images, simultaneously making use of both stable and variable features.
Abstract: With the advent of high spatial resolution satellite imagery, automatic and semiautomatic building extractions have turned into one of the outstanding research topics in the field of remote sensing and machine vision. To this date, various algorithms have been presented for extracting the buildings from satellite images. Such methods lend their bases to diverse criteria such as radiometric, geometric, edge detection, and shadow. In this paper, a novel object based approach has been proposed for automatic and robust detections as well as extraction of the building in high spatial resolution images. To fulfill this, we simultaneously made use of both stable and variable features. While the former can be derived from inherent characteristics of the buildings, the latter is extracted using a feature analysis tool. In addition, a novel perspective has been recommended to boost the automation degree of the segmentation part in the object based analysis of remote sensing imagery. The proposed method was applied to a QuickBird imagery of an urban area in Isfahan city and the results of the quantitative evaluation demonstrated that the proposed method could yield promising results. Moreover, in another section of this study, for assessing the algorithm transferability, the rule set was implemented to a part of the WorldView image of Yazd city, proving that the proposed approach is capable of transferability in different types of case studies.
TL;DR: It was found that object-based approach outperformed the traditional pixel- based approach for all cases (up to 18% improvement), and the RF classifier produced significantly more accurate results than the NN classifier.
Abstract: Object-based image analysis (OBIA) has attained great importance for the delineation of landscape features, particularly with the accessibility to satellite images with high spatial resolution acquired by recent sensors. Statistical parametric classifiers have become ineffective mainly due to their assumption of normal distribution, vast increase in the dimensions of the data and availability of limited ground sample data. Despite pixel-based approaches, OBIA takes semantic information of extracted image objects into consideration, and thus provides more comprehensive image analysis. In this study, Indian Pines hyperspectral data set, which was recorded by the AVIRIS hyperspectral sensor, was used to analyse the effects of high dimensional data with limited ground reference data. To avoid the dimensionality curse, principal component analysis (PCA) and feature selection based on Jeffries–Matusita (JM) distance were utilized. First 19 principal components representing 98.5% of the image were selected using the PCA technique whilst 30 spectral bands of the image were determined using JM distance. Nearest neighbour (NN) and random forest (RF) classifiers were employed to test the performances of pixel- and object-based classification using conventional accuracy metrics. It was found that object-based approach outperformed the traditional pixel-based approach for all cases (up to 18% improvement). Also, the RF classifier produced significantly more accurate results (up to 10%) than the NN classifier.
TL;DR: In this paper, where and when temporal and spatial heterogeneity occurs is tried to be explained by taking human intervention in landscape pattern and processes in and around the city of Denizli into account and how this heterogeneity affects habitat conditions within the scope of landscape ecology.
Abstract: Perforation, dissection, fragmentation, shrinkage and attrition in ecosystems take place due to urbanization In this study, where and when temporal and spatial heterogeneity occurs is tried to be explained by taking human intervention in landscape pattern and processes in and around the city of Denizli into account and how this heterogeneity affects habitat conditions within the scope of landscape ecology Landscape pattern metrics were estimated in order to reveal the change in habitats and present the properties of the landscape 30 pattern indicators on class and pattern levels, which are important to show human–environment interaction, were analyzed in order to indicate the features of the landscape such as area, side, shape and dispersion To this end, LANDSAT TM/7–ETM/8-OLI satellite images of 1987 and 2013 were classified for laying the foundations of the analysis Analyses showed that between 1987 and 2013, complicated shape features, increase in edge habitats, de-growth in core areas and eventually fragmentation in landscape have been dominant Heterogenic structure in landscape has increased This points not to the self-functioning of the landscape, but to the domination of human intervention over the landscape Particularly, due to urban growth and sprawl, fragmentation, isolation and habitat loss in croplands have increased This study sets forth the usefulness of remote sensing, GIS and landscape metrics in understanding how urban dynamics and ecosystems change in developing urban politics