Browsing by Author "Mustak, Sk."

Now showing 1 - 3 of 3

Building Extraction of Kolkata Metropolitan Area Using Machine Learning and Earth Observation Datasets
(Springer Science and Business Media Deutschland GmbH, 2023-05-23T00:00:00) Barman, Prosenjit; Mustak, Sk.
Rural-to-urban migration and increasing population has created urban agglomeration, particularly in metropolitan cities. This agglomeration creates pressure on cities which interrupts the city to become sustainable. Evaluation of the pattern of urban growth pattern is a crucial task for a city�s long-term development. The building footprint is one of the most important features of a city form to support urban management and development. Previous studies show that high-resolution images are robust to extract building footprints using machine learning algorithms. The main objective of this study is to extract the building footprint from the satellite imagery using machine learning algorithms. In this study, Sentinel-2 multispectral satellite imagery and support vector machine (SVM) linear and radial basis function (RBF) have been used to extract the building footprints in Kolkata metropolitan area. In addition, both pixel-based and object-based image classification approaches have been applied and compared in this study. This result shows that in pixel-based image classification SVM linear gives a high accuracy than the SVM RBF. The accuracy level of SVM linear is 92.58% while Kappa is 0.89. On the other hand, object-based image analysis LULC classification has been done using the SVM ML algorithm. In this image classification, the SVM RBF kernel type gives high accuracy. The overall accuracy of this OBIA image classification is 91.58% and the Kappa is 0.87. For the building extraction in an urban area from the medium-resolution image Sentinel 2 using a machine learning algorithm with high accuracy gives a significant approach. Policymakers and planners can develop the city sustainably from this building footprint in an urban region and use sustainable urban planning to achieve the Sustainable Development Goal. � 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.
Prediction of soil erosion risk using earth observation data under recent emission scenarios of CMIP6
(Taylor and Francis Ltd., 2021-08-26T00:00:00) Kumar, Nirmal; Singh, Sudhir Kumar; Dubey, Amit Kumar; Ray, Ram L.; Mustak, Sk.; Rawat, Kishan Singh
The earth observation data and CMIP6 models were used to predict plausible soil loss from the Ghaghara river basin. The decadal prediction of soil loss (28.64 ton/ha/year) was found high for SSP585 of CanESM5 during 2015�2025. However, the lower value was reported as 21.71 ton/ha/year for SSP245 of MRI-ESM2-0 during 2035�2045. The century level future rainfall erosivity factor was found lowest for SSP245, however highest for SSP585 of Access-ESM1-5, CanESM5, and IPSL-CM6A-LR. The SSP585 (Access-ESM1-5, CanESM5, and IPSL-CM6A-LR) have maximum soil erosion rate as 29.07, 28.03, and 28.0 ton/ha/year, respectively. For the SSP585, increments were observed as 35.93%, 31.04%, and 30%, respectively, compared to the baseline year (2014). Whereas, lowest was reported as 21.7 and 24.9 ton/ha/year and consequently the low increment as 1.31% and 16.55% for both scenarios of MRI-ESM2-0 compared to baseline. We observed that the soil erosion rate is aligned with the predicted rainfall erosivity factor. � 2021 Informa UK Limited, trading as Taylor & Francis Group.
Vulnerability and risk assessment mapping of Bhitarkanika national park, Odisha, India using machine-based embedded decision support system
(Frontiers Media SA, 2023-09-29T00:00:00) Mohanty, Shantakar; Mustak, Sk.; Singh, Dharmaveer; Van Hoang, Thanh; Mondal, Manishree; Wang, Chun-Tse
The vulnerability and flood risk assessment of Bhitarkanika National Park in Odisha, India, was conducted using a data-driven approach and a machine-based embedded decision support system. The park, located in the estuaries of the Brahmani, Baitarani, Dharma, and Mahanadi river systems, is home to India�s second-largest mangrove environment and the world�s most active and diverse saline wetland. To evaluate its vulnerability and risk, various threats were considered, with a focus on floods. Satellite imageries, such as Landsat 8 OLI, SRTM digital elevation model, open street map, Google pro image, reference map, field survey, and other ancillary data, were utilized to develop vulnerability and risk indicators. These indicators were then reclassified into �Cost� and �Benefit� categories for better understanding. The factors were standardized using the max-min standardization method before being fed into the vulnerability and risk model. Initially, an analytical hierarchy approach was used to develop the model, which was later compared with machine learning algorithms (e.g., SVM) and uncertainty analysis indices (e.g., overall accuracy, kappa, map quality, etc.). The results showed that the SVM-RBF machine learning algorithm outperformed the traditional geostatistical model (AHP), with an overall accuracy of 99.54% for flood risk mapping compared to AHP�s 91.12%. The final output reveals that a large area of Bhitarkanika National park falls under high flood risk zone. The Eastern coastal regions of Govindapur, Kanhupur, Chinchri, Gobardhanpur and Barunei fall under high risk zone of tidal floods, The Northern and western regions of Ramachandrapur, Jaganathpur, Kamalpur, Subarnapur, Paramanandapur, etc., Fall under high risk region of riverine floods. The study also revealed that the areas covered with mangroves have a higher elevation and hence are repellent to any kind of flood. In the event of a flood high priority conservation measures should be taken along all high flood risk areas. This study is helpful for decision-making and carrying out programs for the conservation of natural resources and flood management in the national park and reserve forest for ecological sustainability to support sustainable development goals (e.g., SDGs-14, 15). Copyright � 2023 Mohanty, Mustak, Singh, Van Hoang, Mondal and Wang.