Rohit Sharma

Rohit Sharma

Auto-generated profile, based on publicly available information

Rohit Sharma

Department of Civil Engineering, Indian Institute of Technology Roorkee, India

Research Focus & Metrics

Academic Career Length

2Years (2025-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
12 July 2025 in ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Verified
Journal Article
Open access
PDF file
RSRohit Sharma
Rohit Sharma
PRPritipadmaja
Pritipadmaja
ZAZulfequar Ahmad
Zulfequar Ahmad
RGRahul Dev Garg
Rahul Dev Garg
Accurate land cover classification in riverine environments is essential for understanding hydrological dynamics, ecological health, and resource management. This study utilizes high-resolution LISS-4 imagery from the IRS-R2 satellite to address the complex classification challenges in the dynamic confluence of the Mahanadi and Shivnath Rivers. To enhance classification accuracy, two distinct feature sets were developed: the first feature set (FS-1) contained the original three spectral bands of the imagery, and the second feature set (FS-2) added eight derived spectral indices to improve class separation. Three machine learning classifiers, Random Forest (RF), Support Vector Machine (SVM), and Gradient Tree Boosting (GTB), were applied to assess the effectiveness of each feature set. The results demonstrated that the combination of FS-2 and RF yielded the most accurate and interpretable classification, achieving an overall accuracy of 84.48% and a Cohen’s Kappa coefficient of 0.81. Visual results demonstrated precise delineation of narrow water channels and sandbars, when using the enhanced feature set FS-2 with the Random Forest (RF) classifier, the model achieved an F1-score of 86.96% for dense vegetation and 83.78% for water. FS-2 consistently improved F1-scores across all classifiers, aiding in distinguishing visually similar classes like wet sand, dry sand, and sparse vegetation. These findings highlight the value of combining spectral indices with high-resolution imagery to achieve accurate land cover classification in complex landscapes, supporting applications in ecological monitoring, flood risk assessment, and resource management for riverine ecosystems.

30 December 2025 in Geografický časopis / Geographical Journal
Verified
Journal Article
Open access
MRMiloš Rusnák
Miloš Rusnák
RSRohit Sharma
Rohit Sharma
VSValentín Sočuvka
Valentín Sočuvka
ASAshraf
Ashraf
TGTomáš Goga
Tomáš Goga
DRDev Garg Rahul
Dev Garg Rahul
Slovenská akadémia vied vykonáva špičkový základný výskum, dlhodobý strategický a aplikovaný výskum a vývoj a poskytuje prístup k vedeckej infraštruktúre.

18 March 2025
Verified
Other
Open access
RSRohit Sharma
Rohit Sharma
ZAZulfequar Ahmad
Zulfequar Ahmad
RGRahul Dev Garg
Rahul Dev Garg
Precise land cover classification in riverine environments is vital for effective hydrological studies and resource management. This study evaluates the influence of feature set design and machine learning classifiers on land cover classification accuracy in the confluence of the Mahanadi and Shivnath Rivers using high-resolution LISS-4 imagery from the IRS-R2 satellite. Two feature sets were developed. Feature Set 1 includes the original three spectral bands and twelve derived spectral indices, making a total of fifteen features. Feature Set 2 incorporates the original bands along with twenty-one Gray-Level Co-occurrence Matrix features computed from the original three LISS-4 bands, resulting in a total of twenty-four features. The GLCM features were derived from the original bands and include metrics such as Angular Second Moment, Contrast, Correlation, Entropy, and Variance. Four machine learning classifiers were applied to each feature set, namely Random Forest, Support Vector Machine, Gradient Tree Boosting, and Classification and Regression Tree, generating eight classified images. The best performance was observed with Feature Set 2 using the Support Vector Machine classifier, achieving an overall accuracy of 86.2 percent and a Cohen’s Kappa coefficient of 0.827. Feature Set 1 with Random Forest also performed well, with an overall accuracy of 84.5 percent and a Kappa coefficient of 0.806, and was particularly effective in delineating narrow water channels. Feature Set 2 consistently improved class separability for challenging classes such as wet sand, dry sand, and sparse vegetation, highlighting the potential of texture-based features in enhancing classification accuracy. The findings demonstrate the value of integrating spectral indices and GLCM texture features with high-resolution imagery for detailed land cover classification in complex riverine systems. This approach provides valuable insights for ecological monitoring, flood risk assessment, and sustainable management of riverine resources.

21 January 2025
Verified
Book Chapter
Closed access
PDF file
RSRohit Sharma
Rohit Sharma
ZAZ. Ahmad
Z. Ahmad
RGR. D. Garg
R. D. Garg
PGP. K. Garg
P. K. Garg

31 July 2025 in The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Verified
Journal Article
Open access
PDF file
PRPritipadmaja
Pritipadmaja
RSRohit Sharma
Rohit Sharma
RGRahul Dev Garg
Rahul Dev Garg
India is experiencing a significant intensification of heat extremes, with both hot days and warm nights becoming more frequent and severe in recent years. Studying these evolving patterns is crucial for understanding regional climate trends and addressing the growing vulnerabilities. This study focuses on the temporal trends in heat extremes by analysing the increase in the frequency of very hot days and very warm nights across Indian districts. By comparing data from a baseline period (2003–2012) to a recent period (2019–2023), the research aims to capture the evolving spatial patterns of heat extremes and assess their implications for vulnerable populations. The analysis was conducted using MODIS Aqua Land Surface Temperature (LST) daily products available on Google Earth Engine. The study derived thresholds for very hot days (90th percentile of daytime LST) and very warm nights (90th percentile of nighttime LST) based on the baseline period. These thresholds were applied to the recent period to quantify the increasing frequency of very hot days and warm nights across all districts. Temporal and spatial variations in the frequency of very hot days and warm nights were analysed to identify emerging hotspots and trends. The analysis reveals a significant increase in heat extremes across India, with both hot days and warm nights becoming more frequent and intense in recent years. These patterns highlight the complex thermal dynamics shaped by topography, urbanisation, and atmospheric processes, emphasising the need for targeted adaptation and mitigation measures. This study comprehensively assesses the temporal and spatial trends in heat extremes across India with a scientific foundation for targeted heat mitigation strategies and evidence-based policy-making for planning, agricultural strategies, and disaster management. Mapping heat extremes provides a robust foundation for designing policies.