ItemArticle

A generic framework for forecasting lake surface area dynamics using level set segmentation and double exponential smoothing

Title
A generic framework for forecasting lake surface area dynamics using level set segmentation and double exponential smoothing
Creator
Bijeesh, T.V.; Narasimhamurthy, K.N.; Sathish, P.K.; Bejoy, B.J.; Michael Moses, J.
Description
Water has been a crucial element for the sustenance of civilization throughout history and civilizations have sprung up around a body of water in one form or another. It becomes imperative to address the pressing issue of water shortage and the shrinking size of urban water bodies, which is particularly relevant in Indian cities like Bangalore. The effective management and preservation of these invaluable resources depend on the development of accurate and automated tools to monitor them. The proposed framework introduces a novel approach, combining a level set-based segmentation algorithm with double exponential smoothing to monitor water bodies using multispectral satellite images. In-depth review of nine lakes within Bangalore was carried out using a Landsat time series data set spanning 1987 to 2020. The resulting forecasting model, employing a univariate smoothing methodology, showcased exceptional performance metrics. Notably, it yielded an average error of 0.072 and exhibited a robust correlation coefficient of 0.94 when cross-referenced with proven results. The proposed framework holds great potential for practical implementation in the domain of long-term water body analysis, effectively catering to the requirements of administrative and decision-making entities. Moreover, the adaptability of this framework for the incorporation of additional external factors, as well as its potential to analyze seasonal dynamics, offers exciting avenues for further exploration. The dataset of delineated lake images prepared in this study presents an opportunity for the advancement of image-to-image regression networks, enabling the prediction of both area and shape variations for lakes, thereby enhancing predictive accuracy and insights. The Author(s), under exclusive licence to Springer Nature Switzerland AG 2025.
Source
Sustainable Water Resources Management;Volume;11;Issue;4;Article No.;84;
Date
01-01-2025
Publisher
Springer Science and Business Media Deutschland GmbH
Subject
Change detection framework; Double exponential smoothing; Landsat multi-temporal data; Water body change monitoring
Coverage
Bijeesh T.V., Department of CSE, CHRIST University, Kengeri Campus, Karnataka, Bangalore, 560060, India; Narasimhamurthy K.N., Department of Civil Engineering, CHRIST University, Kengeri Campus, Karnataka, Bangalore, 560060, India; Sathish P.K., Department of CSE, CHRIST University, Kengeri Campus, Karnataka, Bangalore, 560060, India; Bejoy B.J., Department of CSE, CHRIST University, Kengeri Campus, Karnataka, Bangalore, 560060, India; Michael Moses J., Department of CSE, CHRIST University, Kengeri Campus, Karnataka, Bangalore, 560060, India
Rights
Restricted Access; Hardcopy may be available in the library
Relation
ISSN: 23635037;
Format
online
Language
English
Type
Article

Citation

Bijeesh, T.V.; Narasimhamurthy, K.N.; Sathish, P.K.; Bejoy, B.J.; Michael Moses, J., “A generic framework for forecasting lake surface area dynamics using level set segmentation and double exponential smoothing,” CHRIST (Deemed To Be University) Institutional Repository, accessed June 18, 2026, https://archives.christuniversity.in/items/show/22081.

Output Formats