Abstract
Land cover classification and investigation of temporal changes are considered to be common applications of remote sensing. Water/non-water region estimation is one of the most fundamental classification tasks, analyzing the occurrence of water on the Earth’s surface. However, common remote sensing practices such as thresholding, spectral analysis, and statistical approaches are not sufficient to produce a globally adaptable water classification. The aim of this study is to develop a formula with automatically derived tuning parameters using perceptron neural networks for water/non-water region estimation, which we call the Perceptron-Derived Water Formula (PDWF), using Landsat-8 images. Water/non-water region estimates derived from PDWF were compared with three different approaches—Modified Normalized Difference Water Index (MNDWI), Automatic Water Extraction Index (AWEI), and Deep Convolutional Neural Network—using various case studies. Our proposed method outperforms all three approaches, showing a significant improvement in water/non-water region estimation. PDWF performance is consistently better even in cases of challenging conditions such as low reflectance due to hill shadows, building-shadows, and dark soils. Moreover, our study implemented a sunglint correction to adapt water/non-water region estimation over sunglint-affected pixels.
Highlights
The location and persistence of Earth’s surface water is changing due to various factors such as change in climate, seasons, and human activities
In order to overcome these challenges, we propose a formula with perceptron-learning-derived parameters for water/non-water region estimation without a need for optimizing the threshold regionally
This study investigated perceptron learning aspects which affect the performance of the perceptron learning and performed several comparisons to generate suitable parameters and thresholds
Summary
The location and persistence of Earth’s surface water is changing due to various factors such as change in climate, seasons, and human activities. As Earth’s surface water is a vital resource, it is important to analyze its spatial and temporal changes. Accurate water/non-water region estimation is a key task for coastal change analysis, river channel change analysis [1], and shore/coast line extraction [2]. Accurate and automatic water/non-water region estimation is an important task. Several remote sensing image datasets are available that capture water bodies from around the globe, such as the Landsat mission and the Moderate Resolution Imaging Spectroradiometer (MODIS) [3,4,5]. Landsat-8 is a multispectral remote sensing sensor with 30 m spatial resolution and 16 days revisit capacity and is the latest in a continuous series of Landsat missions that began in
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
