Abstract
Abstract. Flood contributes a key role in devastating natural and man-made areas. Floods usually are occurred when there is a considerable number of clouds in the sky making optic data useless. Synthetic aperture radar (SAR) images can be a valuable data source in earth observation tasks. The most important characteristic of the radar image is its ability to penetrate the cloud and dust. Therefore, monitoring earth in cloudy or rainy weather can be available by this kind of dataset. In the last few years by improving machine learning methods and development of convolutional neural networks in remote sensing applications we are facing with extremely high improvement in classification tasks. In this paper, we use dual-polarized VV and VH backscatter values of Sentinel-1 and Shuttle Radar Topography Mission (SRTM) digital elevation model (DEM) dataset in a proposed convolutional neural network to generate a land cover map of a flooded area before and after happening. Obtained classification results vary between 93.3% to 98.5% for different training sizes. By comparing the generated classified maps, flooded areas of each class can be extracted.
Highlights
Flood contributes a key role in devastating natural and man-made areas
Convolutional Neural Networks (CNN) is popular in remote sensing image classification because it alleviates the problem of big data analysis that has been always an issue in satellite image processing
We investigate the use of a 2D CNN in flood mapping of Pol-e-Dokhtar city, Lorestan, Iran using two –time Sentinel-1 dataset, taken before and after the flood, one at 3rd of March of 2019 and another at 2rd of April of the same year
Summary
Flood contributes a key role in devastating natural and man-made areas. In March 2019, a destructive inundation occurred in Iran which put down the demise of hundreds of people and extinction of properties. Flood mapping using remote sensing techniques has the merit of helping authorities to have a thorough overview of working out the amount of damage, and it alleviates the emergency procedures (Benoudjit and Guida, 2019; Domeneghetti et al, 2019). Flood mapping using SAR images can be a necessity. One way to map flood is by using Change detection approaches when two-time data of a region are available (Zhao et al, 2019). One of conventional change detection approaches is post-classification comparison technique where we compare the classification results of the data at two different times. Conventional classification algorithms like SVM and Neural Network cannot handle the disturbances and distortions existence in the SAR dataset and we need an integrated classification system capable of doing feature extraction and partitioning simultaneously. CNN is popular in remote sensing image classification because it alleviates the problem of big data analysis that has been always an issue in satellite image processing
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
