The Efficiency of Random Forest Method for Shoreline Extraction From Landsat-8 and Gokturk-2 Imageries

Abstract

ABSTRACT:Coastal monitoring plays a vital role in environmental planning and hazard management related issues. Since shorelines are fundamental data for environment management, disaster management, coastal erosion studies, modelling of sediment transport and coastal morphodynamics, various techniques have been developed to extract shorelines. Random Forest is one of these techniques which is used in this study for shoreline extraction.. This algorithm is a machine learning method based on decision trees. Decision trees analyse classes of training data creates rules for classification. In this study, Terkos region has been chosen for the proposed method within the scope of "TUBITAK Project (Project No: 115Y718) titled" Integration of Unmanned Aerial Vehicles for Sustainable Coastal Zone Monitoring Model -Three-Dimensional Automatic Coastline Extraction and Analysis: Istanbul-Terkos Example ". Random Forest algorithm has been implemented to extract the shoreline of the Black Sea where near the lake from LANDSAT-8 and GOKTURK-2 satellite imageries taken in 2015. The MATLAB environment was used for classification. To obtain land and waterbody classes, the Random Forest method has been applied to NIR bands of LANDSAT-8 (5 th band) and GOKTURK-2 (4 th band) imageries. Each image has been digitized manually and shorelines obtained for accuracy assessment. According to accuracy assessment results, Random Forest method is efficient for both medium and high resolution images for shoreline extraction studies.

Semi-Automated Semantic Segmentation of Arctic Shorelines Using Very High-Resolution Airborne Imagery, Spectral Indices and Weakly Supervised Machine Learning Approaches

“…Recently, high performance computing, ML, and deep learning approaches have provided solutions for efficient and accurate landscape feature mapping across difference ecosystems. In the Arctic, studies have delineated polygonal tundra geomorphologies [45,79], arctic lake features [80], glacier extents [48,81,82], and coastal features [40][41][42]83]. In this research, we propose an automated pipeline using traditional ML based methods-random forest, and xgboost, and a deep neural network based U-Net architecture for arctic coastal mapping and compare their performances.…”

“…Developing effective automated/semi-automated techniques for accurate land cover mapping is just as necessary as the availability of VHR remote sensing images since the periodic manual labeling of remote sensing images over a large area is practically impossible. Fortunately, over the years, there have been advancements in areas of computer vision and remote sensing that allow researchers to automate this task using various ML methods [40][41][42]. The introduction of geospatial cloud computing platforms, like Google Earth Engine, have allowed researchers to harness the power of distributed computing for processing dense stacks of satellite imagery and performing machine learning routines for the extraction of coastline data [43].…”

On the shoreline positioning via remote sensing imagery: an isoradiometric approach

“…Pixel-based approaches can be divided into categories including band thresholding (e.g., Ref. 21); water indices, such as the normalized difference water index; 22 raster-based approaches, such as random forest; 23 and vector-based approaches, such as object-based image analysis 24 . These approaches usually produce a shoreline that follows the shape of the pixels, even though this pixel-bounded shoreline can be smoothed using vector generalizing algorithms.…”

A radiometric contouring approach to map the shoreline

“…Pixel-based approaches can be further categorized into image thresholding, water indices 10 , raster-based approaches 11 , and vector-based 12 . We introduce a novel methodology based on raster contouring, wherein the shoreline is delineated by treating pixels as concurrent points of radiometric remote measurements.…”