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

Abstract

Precise coastal shoreline mapping is essential for monitoring changes in erosion rates, surface hydrology, and ecosystem structure and function. Monitoring water bodies in the Arctic National Wildlife Refuge (ANWR) is of high importance, especially considering the potential for oil and natural gas exploration in the region. In this work, we propose a modified variant of the Deep Neural Network based U-Net Architecture for the automated mapping of 4 Band Orthorectified NOAA Airborne Imagery using sparsely labeled training data and compare it to the performance of traditional Machine Learning (ML) based approaches—namely, random forest, xgboost—and spectral water indices—Normalized Difference Water Index (NDWI), and Normalized Difference Surface Water Index (NDSWI)—to support shoreline mapping of Arctic coastlines. We conclude that it is possible to modify the U-Net model to accept sparse labels as input and the results are comparable to other ML methods (an Intersection-over-Union (IoU) of 94.86% using U-Net vs. an IoU of 95.05% using the best performing method).

Shoreline change rates and land to sea sediment and soil organic carbon transfer in eastern Parry Peninsula from 1965 to 2020 (Amundsen Gulf, Canada)

“…The application of automatic coastal delineation algorithms using remote sensing imagery is usually based on the detection of the waterline through spectral indices, pixel segmentation or and various classification techniques, with recent examples being CoastSat, Waterdetect or XGBoost (Vos et al, 2019;Aryal et al, 2021;Cordeiro et al, 2021). Our comparative analysis has shown that the shoreline changes and volume transfer estimates can be very different depending on the shoreline proxy used -backshore vs. foreshore.…”

Convolutional Neural Network and Optical Flow for the Assessment of Wave and Tide Parameters from Video Analysis (LEUCOTEA): An Innovative Tool for Coastal Monitoring

“…Although LEUCOTEA allowed to assess some parameters of the energetic balance on the coasts, further studies are needed to integrate the deep learning techniques with other architectures to assess the mass balance [16,91,107]. The assessment of mass balance requires the implementation of different machine learning models, which could involve classification and segmentation techniques for the quantitative evaluation of coastal landforms and sediment typologies [108,109]. For the energetic balance, a full framework of all components must be realized through a joint assessment of current, tide, wave, and wind parameters.…”

Multiscale Object-Based Classification and Feature Extraction along Arctic Coasts

“…The scale parameter is used to control average image object size [41,42] where a higher value results in bigger objects and a smaller value results in smaller objects. The scale parameter has been considered the primary factor for segmentation in object-based research [43,44] and the determination of the scale parameter depends on factors such as the sensor type, resolution, the purpose of the segmentation, and objects of interest. As a result, we incremented our scale parameter by 100 between 100 and 800 in our threshold-based and supervised classifications to better understand the influence of object size on accuracy.…”

“…Object-based and deep learning applications have been applied to infrastructure detection in the Arctic [32], ice-wedge polygon mapping [33][34][35], and recently in detecting RTS [29]. Broadly, deep learning neural networks have been successfully demonstrated in image classification, segmentation, and object detection, leading to substantial application in remote sensing [36][37][38][39][40][41], coastal erosion [42][43][44][45], and geomorphology [46][47][48].…”