Deep Learning Applied to Water Segmentation

Abstract

Abstract. The use of deep learning (DL) with convolutional neural networks (CNN) to monitor surface water can be a valuable supplement to costly and labour-intense standard gauging stations. This paper presents the application of a recent CNN semantic segmentation method (SegNet) to automatically segment river water in imagery acquired by RGB sensors. This approach can be used as a new supporting tool because there are only a few studies using DL techniques to monitor water resources. The study area is a medium-scale river (Wesenitz) located in the East of Germany. The captured images reflect different periods of the day over a period of approximately 50 days, allowing for the analysis of the river in different environmental conditions and situations. In the experiments, we evaluated the input image resolutions of 256 × 256 and 512 × 512 pixels to assess their influence on the performance of river segmentation. The performance of the CNN was measured with the pixel accuracy and IoU metrics revealing an accuracy of 98% and 97%, respectively, for both resolutions, indicating that our approach is efficient to segment water in RGB imagery.

East Asia Hydrographic Commission - Fifty Years of Progress

“…Sometimes, problems can occur when the image is over-or underexposed in the bank area. Recently, however, novel routines have been published to recognise and segment water surfaces in images via deep learning (see Figure 5 by Akiyama et al, 2020). Assuming a model being trained on various images showing running water surfaces, this approach might, on the one hand, solve the exposure problem and, on the other hand, further reduce the effort for water line detection using a single image of the riverbank instead of an image sequence.…”

Toward Transferable Water Level Trend Monitoring Using River Cameras: Integrating Domain-Specific Models and Segment Anything Model (SAM) for Water Segmentation

“…Currently, water segmentation mainly relies on semantic segmentation deep learning models for their high automation and scalability (Eltner et al, 2018). A series of classic model structures, including SegNet, Fully Convolutional Networks, Fully Convolutional DenseNets, and Conditional Adversarial Networks, have already been applied to water segmentation for river camera images (Akiyama et al, 2020;Erfani et al, 2022;Lopez-Fuentes et al, 2017). However, deep learning models are significantly influenced by the amount of training data.…”

Classification of River Sediment Fractions in a River Segment including Shallow Water Areas Based on Aerial Images from Unmanned Aerial Vehicles with Convolution Neural Networks

“…Major examples of semantic segmentation as applied to rivers are the river channel detection and the river width measurement using satellite images [32][33][34]. As for the examples related to river management, there have been attempts to detect estuary sandbars [35], to monitor water levels during floods [36][37][38][39], to conduct the water binary segmentation task to aid in fluvial scene autonomous navigation [40], and to detect fine-grained river ice [41]. The number of research on semantic segmentation for river scenes is less than that for terrestrial areas, because the benchmark datasets are smaller than those for terrestrial data.…”