Realistic River Image Synthesis Using Deep Generative Adversarial Networks

Akshat Gautam; Muhammed Sit; Ibrahim Demir

doi:10.3389/frwa.2022.784441

Back

Realistic River Image Synthesis Using Deep Generative Adversarial Networks

Journal article

Open access

Peer reviewed

Realistic River Image Synthesis Using Deep Generative Adversarial Networks

Akshat Gautam, Muhammed Sit and Ibrahim Demir

Frontiers in water, Vol.4, 784441

02/01/2022

DOI: 10.3389/frwa.2022.784441

Files and links (1)

url

https://doi.org/10.3389/frwa.2022.784441View

Published (Version of record) Open Access

Abstract

In this paper, we demonstrated a practical application of realistic river image generation using deep learning. Specifically, we explored a generative adversarial network (GAN) model capable of generating high-resolution and realistic river images that can be used to support modeling and analysis in surface water estimation, river meandering, wetland loss, and other hydrological research studies. First, we have created an extensive repository of overhead river images to be used in training. Second, we incorporated the Progressive Growing GAN (PGGAN), a network architecture that iteratively trains smaller-resolution GANs to gradually build up to a very high resolution to generate high quality (i.e., 1,024 × 1,024) synthetic river imagery. With simpler GAN architectures, difficulties arose in terms of exponential increase of training time and vanishing/exploding gradient issues, which the PGGAN implementation seemed to significantly reduce. The results presented in this study show great promise in generating high-quality images and capturing the details of river structure and flow to support hydrological modeling and research.

deep learning

generative adversarial networks

hydrological datasets

imagery synthesis

river image

Details

Title: Subtitle: Realistic River Image Synthesis Using Deep Generative Adversarial Networks
Creators: Akshat Gautam - School of Engineering and Applied Sciences, Columbia University, New York, NY, United States
Muhammed Sit - Interdisciplinary Graduate Program in Informatics, University of Iowa, Iowa City, IA, United States
Ibrahim Demir - Department of Civil and Environmental Engineering, University of Iowa, Iowa City, IA, United States
Resource Type: Journal article
Publication Details: Frontiers in water, Vol.4, 784441
DOI: 10.3389/frwa.2022.784441
eISSN: 2624-9375
Publisher: Frontiers Media S.A
Language: English
Date published: 02/01/2022
Academic Unit: Electrical and Computer Engineering; Civil and Environmental Engineering; IIHR--Hydroscience and Engineering; Injury Prevention Research Center
Record Identifier: 9984222758902771

Metrics

123 Record Views

25 Times Cited - Web of Science