V-FloodNet: A video segmentation system for urban flood detection and quantification

Article Computer Science, Interdisciplinary Applications

Application of image processing and convolutional neural networks for flood image classification and semantic segmentation

R. J. Pally et al.

Summary: Researchers leverage deep learning algorithms, specifically convolutional neural networks, to analyze and detect flood objects in flood images. They developed a python package called FloodImageClassifier that can perform object detection, segmentation, flood water level estimation, and classification. This technology can be integrated with existing cameras to monitor flood conditions and provide early intelligence to emergency response authorities.

ENVIRONMENTAL MODELLING & SOFTWARE (2022)

Add to Collection

Article Computer Science, Interdisciplinary Applications

ATLANTIS: A benchmark for semantic segmentation of waterbody images

Seyed Mohammad Hassan Erfani et al.

Summary: This article introduces a new benchmark for semantic segmentation of waterbodies and related objects, ATLANTIS, which includes a large number of images and pixel-level manual annotations. Researchers also developed a novel deep neural network, AQUANet, demonstrating its superiority over other state-of-the-art semantic segmentation networks on the ATLANTIS dataset.

ENVIRONMENTAL MODELLING & SOFTWARE (2022)

Add to Collection

Article Computer Science, Interdisciplinary Applications

A deep learning model for predicting river flood depth and extent

Hossein Hosseiny

Summary: This study introduces an innovative DL framework named U-NetRiver, which can automatically identify river geometry, flood extent, and predict river flooding depth. The model shows promising accuracy in predicting maximum flood depth in rivers.

ENVIRONMENTAL MODELLING & SOFTWARE (2021)

Add to Collection

Article Computer Science, Interdisciplinary Applications

Deep learning-enhanced extraction of drainage networks from digital elevation models

Xin Mao et al.

Summary: The study introduced distributed representations of aspect features, adopted the U-Net model and postprocessing method to improve the accuracy and efficiency of deep learning models in drainage network extraction.

ENVIRONMENTAL MODELLING & SOFTWARE (2021)

Add to Collection

Article Computer Science, Interdisciplinary Applications

Real-time water level monitoring using live cameras and computer vision techniques

Navid H. Jafari et al.

Summary: The study aimed to develop technology for accurate and timely flood hydrographs using big data generated from videos and images, predicting water levels from images by leveraging object-based image analysis and deep learning techniques.

COMPUTERS & GEOSCIENCES (2021)

Add to Collection

Article Computer Science, Interdisciplinary Applications

Computer Vision Based Estimation of Flood Depth in Flooded-Vehicle Images

Somin Park et al.

Summary: This study introduces a vision-based method for flood depth estimation using flooded-vehicle images, consisting of vehicle object segmentation, image retrieval, and depth estimation processes. The method has been validated with 500 flooding images, showing promising results in terms of accuracy and effectiveness.

JOURNAL OF COMPUTING IN CIVIL ENGINEERING (2021)

Add to Collection

Article Computer Science, Interdisciplinary Applications

An attention U-Net model for detection of fine-scale hydrologic streamlines

Zewei Xu et al.

Summary: This study introduces an attention U-net model using high-accuracy lidar data for hydrologic streamline detection, showing higher accuracy and smoothness compared to traditional machine learning methods. These results suggest that deep learning approaches can effectively utilize fine-scale high-accuracy data for hydrologic studies.

ENVIRONMENTAL MODELLING & SOFTWARE (2021)

Add to Collection

Article Computer Science, Information Systems