HED-UNet: Combined Segmentation and Edge Detection for Monitoring the Antarctic Coastline

@article{Heidler2022HEDUNetCS,
  title={HED-UNet: Combined Segmentation and Edge Detection for Monitoring the Antarctic Coastline},
  author={Konrad Heidler and Lichao Mou and Celia A. Baumhoer and Andreas J. Dietz and Xiaoxiang Zhu},
  journal={IEEE Transactions on Geoscience and Remote Sensing},
  year={2022},
  volume={60},
  pages={1-14}
}
Deep learning-based coastline detection algorithms have begun to outshine traditional statistical methods in recent years. However, they are usually trained only as single-purpose models to either segment land and water or delineate the coastline. In contrast to this, a human annotator will usually keep a mental map of both segmentation and delineation when performing manual coastline detection. To take into account this task duality, we, therefore, devise a new model to unite these two… 

Figures and Tables from this paper

Localization and Edge-Based Segmentation of Lumbar Spine Vertebrae to Identify the Deformities Using Deep Learning Models

TLDR
The methods that will help clinicians to grade the severity of the disease with confidence are presented, as the current manual diagnosis by different doctors has dissimilarity and variations in the analysis of diseases.

Calving fronts and where to find them: a benchmark dataset and methodology for automatic glacier calving front extraction from synthetic aperture radar imagery

TLDR
A benchmark dataset of SAR images from multiple regions of the globe with corresponding manually defined labels providing information on the position of the calving front, enabling comparability between different front detection algorithms and improving the reproducibility of front detection studies is presented.

Decision Calibration Network for Semantic Labeling of High-Resolution Remote Sensing Images

TLDR
A dual-decoder semantic labeling neural network based on the atrous spatial pyramid pooling module and attention mechanism to achieve the high-precision classification of different ground objects and a decision calibration auxiliary loss is proposed to improve the models’s ability to classify examples of highly ambiguous output by different decoders.

A C OMPREHENSIVE R EVIEW ON D EEP S UPERVISION : T HEORIES AND A PPLICATIONS

TLDR
A comprehensive in-depth review of deep supervision in both theories and applications is provided, which proposes a new classification of different deep supervision networks, and discusses advantages and limitations of currentDeep supervision networks in computer vision applications.

A Comprehensive Review on Deep Supervision: Theories and Applications

TLDR
A comprehensive in-depth review of deep supervision in both theories and applications is provided, which proposes a new classification of different deep supervision networks, and discusses advantages and limitations of currentDeep supervision networks in computer vision applications.

Calving Fronts and Where to Find Them: A Benchmark Dataset and Methodology for Automatic Glacier Calving Front Extraction from SAR Imagery

TLDR
This paper presents a benchmark dataset of SAR images from multiple regions of the globe with corresponding manually defined labels to train and test approaches for the detection of glacier calving fronts, and is the first dataset to provide long-term glacier calve front information from multi-mission data.

Dam Reservoir Extraction From Remote Sensing Imagery Using Tailored Metric Learning Strategies

Dam reservoirs play an important role in meeting sustainable development goals (SDGs) and global climate targets. However, particularly for small dam reservoirs, there is a lack of consistent data on

Physics-aware deep edge detection network

In this paper, we describe an effort to build a new deep edge detection method designed to detect weather-related phenomena such as clouds and planetary boundary layer heights present in backscatter

Hed-Unet: A Multi-Scale Framework for Simultaneous Segmentation and Edge Detection

TLDR
The proposed network architecture follows the successful encoder-decoder approach, and is improved by employing deep supervision at multiple resolution levels, as well as merging these resolution levels into a final prediction using a hierarchical attention mechanism.

References

SHOWING 1-10 OF 72 REFERENCES

Gated-SCNN: Gated Shape CNNs for Semantic Segmentation

TLDR
A new two-stream CNN architecture for semantic segmentation that explicitly wires shape information as a separate processing branch, i.e. shape stream, that processes information in parallel to the classical stream is proposed.

Coastline Detection And Tracing In SAr Images

Holistically-Nested Edge Detection

  • Saining XieZ. Tu
  • Computer Science
    2015 IEEE International Conference on Computer Vision (ICCV)
  • 2015
TLDR
HED performs image-to-image prediction by means of a deep learning model that leverages fully convolutional neural networks and deeply-supervised nets, and automatically learns rich hierarchical representations that are important in order to resolve the challenging ambiguity in edge and object boundary detection.

A Complete High-Resolution Coastline of Antarctica Extracted from Orthorectified Radarsat SAR Imagery

A complete, high-resolution coastline of Antarctica, extracted from an orthorectified mosaic of Radarsat-1 synthetic aperture radar (SAR) images through a sequence of automated image processing

U-Net: Convolutional Networks for Biomedical Image Segmentation

TLDR
It is shown that such a network can be trained end-to-end from very few images and outperforms the prior best method (a sliding-window convolutional network) on the ISBI challenge for segmentation of neuronal structures in electron microscopic stacks.

Relation Matters: Relational Context-Aware Fully Convolutional Network for Semantic Segmentation of High-Resolution Aerial Images

TLDR
Two simple yet effective network units are introduced, the spatial relation module, and the channel relation module to learn and reason about global relationships between any two spatial positions or feature maps, and then produce Relation-Augmented (RA) feature representations.

Bathymetry Time Series Using High Spatial Resolution Satellite Images

The use of the new generation of remote sensors, such as echo sounders and Global Navigation Satellite System (GNSS) receivers with differential correction installed in a drone, allows the

Semantic segmentation network combined with edge detection for building extraction in remote sensing images

TLDR
An end-to-end convolutional neural network that combines semantic segmentation and edge detection for building extraction and a residual unit combined with spatial pyramid pooling (SPP-RU) to yield representations of different size receptive fields by multi-branch network are designed.

Building Footprint Generation by Integrating Convolution Neural Network With Feature Pairwise Conditional Random Field (FPCRF)

TLDR
It is found that the proposed end-to-end building footprint generation framework with the FPCRF as the graph model can further improve the accuracy of building footprintgeneration by using only CNN, which is the current state of the art.

Sea-Land Segmentation With Res-UNet And Fully Connected CRF

TLDR
A method based on UNet for sea-land segmentation is proposed, replacing its contraction part with ResNet which specializes in handling complicated scenes, and constructing a new network structure Res-UNet, which is based on Conditional Random Field model.
...