Skip to main content
U.S. Department of Energy
Office of Scientific and Technical Information

Coarse-to-fine Task-driven Inpainting for Geoscience Images

Journal Article · · IEEE Transactions on Circuits and Systems for Video Technology
 [1];  [1];  [2];  [3];  [4];  [5];  [1]
  1. Cleveland State Univ., Cleveland, OH (United States)
  2. Agency for Science, Technology and Research (A*STAR), Singapore. Centre for Frontier AI Research (CFAR)
  3. Wuhan Univ. (China)
  4. Beijing University of Technology (China)
  5. Brookhaven National Laboratory (BNL), Upton, NY (United States)
The processing and recognition of geoscience images have wide applications. Most of existing researches focus on understanding the high-quality geoscience images by assuming that all the images are clear. However, in many real-world cases, the geoscience images might contain occlusions during the image acquisition. This problem actually implies the image inpainting problem in computer vision and multimedia. As far as we know, all the existing image inpainting algorithms learn to repair the occluded regions for a better visualization quality, they are excellent for natural images but not good enough for geoscience images, and they never consider the following geoscience task when developing inpainting methods. Here, this paper aims to repair the occluded regions for a better geoscience task performance and advanced visualization quality simultaneously, without changing the current deployed deep learning based geoscience models. Because of the complex context of geoscience images, we propose a coarse-to-fine encoder-decoder network with the help of designed coarse-to-fine adversarial context discriminators to reconstruct the occluded image regions. Due to the limited data of geoscience images, we propose a MaskMix based data augmentation method, which augments inpainting masks instead of augmenting original images, to exploit the limited geoscience image data. The experimental results on three public geoscience datasets for remote sensing scene recognition, cross-view geolocation and semantic segmentation tasks respectively show the effectiveness and accuracy of the proposed method. The code is available at: https://github.com/HMS97/Task-driven-Inpainting.
Research Organization:
Brookhaven National Laboratory (BNL), Upton, NY (United States)
Sponsoring Organization:
National Science Foundation (NSF); USDOE Office of Science (SC), Advanced Scientific Computing Research (ASCR)
Grant/Contract Number:
SC0012704
OSTI ID:
1993157
Report Number(s):
BNL-224638-2023-JAAM
Journal Information:
IEEE Transactions on Circuits and Systems for Video Technology, Journal Name: IEEE Transactions on Circuits and Systems for Video Technology Journal Issue: 12 Vol. 33; ISSN 1051-8215
Publisher:
IEEECopyright Statement
Country of Publication:
United States
Language:
English

References (62)

Image Inpainting for Irregular Holes Using Partial Convolutions book January 2018
VCNet: A Robust Approach to Blind Image Inpainting book January 2020
Automatic pancreas segmentation based on lightweight DCNN modules and spatial prior propagation journal June 2021
Context Encoders: Feature Learning by Inpainting conference June 2016
Deep Residual Learning for Image Recognition conference June 2016
Image-to-Image Translation with Conditional Adversarial Networks conference July 2017
The Unreasonable Effectiveness of Deep Features as a Perceptual Metric conference June 2018
DOTA: A Large-Scale Dataset for Object Detection in Aerial Images conference June 2018
Generative Image Inpainting with Contextual Attention conference June 2018
Learning RoI Transformer for Oriented Object Detection in Aerial Images conference June 2019
BASNet: Boundary-Aware Salient Object Detection conference June 2019
Contextual Residual Aggregation for Ultra High-Resolution Image Inpainting conference June 2020
Recurrent Feature Reasoning for Image Inpainting conference June 2020
PD-GAN: Probabilistic Diverse GAN for Image Inpainting conference June 2021
MAT: Mask-Aware Transformer for Large Hole Image Inpainting conference June 2022
Coherent Semantic Attention for Image Inpainting conference October 2019
Dense-Add Net: An Novel Convolutional Neural Network for Remote Sensing Image Inpainting conference July 2018
Remote Sensing Image Inpainting with Generative Adversarial Networks conference June 2020
Remote Sensing Image Scene Classification: Benchmark and State of the Art journal October 2017
Multilevel Cloud Detection in Remote Sensing Images Based on Deep Learning journal August 2017
Deep Learning Based Feature Selection for Remote Sensing Scene Classification journal November 2015
Inpainting of Remote Sensing SST Images With Deep Convolutional Generative Adversarial Network journal February 2019
Domain Adaptation for Convolutional Neural Networks-Based Remote Sensing Scene Classification journal August 2019
Denoising-Based Multiscale Feature Fusion for Remote Sensing Image Captioning journal March 2021
Object Detection Under Rainy Conditions for Autonomous Vehicles: A Review of State-of-the-Art and Emerging Techniques journal January 2021
Patch-Sparsity-Based Image Inpainting Through a Facet Deduced Directional Derivative journal May 2019
Low Rank Component Induced Spatial-Spectral Kernel Method for Hyperspectral Image Classification journal October 2020
Multiscale Densely-Connected Fusion Networks for Hyperspectral Images Classification journal January 2021
E2I: Generative Inpainting From Edge to Image journal April 2021
Structure-Guided Deep Video Inpainting journal August 2021
Each Part Matters: Local Patterns Facilitate Cross-View Geo-Localization journal February 2022
IID-Net: Image Inpainting Detection Network via Neural Architecture Search and Attention journal March 2022
Global-Local Balanced Low-Rank Approximation of Hyperspectral Images for Classification journal April 2022
Dual-Pyramidal Image Inpainting With Dynamic Normalization journal September 2022
Deep Image Inpainting With Enhanced Normalization and Contextual Attention journal October 2022
Reference-Guided Landmark Image Inpainting With Deep Feature Matching journal December 2022
A MAP-Based Algorithm for Destriping and Inpainting of Remotely Sensed Images journal May 2009
Inpainting for Remotely Sensed Images With a Multichannel Nonlocal Total Variation Model journal January 2014
Unsupervised Deep Slow Feature Analysis for Change Detection in Multi-Temporal Remote Sensing Images journal December 2019
Truncation Cross Entropy Loss for Remote Sensing Image Captioning journal June 2021
AFNet: Adaptive Fusion Network for Remote Sensing Image Semantic Segmentation journal September 2021
VD-GAN: A Unified Framework for Joint Prototype and Representation Learning From Contaminated Single Sample per Person journal January 2021
An adaptive semiparametric and context-based approach to unsupervised change detection in multitemporal remote-sensing images journal April 2002
Image Quality Assessment: From Error Visibility to Structural Similarity journal April 2004
No-Reference Image Quality Assessment in the Spatial Domain journal December 2012
Polarimetric SAR Image Semantic Segmentation With 3D Discrete Wavelet Transform and Markov Random Field journal January 2020
X-Invariant Contrastive Augmentation and Representation Learning for Semi-Supervised Skeleton-Based Action Recognition journal January 2022
Salient Band Selection for Hyperspectral Image Classification via Manifold Ranking journal June 2016
DisP+V: A Unified Framework for Disentangling Prototype and Variation From Single Sample per Person journal February 2023
SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation journal December 2017
Places: A 10 Million Image Database for Scene Recognition journal June 2018
Effects of Image Degradation and Degradation Removal to CNN-Based Image Classification journal April 2021
Deep High-Resolution Representation Learning for Visual Recognition journal October 2021
Evaluation of Image Inpainting for Classification and Retrieval conference March 2020
Cross-View Image Matching for Geo-Localization in Urban Environments conference July 2017
Predicting Ground-Level Scene Layout from Aerial Imagery conference July 2017
MISF:Multi-level Interactive Siamese Filtering for High-Fidelity Image Inpainting conference June 2022
JPGNet: Joint Predictive Filtering and Generative Network for Image Inpainting conference October 2021
R³Net: Recurrent Residual Refinement Network for Saliency Detection conference July 2018
Context-Aware Image Inpainting with Learned Semantic Priors conference August 2021
Cloud Extraction from Chinese High Resolution Satellite Imagery by Probabilistic Latent Semantic Analysis and Object-Based Machine Learning journal November 2016
LoveDA: A Remote Sensing Land-Cover Dataset for Domain Adaptive Semantic Segmentation dataset January 2021

Similar Records

Inpainting with sparse linear combinations of exemplars
Journal Article · Mon Dec 31 23:00:00 EST 2007 · OSTI ID:960615

Load Profile Inpainting for Missing Load Data Restoration and Baseline Estimation
Journal Article · Thu Jul 06 20:00:00 EDT 2023 · IEEE Transactions on Smart Grid · OSTI ID:2329385