Inpainting Strategies for Reconstruction of Missing Data in VHR Images

  title={Inpainting Strategies for Reconstruction of Missing Data in VHR Images},
  author={Luca Lorenzi and Farid Melgani and Gr{\'e}goire Mercier},
  journal={IEEE Geoscience and Remote Sensing Letters},
Missing data in very high spatial resolution (VHR) optical imagery take origin mainly from the acquisition conditions. Their accurate reconstruction represents a great methodological challenge because of the complexity and the ill-posed nature of the problem. In this letter, we present three different solutions, with all based on the inpainting approach, which consists in reconstructing the missing regions in a given image by propagating the spectrogeometrical information retrieved from the… 

Figures and Tables from this paper

Orthogonal matching pursuit for VHR image reconstruction

A new method for the reconstruction of areas obscured by clouds is proposed, based on compressive sensing (CS) theory, which allows to find sparse signal representations in underdetermined linear equation systems.

Contextual genetic algorithm for compressive sensing reconstruction of VHR images

A novel implementation which exploits genetic algorithms (GAs) and a new strategy for the selection of atoms belonging to the dictionary is proposed based on compressive sensing theory, which allows to find sparse signal representations in underdetermined linear equation systems.

Inpainting for Remotely Sensed Images With a Multichannel Nonlocal Total Variation Model

An effective image inpainting technology is presented to solve this task, based on multichannel nonlocal total variation, which takes advantage of a nonlocal method, which has a superior performance in dealing with textured images and reconstructing large-scale areas.


The presence of shadows in Remote Sensing images leads to misinterpretation of objects in several real-world applications, which includes Very High Resolution (VHR) image data from urban areas.

Image inpainting: theoretical analysis and comparison of algorithms

Analyzing this method using the concept of clustered sparsity leads to theoretical bounds and results, which will be presented and necessary conditions for the frames to lead to sufficiently good solutions to show that shearlets are able to inpaint larger gaps than wavelets.

Image Enhancement and Restoration by Image Inpainting

This paper presents efficient algorithm that combines the advantages of structural, texture and exemplar techniques and proposes a best-first algorithm in which the confidence in the synthesized pixel values is propagated in a manner similar to the propagation of information in inpainting.

Support vector regression with kernel combination for missing data reconstruction

New combinations of kernel functions with which a better reconstruction of missing data due to the presence of clouds is obtained, by adding to the radiometric information, i.e., the position information of the pixels in the image.


An automated methodology for both detection and recovery of shadows is proposed, which attains a high performance and substantial gain in recovering anthropic targets such as streets, roofs and buildings initially damaged by shadows.



Contextual Spatiospectral Postreconstruction of Cloud-Contaminated Images

This letter presents a postreconstruction methodology for improving the contextual reconstruction process by opportunely capturing spatial and spectral correlations characterizing the considered image and proposes a solution to a problem that has not been addressed in the remote sensing literature, i.e., the generation of an error map beside the reconstructed images to provide end-users with helpful indications about reconstruction reliability.

A Bandelet-Based Inpainting Technique for Clouds Removal From Remotely Sensed Images

An efficient inpainting technique for the reconstruction of areas obscured by clouds or cloud shadows in remotely sensed images is presented, based on the Bandelet transform and the multiscale geometrical grouping.

Region filling and object removal by exemplar-based image inpainting

The simultaneous propagation of texture and structure information is achieved by a single, efficient algorithm that combines the advantages of two approaches: exemplar-based texture synthesis and block-based sampling process.

Image inpainting

A novel algorithm for digital inpainting of still images that attempts to replicate the basic techniques used by professional restorators, and does not require the user to specify where the novel information comes from.

Block-based image inpainting in the wavelet domain

Experimental results indicate that the proposed algorithm can fill large inPainting regions with good visual quality, presenting results comparable to or better than other competitive approaches for image inpainting.

Image inpainting using wavelet-based inter- and intra-scale dependency

  • D. ChoT. Bui
  • Computer Science
    2008 19th International Conference on Pattern Recognition
  • 2008
This work proposes to utilize the advantages of wavelet transforms for image inpainting, and can expect better global structure estimation of a damaged region in addition to shape and texture properties.

On pixel-based texture synthesis by non-parametric sampling

Contextual reconstruction of cloud-contaminated multitemporal multispectral images

  • F. Melgani
  • Environmental Science, Mathematics
    IEEE Transactions on Geoscience and Remote Sensing
  • 2006
The proposed methods for the reconstruction of areas obscured by clouds in a sequence of multitemporal multispectral images show a clear superiority, which makes them a promising and useful tool in solving the considered problem, whose great complexity is commensurate with its practical importance.

Clouds and cloud shadows removal from high-resolution remote sensing images

This paper proposed an improved fast fragment-based image completion technique to accomplish this hard work automatically based on Drori’s work and can reduce computing time remarkably with almost identical results.

Multiresolution Remote Sensing Image Clustering

The goal of this multiresolution image method is to automatically build a classification using knowledge extracted from both images, by unsupervised way and without preprocessing image fusion.