Cloud Removal from Satellite Images

Chauhan, Rutvik; Singh, Antarpuneet; Saha, Sujoy Kumar

doi:10.48550/arxiv.2112.15483

Cited by 1 publication

(4 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Regarding the quantitative evaluation metrics, the works by [2,4,6,7,20] also adopt the indicators structural similarity (SSIM), square root of the average relative global error (ERGAS), root mean squares error (RMSE), normalized root mean squares error (NRMSE), spectral angle mapper (SAM) and universal image quality index (UIQI) to evaluate the performance of satellite image cloud removal methods. The work by [21] highlights the usefulness of the SSIM indicator in the phenoms related to the visual quality of the object's structural details.…”

Section: Related Workmentioning

confidence: 99%

“…The Acre state, localized in the northern region of Brazil, was chosen as the study area for this research. According to the Brazilian Institute of Geography and Statistics (IBGE (https://cidades.ibge.gov.br/brasil/ac/panorama, accessed on 13 October 2023)), Acre has 830,026 inhabitants, with a population density of 5.06 inhabitants per km 2 , with a human development index (HDI) of 0.71, an urbanized area of 216.14 km 2 and a land area of 164,173.429 km 2 . Figure 6 shows the study area.…”

Section: Study Areamentioning

confidence: 99%

“…The images available for remote sensing are mostly observed from a distance and are affected by atmospheric conditions and climatic factors, such as clouds [1]. Cloud cover alters the useful information in images, which is an adversity for subsequent image interpretation, as it can weaken or even cause the loss of essential information for analysis [2].…”

Section: Introductionmentioning

confidence: 99%

“…Similarly, ref. [2] uses GANs to remove clouds from Sentinel images captured on the Google Earth platform, employing spatial-based methods. In [5], they apply spatial-based methods using convolutional neural networks (CNNs); the network is structured in two stages, one to extract cloud transparency information and the other to retrieve information from the ground surface.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

A Cloud Coverage Image Reconstruction Approach for Remote Sensing of Temperature and Vegetation in Amazon Rainforest

Bezerra,

Mafalda,

Alvarez

et al. 2023

Applied Sciences

View full text Add to dashboard Cite

Remote sensing involves actions to obtain information about an area located on Earth. In the Amazon region, the presence of clouds is a common occurrence, and the visualization of important terrestrial information in the image, like vegetation and temperature, can be difficult. In order to estimate land surface temperature (LST) and the normalized difference vegetation index (NDVI) from satellite images with cloud coverage, the inpainting approach will be applied to remove clouds and restore the image of the removed region. This paper proposes the use of the neural network LaMa (large mask inpainting) and the scalable model named Big LaMa for the automatic reconstruction process in satellite images. Experiments are conducted on Landsat-8 satellite images of the Amazon rainforest in the state of Acre, Brazil. To evaluate the architecture’s accuracy, the RMSE (root mean squared error), SSIM (structural similarity index) and PSNR (peak signal-to-noise ratio) metrics were used. The LST and NDVI of the reconstructed image were calculated and compared qualitatively and quantitatively, using scatter plots and the chosen metrics, respectively. The experimental results show that the Big LaMa architecture performs more effectively and robustly in restoring images in terms of visual quality. And the LaMa network shows minimal superiority for the measured metrics when addressing medium marked areas. When comparing the results achieved in NDVI and LST of the reconstructed images with real cloud coverage, great visual results were obtained with Big LaMa.

show abstract