Extendibility of a Thin-Cloud Removal Algorithm to Hi-Resolution Visible Bands of Sentinel-2 Data

Gao, Yue; Wang, Yong; Lv, Haitao

doi:10.1109/igarss.2018.8518878

“…Various studies [22,[26][27][28][29][30] demonstrate the use of GANs-generated data in the training process of deep learning models to improve their performance in case of insufficient training data. GANs have been utilized in the fields of remote sensing not only to generate missing data [31,32], but also to translate data among various domains [23,[33][34][35][36][37][38]. However, in applications that require control over the data translation, conditional GANs by Mirza et al (2014) is the suitable framework to allow the generation of data conditioned by auxiliary information such as text, tags, class labels, images, etc.…”

Section: Introductionmentioning

confidence: 99%

Generation Of Dense Urban Features Using Conditional GAN

Nihal,

Kumar

2023

Preprint

0

View full text Add to dashboard Cite

This paper discusses the use of conditional Generative Adversarial Networks (GANs) to generate dense urban features in satellite images and evaluate their effectiveness in semantic segmentation tasks. High-resolution true-color satellite imagery of Mumbai, obtained from Pleiades-1A at a 0.5m resolution, is utilized for the study. The proposed Multiple Discriminator pix2pix (MD-pix2pix) model, which employs multiple discriminators and a modified training procedure, is introduced to generate realistic satellite images. The performance of the MD-pix2pix model is compared to the traditional pix2pix model using a mix dataset of real and generated satellite images. The synthesized satellite images are assessed for their effectiveness in semantic segmentation tasks using various CNN models, including VGG16-UNet, MobileNetV2-UNet, and DeepLabV3+. The study aims to overcome the limitations of existing datasets that do not include informal settlements, such as slum areas, which are common in many cities. The results indicate that the MD-pix2pix model produces more realistic satellite images with greater variability in vegetation type, slum arrangements, and built-ups than the traditional pix2pix model. The synthetically generated satellite images are also effective in semantic segmentation tasks, with better segmentation accuracy achieved using the MD-pix2pix generated images for computationally less expensive architectures such as MobileNetV2-UNet. This study highlights the potential of GANs to generate realistic satellite images for urban feature mapping and monitoring applications.

show abstract

“…Various studies (Kim & Hwang, 2022;Lv et al, 2020;Nabati, Navidan, Shahbazian, Ghorashi, & Windridge, 2020;Park, Tran, Jung, & Park, 2020;Rashid, Tanveer, & Aqeel Khan, 2019;Sedigh, Sadeghian, & Masouleh, 2019) demonstrate the use of GANs-generated data in the training process of deep learning models to improve their performance in case of insufficient training data. GANs have been utilized in the fields of remote sensing not only to generate missing data (Panchal et al, 2021;Shao, Wang, Zuo, & Meng, 2022), but also to translate data among various domains (Bermudez et al, 2019;Bermudez, Happ, Oliveira, & Feitosa, 2018;Enomoto et al, 2017Enomoto et al, , 2018Gao, Wang, & Lv, 2018;Ley, Dhondt, Valade, Haensch, & Hellwich, 2018;Singh & Komodakis, 2018). However, in applications that require control over the data translation, conditional GANs by Mirza and Osindero (2014) is the suitable framework to allow the generation of data conditioned by auxiliary information such as text, tags, class labels, images, etc.…”

Section: Introductionmentioning

confidence: 99%

Generation Of Dense Urban Features Using Conditional GAN

Nihal,

Kumar

2023

Preprint

0

View full text Add to dashboard Cite

This paper discusses the use of conditional Generative Adversarial Networks (GANs) to generate dense urban features in satellite images and evaluate their effectiveness in semantic segmentation tasks. High-resolution true-color satellite imagery of Mumbai, obtained from Pleiades-1A at a 0.5m resolution, is utilized for the study. The proposed Multiple Discriminator pix2pix (MD-pix2pix) model, which employs multiple discriminators and a modified training procedure, is introduced to generate realistic satellite images. The performance of the MD-pix2pix model is compared to the traditional pix2pix model using a mix dataset of real and generated satellite images. The synthesized satellite images are assessed for their effectiveness in semantic segmentation tasks using various CNN models, including VGG16-UNet, MobileNetV2-UNet, and DeepLabV3+. The study aims to overcome the limitations of existing datasets that do not include informal settlements, such as slum areas, which are common in many cities. The results indicate that the MD-pix2pix model produces more realistic satellite images with greater variability in vegetation type, slum arrangements, and built-ups than the traditional pix2pix model. The synthetically generated satellite images are also effective in semantic segmentation tasks, with better segmentation accuracy achieved using the MD-pix2pix generated images for computationally less expensive architectures such as MobileNetV2-UNet. This study highlights the potential of GANs to generate realistic satellite images for urban feature mapping and monitoring applications.

show abstract

“…This ensures that the thinner clouds are also removed. The authors of [4] extend the radiative transfer model to Sentinel-2 images to remove thin clouds. In [5], the authors use a convolutional neural network with multi-scale prediction scheme to remove the clouds from Sentinel-2 images.…”

Section: Introductionmentioning

confidence: 99%

Estimating NDVI from SAR Images Using DNN

Calota

¹

,

Faur

²

,

Datcu

³

2022

IGARSS 2022 - 2022 IEEE International Geoscience and Remote Sensing Symposium

View full text Add to dashboard Cite

The Normalized Difference Vegetation Index (NDVI) is an important factor to be considered in vegetation tracking and analysis, which can be easily derived from multispectral (MS) images. However, the limitation imposed by the atmospheric conditions makes the calculation of this index difficult. Because of the clouds, only a limited number of multispectral bands can capture the land appropriately. Furthermore, the multispectral sensors are dependent on the sunlight, which makes the acquisition of data more limited. These limitations do not hinder other types of Earth Observation (EO) data, like the scenes captured by the Synthetic Aperture Radar (SAR). However, SAR images cannot be used in NDVI calculation. In this article, we propose a deep learning (DL) based method for NDVI estimation from SAR data. Using a database with corresponding MS and SAR patches, we calculate the NDVI for each sample, then use a convolutional neural network (CNN) for predicting the NDVI of SAR images. This simple method leads to a precision of 70% in NDVI estimation from SAR images.

show abstract

Extendibility of a Thin-Cloud Removal Algorithm to Hi-Resolution Visible Bands of Sentinel-2 Data

Cited by 5 publications

References 9 publications

Generation Of Dense Urban Features Using Conditional GAN

Generation Of Dense Urban Features Using Conditional GAN

Generation Of Dense Urban Features Using Conditional GAN

Estimating NDVI from SAR Images Using DNN

Contact Info

Product

Resources

About