Centre de documentation
scientifique

A high-resolution panchromatic-multispectral satellite image fusion method assisted with building segmentation / Fang Gao in Computers & geosciences, vol 168 (November 2022)  

Public [article]  inComputers & geosciences > vol 168 (November 2022) . - n° 105219 Titre : A high-resolution panchromatic-multispectral satellite image fusion method assisted with building segmentation Type de document : Article/Communication Auteurs : Fang Gao, Auteur ; Yihui Li, Auteur ; Peng Zhang, Auteur ; et al., Auteur Année de publication : 2022 Article en page(s) : n° 105219 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique [Termes IGN] bâtiment [Termes IGN] filtre de Gauss [Termes IGN] fusion d'images [Termes IGN] image à haute résolution [Termes IGN] image Jilin [Termes IGN] image multibande [Termes IGN] image panchromatique [Termes IGN] image satellite [Termes IGN] lissage de données Résumé : (auteur) The main difficulty of panchromatic-multispectral image fusion is to balance the quality of spatial information and the spectral fidelity. Most of the practical fusion methods determine the optimal parameters based on the spatial and spectral characteristics of all original panchromatic and multispectral bands. However, for built-up and non-built-up areas (like cropland, forest) in one image, there may be large differences in their spatial and spectral characteristics, so their fused results are not optimal respectively with same parameters. To address above issues, this paper presents a high-resolution satellite image fusion method assisted with building segmentation. First, the proposed approach computes the average gradient and Gaussian filtering parameters of built-up and non-built-up areas separately according to the building segmentation results, on the basis of smoothing filter-based intensity modulation (SFIM). Then the intermediate data of two types of areas are computed in parallel and they are composited to obtain the final fused image, weighted by the pixel-wise “building factors” derived from the building segmentation results. Moreover, to better simulate the spatial characteristics of the multispectral image, we perform the “gradient simulation” operation to extract the gradient values in the multispectral image. Experimental results on Jilin-1 satellite images show that the proposed method provides competitive performance in spatial resolution, multispectral fidelity and quantity of information, as compared to the state-of-the-art methods in mainstream commercial software. Numéro de notice : A2022-721 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article DOI : 10.1016/j.cageo.2022.105219 Date de publication en ligne : 11/09/2022 En ligne : https://doi.org/10.1016/j.cageo.2022.105219 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=101657 [article]

Comparison between Gaussian and decorrelation filters of GRACE-based RL05 temporal gravity solutions over Egypt / Basem Elsaka in Survey review, vol 54 n° 384 (May 2022)  

Public [article]  inSurvey review > vol 54 n° 384 (May 2022) . - pp 233 - 242 Titre : Comparison between Gaussian and decorrelation filters of GRACE-based RL05 temporal gravity solutions over Egypt Type de document : Article/Communication Auteurs : Basem Elsaka, Auteur ; Mohamed El-Ashquer, Auteur Année de publication : 2022 Article en page(s) : pp 233 - 242 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Géodésie physique [Termes IGN] analyse comparative [Termes IGN] champ de pesanteur local [Termes IGN] décorrélation [Termes IGN] données GRACE [Termes IGN] Egypte [Termes IGN] filtre de Gauss Résumé : (auteur) This contribution provides a comparison between the Gaussian and decorrelation filters as derived from GRACE products (RL05) estimated by the official GRACE Science Data System centres (GFZ, CSR and JPL) as well as the ITSG-GRACE2016 solutions over Egypt. The outcome of this study will help in finding out which of these centres provides improved temporal gravity solutions as well as the most promising GRACE time series over Egypt. The obtained results regarding Gaussian filters show that the GFZ centre provides the most promising solutions w.r.t. CSR and JPL. Whereas the ITSG-GRACE2016 products provide improvements, especially at Gaussian radius 200 km, of about 56%, 68% and 60% w.r.t. CSR, JPL and GFZ solutions, respectively. Regarding the decorrelation filtering, the ITSG-GRACE2016 provides the least Std. w.r.t. CSR, JPL and GFZ solutions showing for the DDK8 improvement of about 48%, 64% and 68% w.r.t. the three centres GFZ, JPL and CSR, respectively. Numéro de notice : A2022-355 Affiliation des auteurs : non IGN Thématique : POSITIONNEMENT Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1080/00396265.2021.1919841 Date de publication en ligne : 04/05/2021 En ligne : https://doi.org/10.1080/00396265.2021.1919841 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=100553 [article]

Pedestrian trajectory prediction with convolutional neural networks / Simone Zamboni in Pattern recognition, vol 121 (January 2022)  

Public [article]  inPattern recognition > vol 121 (January 2022) . - n° 108252 Titre : Pedestrian trajectory prediction with convolutional neural networks Type de document : Article/Communication Auteurs : Simone Zamboni, Auteur ; Zekarias Tilahun Kefato, Auteur ; Sarunas Girdzijauskas, Auteur ; et al., Auteur Année de publication : 2022 Article en page(s) : n° 108252 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Analyse spatiale [Termes IGN] classification par réseau neuronal convolutif [Termes IGN] classification par réseau neuronal récurrent [Termes IGN] distance euclidienne [Termes IGN] filtre de Gauss [Termes IGN] itinéraire piétionnier [Termes IGN] modèle de simulation [Termes IGN] navigation pédestre [Termes IGN] piéton [Termes IGN] prévision à court terme [Termes IGN] réseau social [Termes IGN] trajet (mobilité) Résumé : (auteur) Predicting the future trajectories of pedestrians is a challenging problem that has a range of application, from crowd surveillance to autonomous driving. In literature, methods to approach pedestrian trajectory prediction have evolved, transitioning from physics-based models to data-driven models based on recurrent neural networks. In this work, we propose a new approach to pedestrian trajectory prediction, with the introduction of a novel 2D convolutional model. This new model outperforms recurrent models, and it achieves state-of-the-art results on the ETH and TrajNet datasets. We also present an effective system to represent pedestrian positions and powerful data augmentation techniques, such as the addition of Gaussian noise and the use of random rotations, which can be applied to any model. As an additional exploratory analysis, we present experimental results on the inclusion of occupancy methods to model social information, which empirically show that these methods are ineffective in capturing social interaction. Numéro de notice : A2022-109 Affiliation des auteurs : non IGN Thématique : GEOMATIQUE Nature : Article DOI : 10.1016/j.patcog.2021.108252 Date de publication en ligne : 13/08/2021 En ligne : https://doi.org/10.1016/j.patcog.2021.108252 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=99615 [article]

Hyperspectral image denoising via clustering-based latent variable in variational Bayesian framework / Peyman Azimpour in IEEE Transactions on geoscience and remote sensing, vol 59 n° 4 (April 2021)  

Public [article]  inIEEE Transactions on geoscience and remote sensing > vol 59 n° 4 (April 2021) . - pp 3266 - 3276 Titre : Hyperspectral image denoising via clustering-based latent variable in variational Bayesian framework Type de document : Article/Communication Auteurs : Peyman Azimpour, Auteur ; Tahereh Bahraini, Auteur ; Hadi Sadoghi Yazdi, Auteur Année de publication : 2021 Article en page(s) : pp 3266 - 3276 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique [Termes IGN] analyse de groupement [Termes IGN] classification bayesienne [Termes IGN] classification floue [Termes IGN] distribution de Gauss [Termes IGN] factorisation de matrice non-négative [Termes IGN] filtrage du bruit [Termes IGN] filtre de Gauss [Termes IGN] image hyperspectrale [Termes IGN] Matlab [Termes IGN] processeur graphique [Termes IGN] qualité des données [Termes IGN] variable Résumé : (auteur) The hyperspectral-image (HSI) noise-reduction step is a very significant preprocessing phase of data-quality enhancement. It has been attracting immense research attention in the remote sensing and image processing domains. Many methods have been developed for HSI restoration, the goal of which is to remove noise from the whole HSI cube simultaneously without considering the spectral–spatial similarity. When a noise-removal algorithm is used globally to the entire data set, it would not eliminate all levels of noise, effectively. Furthermore, most of the existing methods remove independent and identically distributed (i.i.d.) Gaussian noise. The real scenarios are much more complicated than this assumption. The complexity created by natural noise that has a non-i.i.d. structure leads to inefficient methods containing underestimation and invalid performance. In this article, we calculated the spatial–spectral similarity criteria by defining a set of clustering-based latent variables (CLVs) in a Bayesian framework to improve the robustness. These criteria can be extracted using the clustering operators. Then, by applying the CLV to the variational Bayesian model, we investigated a new low-rank matrix factorization denoising approach based on the proposed clustering-based latent variable (CLV-LRMF) to remove noise with the non-i.i.d. mixture of Gaussian structures. Finally, we switched to the GPU for MATLAB implementation to reduce the runtime. The experimental results show that the performance has been improved by applying the proposed CLV and demonstrate the effectiveness of the proposed CLV-LRMF over other state-of-the-art methods. Numéro de notice : A2021-287 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1109/TGRS.2019.2939512 Date de publication en ligne : 24/03/2021 En ligne : https://doi.org/10.1109/TGRS.2019.2939512 Format de la ressource électronique : url article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=97396 [article]

Poststack seismic data denoising based on 3-D convolutional neural network / Dawei Liu in IEEE Transactions on geoscience and remote sensing, vol 58 n° 3 (March 2020)  

Public [article]  inIEEE Transactions on geoscience and remote sensing > vol 58 n° 3 (March 2020) . - pp 1598 - 1629 Titre : Poststack seismic data denoising based on 3-D convolutional neural network Type de document : Article/Communication Auteurs : Dawei Liu, Auteur ; Dawei Liu, Auteur ; Xiaokai Wang, Auteur ; et al., Auteur Année de publication : 2020 Article en page(s) : pp 1598 - 1629 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image [Termes IGN] apprentissage profond [Termes IGN] bruit blanc [Termes IGN] classification par réseau neuronal convolutif [Termes IGN] données localisées 3D [Termes IGN] échantillonnage [Termes IGN] filtrage du bruit [Termes IGN] filtre de Gauss [Termes IGN] post-stratification de données [Termes IGN] séisme [Termes IGN] sismologie Résumé : (Auteur) Deep learning has been successfully applied to image denoising. In this study, we take one step forward by using deep learning to suppress random noise in poststack seismic data from the aspects of network architecture and training samples. On the one hand, poststack seismic data denoising mainly aims at 3-D seismic data. We designed an end-to-end 3-D denoising convolutional neural network (3-D-DnCNN) that takes raw 3-D cubes as input in order to better extract the features of the 3-D spatial structure of poststack seismic data. On the other hand, denoising images with deep learning require noisy–clean sample pairs for training. In the field of seismic data processing, researchers usually try their best to suppress noise by using complex processes that combine different methods, but clean labels of seismic data are not available. In addition, building training samples in field seismic data has become an interesting but challenging problem. Therefore, we propose a training sample selection method that contains a complex workflow to produce comparatively ideal training samples. Experiments in this study demonstrate that deep learning can directly learn the ability to denoise field seismic data from selected samples. Although the building of the training samples may occur through a complex process, the experimental results of synthetic seismic data and field seismic data show that the 3-D-DnCNN has learned the ability to suppress the Gaussian noise and super-Gaussian noise from different training samples. Moreover, the 3-D-DnCNN network has better denoising performance toward arc-like imaging noise. In addition, we adopt residual learning and batch normalization in order to accelerate the training speed. After network training is satisfactorily completed, its processing efficiency can be significantly higher than that of conventional denoising methods. Numéro de notice : A2020-087 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1109/TGRS.2019.2947149 Date de publication en ligne : 06/11/2019 En ligne : https://doi.org/10.1109/TGRS.2019.2947149 Format de la ressource électronique : URL Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=94661 [article]

The guided bilateral filter: When the joint/cross bilateral filter becomes robust / Laurent Caraffa in IEEE Transactions on image processing, vol 24 n° 4 (April 2015)  

Permalink

The Guided Bilateral Filter: When the Joint/Cross Bilateral Filter Becomes Robust / Laurent Caraffa (2015)  

Permalink

Multiscale Haar transform for blur estimation from a set of images / Lâmân Lelégard (2011)    

Permalink

Generalisation methods for propagating updates between cartographic data sets / Lars Harrie (1998)

Permalink