Descripteur



Etendre la recherche sur niveau(x) vers le bas
Aleatoric uncertainty estimation for dense stereo matching via CNN-based cost volume analysis / Max Mehltretter in ISPRS Journal of photogrammetry and remote sensing, vol 171 (January 2021)
![]()
[article]
Titre : Aleatoric uncertainty estimation for dense stereo matching via CNN-based cost volume analysis Type de document : Article/Communication Auteurs : Max Mehltretter, Auteur ; Christian Heipke, Auteur Année de publication : 2021 Article en page(s) : pp 63 - 75 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique
[Termes descripteurs IGN] appariement d'images
[Termes descripteurs IGN] apprentissage profond
[Termes descripteurs IGN] classification par réseau neuronal convolutif
[Termes descripteurs IGN] corrélation épipolaire dense
[Termes descripteurs IGN] couple stéréoscopique
[Termes descripteurs IGN] courbe épipolaire
[Termes descripteurs IGN] disparité
[Termes descripteurs IGN] effet de profondeur cinétique
[Termes descripteurs IGN] image RVB
[Termes descripteurs IGN] modèle d'incertitude
[Termes descripteurs IGN] modèle stochastique
[Termes descripteurs IGN] voxelRésumé : (auteur) Motivated by the need to identify erroneous disparity estimates, various methods for the estimation of aleatoric uncertainty in the context of dense stereo matching have been presented in recent years. Especially, the introduction of deep learning based methods and the accompanying significant improvement in accuracy have greatly increased the popularity of this field. Despite this remarkable development, most of these methods rely on features learned from disparity maps only, neglecting the corresponding 3-dimensional cost volumes. However, conventional hand-crafted methods have already demonstrated that the additional information contained in such cost volumes are beneficial for the task of uncertainty estimation. In this paper, we combine the advantages of deep learning and cost volume based features and present a new Convolutional Neural Network (CNN) architecture to directly learn features for the task of aleatoric uncertainty estimation from volumetric 3D data. Furthermore, we discuss and apply three different uncertainty models to train our CNN without the need to provide ground truth for uncertainty. In an extensive evaluation on three datasets using three common dense stereo matching methods, we investigate the effects of these uncertainty models and demonstrate the generality and state-of-the-art accuracy of the proposed method. Numéro de notice : A2021-012 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1016/j.isprsjprs.2020.11.003 date de publication en ligne : 18/11/2020 En ligne : https://doi.org/10.1016/j.isprsjprs.2020.11.003 Format de la ressource électronique : url article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=96415
in ISPRS Journal of photogrammetry and remote sensing > vol 171 (January 2021) . - pp 63 - 75[article]Hierarchical instance recognition of individual roadside trees in environmentally complex urban areas from UAV laser scanning point clouds / Yongjun Wang in ISPRS International journal of geo-information, vol 9 n° 10 (October 2020)
![]()
[article]
Titre : Hierarchical instance recognition of individual roadside trees in environmentally complex urban areas from UAV laser scanning point clouds Type de document : Article/Communication Auteurs : Yongjun Wang, Auteur ; Tengping Jiang, Auteur ; Jing Liu, Auteur ; et al., Auteur Année de publication : 2020 Article en page(s) : 26 p. Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Lasergrammétrie
[Termes descripteurs IGN] algorithme de filtrage
[Termes descripteurs IGN] apprentissage profond
[Termes descripteurs IGN] arbre hors forêt
[Termes descripteurs IGN] arbre urbain
[Termes descripteurs IGN] détection d'arbres
[Termes descripteurs IGN] données lidar
[Termes descripteurs IGN] données localisées 3D
[Termes descripteurs IGN] gestion urbaine
[Termes descripteurs IGN] image captée par drone
[Termes descripteurs IGN] optimisation (mathématiques)
[Termes descripteurs IGN] reconnaissance d'objets
[Termes descripteurs IGN] segmentation
[Termes descripteurs IGN] semis de points
[Termes descripteurs IGN] voxel
[Termes descripteurs IGN] zone urbaineRésumé : (auteur) Individual tree segmentation is essential for many applications in city management and urban ecology. Light Detection and Ranging (LiDAR) system acquires accurate point clouds in a fast and environmentally-friendly manner, which enables single tree detection. However, the large number of object categories and occlusion from nearby objects in complex environment pose great challenges in urban tree inventory, resulting in omission or commission errors. Therefore, this paper addresses these challenges and increases the accuracy of individual tree segmentation by proposing an automated method for instance recognition urban roadside trees. The proposed algorithm was implemented of unmanned aerial vehicles laser scanning (UAV-LS) data. First, an improved filtering algorithm was developed to identify ground and non-ground points. Second, we extracted tree-like objects via labeling on non-ground points using a deep learning model with a few smaller modifications. Unlike only concentrating on the global features in previous method, the proposed method revises a pointwise semantic learning network to capture both the global and local information at multiple scales, significantly avoiding the information loss in local neighborhoods and reducing useless convolutional computations. Afterwards, the semantic representation is fed into a graph-structured optimization model, which obtains globally optimal classification results by constructing a weighted indirect graph and solving the optimization problem with graph-cuts. The segmented tree points were extracted and consolidated through a series of operations, and they were finally recognized by combining graph embedding learning with a structure-aware loss function and a supervoxel-based normalized cut segmentation method. Experimental results on two public datasets demonstrated that our framework achieved better performance in terms of classification accuracy and recognition ratio of tree. Numéro de notice : A2020-665 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.3390/ijgi9100595 date de publication en ligne : 10/10/2020 En ligne : https://doi.org/10.3390/ijgi9100595 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=96142
in ISPRS International journal of geo-information > vol 9 n° 10 (October 2020) . - 26 p.[article]An improved constrained simultaneous iterative reconstruction technique for ionospheric tomography / Yi Bin Yao in GPS solutions, Vol 24 n° 3 (July 2020)
![]()
[article]
Titre : An improved constrained simultaneous iterative reconstruction technique for ionospheric tomography Type de document : Article/Communication Auteurs : Yi Bin Yao, Auteur ; Changzhi Zhai, Auteur ; Jian Kong, Auteur ; et al., Auteur Année de publication : 2020 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Géodésie spatiale
[Termes descripteurs IGN] données GNSS
[Termes descripteurs IGN] interpolation
[Termes descripteurs IGN] modèle ionosphérique
[Termes descripteurs IGN] reconstruction 3D
[Termes descripteurs IGN] teneur totale en électrons
[Termes descripteurs IGN] tomographie
[Termes descripteurs IGN] voxelRésumé : (auteur) Global Navigation Satellite System (GNSS) is now widely used for continuous ionospheric observations. Three-dimensional computerized ionospheric tomography (3DCIT) is an important tool for the reconstruction of electron density distributions in the ionosphere through effective use of the GNSS data. More specifically, the 3DCIT technique is able to resolve the three-dimensional electron density distributions over the reconstructed area based on the GNSS slant total electron content (STEC) observations. We present an Improved Constrained Simultaneous Iterative Reconstruction Technique (ICSIRT) algorithm that differs from the traditional ionospheric tomography methods in 3 ways. First, the ICSIRT computes the electron density corrections based on the product of the intercept and electron density within voxels so that the assignment of corrections at different heights becomes more reasonable. Second, an Inverse Distance Weighted (IDW) interpolation is used to restrict the electron density values in the voxels not traversed by GNSS rays, thereby ensuring the smoothness of the reconstructed region. Also, to improve the reconstruction accuracy around the HmF2 (the peak height of the F2 layer) altitude, a multiresolution grid is adopted in the vertical direction, with a 10-km resolution from 200 to 420 km and a 50-km resolution at other altitudes. The new algorithm has been applied to the GNSS data over the European and North American regions in different case studies that involve different seasonal conditions as well as a major storm. In the European region experiment, reconstruction results show that the new ICSIRT algorithm can effectively improve the reconstruction of the GNSS data. The electron density profiles retrieved from ICSIRT are much closer to the ionosonde observations than those from its predecessor, namely, the Constrained Simultaneous Iteration Reconstruction Technique (CSIRT). The reconstruction accuracy is significantly improved. In the North American region experiment, the electron density profiles in ICSIRT results show better agreement with incoherent scatter radar observations than CSIRT, even for the topside profiles. Numéro de notice : A2020-227 Affiliation des auteurs : non IGN Thématique : POSITIONNEMENT Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1007/s10291-020-00981-4 date de publication en ligne : 18/04/2020 En ligne : https://doi.org/10.1007/s10291-020-00981-4 Format de la ressource électronique : url article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=94958
in GPS solutions > Vol 24 n° 3 (July 2020)[article]Estimation of tropospheric wet refractivity using tomography method and artificial neural networks in Iranian case study / Mir Reza Ghaffari Razin in GPS solutions, Vol 24 n° 3 (July 2020)
![]()
[article]
Titre : Estimation of tropospheric wet refractivity using tomography method and artificial neural networks in Iranian case study Type de document : Article/Communication Auteurs : Mir Reza Ghaffari Razin, Auteur ; Behzad Voosoghi, Auteur Année de publication : 2020 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Applications de géodésie spatiale
[Termes descripteurs IGN] coefficient de corrélation
[Termes descripteurs IGN] données GPS
[Termes descripteurs IGN] erreur moyenne quadratique
[Termes descripteurs IGN] erreur relative
[Termes descripteurs IGN] Iran
[Termes descripteurs IGN] réfraction atmosphérique
[Termes descripteurs IGN] réseau neuronal artificiel
[Termes descripteurs IGN] retard troposphérique
[Termes descripteurs IGN] retard troposphérique zénithal
[Termes descripteurs IGN] tomographie par GPS
[Termes descripteurs IGN] vapeur d'eau
[Termes descripteurs IGN] voxelRésumé : (auteur) Using the observations from local and regional GPS networks, the estimation of slant wet delays (SWDs) is possible for each line of sight between satellite and receiver. The observations of SWD are used to model horizontal and vertical variations of the wet refractivity in the atmosphere above the study area. This work is done using the tomography method. In tomography, the horizontal variations of tropospheric wet refractivity are modeled with the polynomial in degree and rank of 2 with latitude and longitude as variables. Also, altitude variations are modeled in the form of discrete layers with constant heights. The main innovation is to estimate the tropospheric parameters for each line of sight by the artificial neural networks (ANNs). The SWD obtained from GPS observations for the different signals at each station is compared with the SWD generated by the ANNs (SWDGPS–SWDANNs). The square of the difference between these two values is introduced as the cost function in the ANNs. To evaluate, we used observations from October 27 to 31, 2011. The availability of GPS and radiosonde data is the main reason for choosing this timeframe. The correlation coefficient, root mean square error (RMSE), and relative error allow for evaluation of the proposed model. The results were also compared with the results of the voxel-based troposphere tomography method. For a more detailed evaluation, four test stations are selected and ANN zenith wet delays (ZWDANN) are compared with the ZWDGPS. Observations of test stations are not used in the modeling step. The correlation coefficient in the testing step for TomoANN and Tomovoxel is 0.9006 and 0.8863, respectively. The mean RMSE at 5 days for TomoANN and Tomovoxel is calculated as 0.63 and 0.71 mm/km, respectively. Also, the average relative error at the four test stations for TomoANN is 15.37% and for Tomovoxel it is 19.69%. The results demonstrate the better capability of the proposed method in the modeling of the tropospheric wet refractivity in the region of Iran. Numéro de notice : A2020-238 Affiliation des auteurs : non IGN Thématique : POSITIONNEMENT Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1007/s10291-020-00979-y date de publication en ligne : 10/04/2020 En ligne : https://doi.org/10.1007/s10291-020-00979-y Format de la ressource électronique : url article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=94986
in GPS solutions > Vol 24 n° 3 (July 2020)[article]Tree annotations in LiDAR data using point densities and convolutional neural networks / Ananya Gupta in IEEE Transactions on geoscience and remote sensing, vol 58 n° 2 (February 2020)
![]()
[article]
Titre : Tree annotations in LiDAR data using point densities and convolutional neural networks Type de document : Article/Communication Auteurs : Ananya Gupta, Auteur ; Jonathan Byrne, Auteur ; David Moloney, Auteur Année de publication : 2020 Article en page(s) : pp 971 - 981 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Lasergrammétrie
[Termes descripteurs IGN] classification par réseau neuronal convolutif
[Termes descripteurs IGN] données lidar
[Termes descripteurs IGN] Dublin (Irlande ; ville)
[Termes descripteurs IGN] étiquette
[Termes descripteurs IGN] extraction d'arbres
[Termes descripteurs IGN] image spectrale
[Termes descripteurs IGN] Montréal (Québec)
[Termes descripteurs IGN] segmentation
[Termes descripteurs IGN] semis de points
[Termes descripteurs IGN] voxel
[Termes descripteurs IGN] zone urbaineRésumé : (auteur) LiDAR provides highly accurate 3-D point clouds. However, data need to be manually labeled in order to provide subsequent useful information. Manual annotation of such data is time-consuming, tedious, and error prone, and hence, in this article, we present three automatic methods for annotating trees in LiDAR data. The first method requires high-density point clouds and uses certain LiDAR data attributes for the purpose of tree identification, achieving almost 90% accuracy. The second method uses a voxel-based 3-D convolutional neural network on low-density LiDAR data sets and is able to identify most large trees accurately but struggles with smaller ones due to the voxelization process. The third method is a scaled version of the PointNet++ method and works directly on outdoor point clouds and achieves an F score of 82.1% on the ISPRS benchmark data set, comparable to the state-of-the-art methods but with increased efficiency. Numéro de notice : A2020-095 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1109/TGRS.2019.2942201 date de publication en ligne : 11/10/2019 En ligne : https://doi.org/10.1109/TGRS.2019.2942201 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=94658
in IEEE Transactions on geoscience and remote sensing > vol 58 n° 2 (February 2020) . - pp 971 - 981[article]A new method of equiangular sectorial voxelization of single-scan terrestrial laser scanning data and its applications in forest defoliation estimation / Langning Huo in ISPRS Journal of photogrammetry and remote sensing, vol 151 (May 2019)
PermalinkPairwise coarse registration of point clouds in urban scenes using voxel-based 4-planes congruent sets / Yusheng Xu in ISPRS Journal of photogrammetry and remote sensing, vol 151 (May 2019)
PermalinkConditional random field and deep feature learning for hyperspectral image classification / Fahim Irfan Alam in IEEE Transactions on geoscience and remote sensing, vol 57 n° 3 (March 2019)
PermalinkA time‐geographic approach to quantifying wildlife–road interactions / Rebecca W. Loraamm in Transactions in GIS, vol 23 n° 1 (February 2019)
PermalinkAnalyzing the role of pulse density and voxelization parameters on full-waveform LiDAR-derived metrics / Pablo Crespo-Peremarch in ISPRS Journal of photogrammetry and remote sensing, vol 146 (December 2018)
PermalinkA greyscale voxel model for airborne lidar data applied to building detection / Liying Wang in Photogrammetric record, vol 33 n° 164 (December 2018)
PermalinkAnalyzing the vertical distribution of crown material in mixed stand composed of two temperate tree species / Olivier Martin-Ducup in Forests, vol 9 n° 11 (November 2018)
PermalinkSDF-2-SDF registration for real-time 3D reconstruction from RGB-D data / Miroslava Slavcheva in International journal of computer vision, vol 126 n° 6 (June 2018)
PermalinkA voxel- and graph-based strategy for segmenting man-made infrastructures using perceptual grouping laws: comparison and evaluation / Yusheng Xu in Photogrammetric Engineering & Remote Sensing, PERS, vol 84 n° 6 (juin 2018)
Permalink3D visibility analysis indicating quantitative and qualitative aspects of the visible space / D. Golub in Survey review, vol 50 n° 359 (March 2018)
Permalink