Détail de l'auteur
Auteur Jiasong Zhu |
Documents disponibles écrits par cet auteur (2)
Ajouter le résultat dans votre panier Affiner la recherche Interroger des sources externes
Decomposition of LiDAR waveforms by B-spline-based modeling / Xiang Shen in ISPRS Journal of photogrammetry and remote sensing, vol 128 (June 2017)
[article]
Titre : Decomposition of LiDAR waveforms by B-spline-based modeling Type de document : Article/Communication Auteurs : Xiang Shen, Auteur ; Qing-Quan Li, Auteur ; Guofeng Wu, Auteur ; Jiasong Zhu, Auteur Année de publication : 2017 Article en page(s) : pp 182 - 191 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Lasergrammétrie
[Termes IGN] B-Spline
[Termes IGN] décomposition de Gauss
[Termes IGN] distribution, loi de
[Termes IGN] forme d'onde pleine
[Termes IGN] traitement du signal
[Termes IGN] transformation géométrique
[Termes IGN] translationRésumé : (Auteur) Waveform decomposition is a widely used technique for extracting echoes from full-waveform LiDAR data. Most previous studies recommended the Gaussian decomposition approach, which employs the Gaussian function in laser pulse modeling. As the Gaussian-shape assumption is not always satisfied for real LiDAR waveforms, some other probability distributions (e.g., the lognormal distribution, the generalized normal distribution, and the Burr distribution) have also been introduced by researchers to fit sharply-peaked and/or heavy-tailed pulses. However, these models cannot be universally used, because they are only suitable for processing the LiDAR waveforms in particular shapes. In this paper, we present a new waveform decomposition algorithm based on the B-spline modeling technique. LiDAR waveforms are not assumed to have a priori shapes but rather are modeled by B-splines, and the shape of a received waveform is treated as the mixture of finite transmitted pulses after translation and scaling transformation. The performance of the new model was tested using two full-waveform data sets acquired by a Riegl LMS-Q680i laser scanner and an Optech Aquarius laser bathymeter, comparing with three classical waveform decomposition approaches: the Gaussian, generalized normal, and lognormal distribution-based models. The experimental results show that the B-spline model performed the best in terms of waveform fitting accuracy, while the generalized normal model yielded the worst performance in the two test data sets. Riegl waveforms have nearly Gaussian pulse shapes and were well fitted by the Gaussian mixture model, while the B-spline-based modeling algorithm produced a slightly better result by further reducing 6.4% of fitting residuals, largely benefiting from alleviating the adverse impact of the ringing effect. The pulse shapes of Optech waveforms, on the other hand, are noticeably right-skewed. The Gaussian modeling results deviated significantly from original signals, and the extracted echo parameters were clearly inaccurate and unreliable. The B-spline-based method performed significantly better than the Gaussian and lognormal models by reducing 45.5% and 11.5% of their fitting errors, respectively. Much more precise echo properties can accordingly be retrieved with a high probability. Benefiting from the flexibility of B-splines on fitting arbitrary curves, the new method has the potentiality for accurately modeling various full-waveform LiDAR data, whether they are nearly Gaussian or non-Gaussian in shape. Numéro de notice : A2017-334 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1016/j.isprsjprs.2017.03.006 En ligne : https://doi.org/10.1016/j.isprsjprs.2017.03.006 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=85487
in ISPRS Journal of photogrammetry and remote sensing > vol 128 (June 2017) . - pp 182 - 191[article]Réservation
Réserver ce documentExemplaires(3)
Code-barres Cote Support Localisation Section Disponibilité 081-2017061 RAB Revue Centre de documentation En réserve L003 Disponible 081-2017063 DEP-EXM Revue LASTIG Dépôt en unité Exclu du prêt 081-2017062 DEP-EAF Revue Nancy Dépôt en unité Exclu du prêt Superpixel-based multitask learning framework for hyperspectral image classification / Sen Jia in IEEE Transactions on geoscience and remote sensing, vol 55 n° 5 (May 2017)
[article]
Titre : Superpixel-based multitask learning framework for hyperspectral image classification Type de document : Article/Communication Auteurs : Sen Jia, Auteur ; Bin Deng, Auteur ; Jiasong Zhu, Auteur ; Xiuping Jia, Auteur Année de publication : 2017 Article en page(s) : pp 2575 - 2588 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique
[Termes IGN] apprentissage automatique
[Termes IGN] classification par séparateurs à vaste marge
[Termes IGN] filtre de Gabor
[Termes IGN] image hyperspectraleRésumé : (Auteur) Due to the high spectral dimensionality of hyperspectral images as well as the difficult and time-consuming process of collecting sufficient labeled samples in practice, the small sample size scenario is one crucial problem and a challenging issue for hyperspectral image classification. Fortunately, the structure information of materials, reflecting region of homogeneity in the spatial domain, offers an invaluable complement to the spectral information. Assuming some spatial regularity and locality of surface materials, it is reasonable to segment the image into different homogeneous parts in advance, called superpixel, which can be used to improve the classification performance. In this paper, a superpixel-based multitask learning framework has been proposed for hyperspectral image classification. Specifically, a set of 2-D Gabor filters are first applied to hyperspectral images to extract discriminative features. Meanwhile, a superpixel map is generated from the hyperspectral images. Second, a superpixel-based spatial-spectral Schroedinger eigenmaps (S4E) method is adopted to effectively reduce the dimensions of each extracted Gabor cube. Finally, the classification is carried out by a support vector machine (SVM)-based multitask learning framework. The proposed approach is thus termed Gabor S4E and SVM-based multitask learning (GS4E-MTLSVM). A series of experiments is conducted on three real hyperspectral image data sets to demonstrate the effectiveness of the proposed GS4E-MTLSVM approach. The experimental results show that the performance of the proposed GS4E-MTLSVM is better than those of several state-of-the-art methods, while the computational complexity has been greatly reduced, compared with the pixel-based spatial-spectral Schroedinger eigenmaps method. Numéro de notice : A2017-466 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1109/TGRS.2017.2647815 En ligne : http://dx.doi.org/10.1109/TGRS.2017.2647815 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=86389
in IEEE Transactions on geoscience and remote sensing > vol 55 n° 5 (May 2017) . - pp 2575 - 2588[article]