Hyperspectral and LiDAR fusion using deep three-stream convolutional neural networks
Recently, convolutional neural networks (CNN) have been intensively investigated for the classification of remote sensing data by extracting invariant and abstract features suitable for classification. In this paper, a novel framework is proposed for the fusion of hyperspectral images and LiDAR-deri...
Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Dokumenttyp: | Article (Journal) |
| Sprache: | Englisch |
| Veröffentlicht: |
16 October 2018
|
| In: |
Remote sensing
Year: 2018, Jahrgang: 10, Heft: 10 |
| ISSN: | 2072-4292 |
| DOI: | 10.3390/rs10101649 |
| Online-Zugang: | Verlag, Volltext: https://doi.org/10.3390/rs10101649 Verlag, Volltext: https://www.mdpi.com/2072-4292/10/10/1649 
|
| Verfasserangaben: | Hao Li, Pedram Ghamisi, Uwe Soergel and Xiao Xiang Zhu |
MARC
| LEADER | 00000caa a2200000 c 4500 | ||
|---|---|---|---|
| 001 | 1689781378 | ||
| 003 | DE-627 | ||
| 005 | 20220817225734.0 | ||
| 007 | cr uuu---uuuuu | ||
| 008 | 200211s2018 xx |||||o 00| ||eng c | ||
| 024 | 7 | |a 10.3390/rs10101649 |2 doi | |
| 035 | |a (DE-627)1689781378 | ||
| 035 | |a (DE-599)KXP1689781378 | ||
| 035 | |a (OCoLC)1341305003 | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rda | ||
| 041 | |a eng | ||
| 084 | |a 31 |2 sdnb | ||
| 100 | 1 | |a Li, Hao |e VerfasserIn |0 (DE-588)1197649670 |0 (DE-627)1679340883 |4 aut | |
| 245 | 1 | 0 | |a Hyperspectral and LiDAR fusion using deep three-stream convolutional neural networks |c Hao Li, Pedram Ghamisi, Uwe Soergel and Xiao Xiang Zhu |
| 264 | 1 | |c 16 October 2018 | |
| 300 | |b Illustrationen | ||
| 300 | |a 20 | ||
| 336 | |a Text |b txt |2 rdacontent | ||
| 337 | |a Computermedien |b c |2 rdamedia | ||
| 338 | |a Online-Ressource |b cr |2 rdacarrier | ||
| 500 | |a Gesehen am 11.02.2020 | ||
| 520 | |a Recently, convolutional neural networks (CNN) have been intensively investigated for the classification of remote sensing data by extracting invariant and abstract features suitable for classification. In this paper, a novel framework is proposed for the fusion of hyperspectral images and LiDAR-derived elevation data based on CNN and composite kernels. First, extinction profiles are applied to both data sources in order to extract spatial and elevation features from hyperspectral and LiDAR-derived data, respectively. Second, a three-stream CNN is designed to extract informative spectral, spatial, and elevation features individually from both available sources. The combination of extinction profiles and CNN features enables us to jointly benefit from low-level and high-level features to improve classification performance. To fuse the heterogeneous spectral, spatial, and elevation features extracted by CNN, instead of a simple stacking strategy, a multi-sensor composite kernels (MCK) scheme is designed. This scheme helps us to achieve higher spectral, spatial, and elevation separability of the extracted features and effectively perform multi-sensor data fusion in kernel space. In this context, a support vector machine and extreme learning machine with their composite kernels version are employed to produce the final classification result. The proposed framework is carried out on two widely used data sets with different characteristics: an urban data set captured over Houston, USA, and a rural data set captured over Trento, Italy. The proposed framework yields the highest OA of 92 . 57 % and 97 . 91 % for Houston and Trento data sets. Experimental results confirm that the proposed fusion framework can produce competitive results in both urban and rural areas in terms of classification accuracy, and significantly mitigate the salt and pepper noise in classification maps. | ||
| 650 | 4 | |a composite kernels | |
| 650 | 4 | |a convolutional neural networks (CNN) | |
| 650 | 4 | |a data fusion | |
| 650 | 4 | |a extinction profiles (EPs) | |
| 650 | 4 | |a feature extraction (FE) | |
| 700 | 1 | |a Ghamisi, Pedram |e VerfasserIn |0 (DE-588)1080244328 |0 (DE-627)844076678 |0 (DE-576)434889962 |4 aut | |
| 700 | 1 | |a Sörgel, Uwe |d 1969- |e VerfasserIn |0 (DE-588)12495491X |0 (DE-627)706746392 |0 (DE-576)294587527 |4 aut | |
| 700 | 1 | |a Zhu, Xiaoxiang |d 1984- |e VerfasserIn |0 (DE-588)1013459822 |0 (DE-627)66572800X |0 (DE-576)348363230 |4 aut | |
| 773 | 0 | 8 | |i Enthalten in |t Remote sensing |d Basel : MDPI, 2009 |g 10(2018,10) Artikel-Nummer 1649, Seite 1-20, 20 Seiten |h Online-Ressource |w (DE-627)608937916 |w (DE-600)2513863-7 |w (DE-576)310882532 |x 2072-4292 |7 nnas |a Hyperspectral and LiDAR fusion using deep three-stream convolutional neural networks |
| 773 | 1 | 8 | |g volume:10 |g year:2018 |g number:10 |g extent:20 |a Hyperspectral and LiDAR fusion using deep three-stream convolutional neural networks |
| 856 | 4 | 0 | |u https://doi.org/10.3390/rs10101649 |x Verlag |x Resolving-System |3 Volltext |
| 856 | 4 | 0 | |u https://www.mdpi.com/2072-4292/10/10/1649 |x Verlag |3 Volltext |
| 951 | |a AR | ||
| 992 | |a 20200211 | ||
| 993 | |a Article | ||
| 994 | |a 2018 | ||
| 998 | |g 1197649670 |a Li, Hao |m 1197649670:Li, Hao |d 120000 |d 120700 |e 120000PL1197649670 |e 120700PL1197649670 |k 0/120000/ |k 1/120000/120700/ |p 1 |x j | ||
| 999 | |a KXP-PPN1689781378 |e 3592805215 | ||
| BIB | |a Y | ||
| SER | |a journal | ||
| JSO | |a {"relHost":[{"type":{"bibl":"periodical","media":"Online-Ressource"},"disp":"Hyperspectral and LiDAR fusion using deep three-stream convolutional neural networksRemote sensing","pubHistory":["1.2009 -"],"note":["Gesehen am 06.09.10"],"part":{"year":"2018","volume":"10","text":"10(2018,10) Artikel-Nummer 1649, Seite 1-20, 20 Seiten","extent":"20","issue":"10"},"language":["eng"],"origin":[{"dateIssuedDisp":"2009-","publisher":"MDPI","publisherPlace":"Basel","dateIssuedKey":"2009"}],"id":{"issn":["2072-4292"],"eki":["608937916"],"zdb":["2513863-7"]},"physDesc":[{"extent":"Online-Ressource"}],"title":[{"title_sort":"Remote sensing","title":"Remote sensing"}],"recId":"608937916","name":{"displayForm":["Molecular Diversity Preservation International (MDPI)"]}}],"origin":[{"dateIssuedDisp":"16 October 2018","dateIssuedKey":"2018"}],"title":[{"title_sort":"Hyperspectral and LiDAR fusion using deep three-stream convolutional neural networks","title":"Hyperspectral and LiDAR fusion using deep three-stream convolutional neural networks"}],"type":{"bibl":"article-journal","media":"Online-Ressource"},"id":{"eki":["1689781378"],"doi":["10.3390/rs10101649"]},"physDesc":[{"extent":"20 S.","noteIll":"Illustrationen"}],"recId":"1689781378","person":[{"role":"aut","family":"Li","given":"Hao","display":"Li, Hao"},{"family":"Ghamisi","role":"aut","given":"Pedram","display":"Ghamisi, Pedram"},{"given":"Uwe","display":"Sörgel, Uwe","family":"Sörgel","role":"aut"},{"role":"aut","family":"Zhu","given":"Xiaoxiang","display":"Zhu, Xiaoxiang"}],"note":["Gesehen am 11.02.2020"],"name":{"displayForm":["Hao Li, Pedram Ghamisi, Uwe Soergel and Xiao Xiang Zhu"]},"language":["eng"]} | ||
| SRT | |a LIHAOGHAMIHYPERSPECT1620 | ||