Cristin-resultat-ID: 1720499
Sist endret: 20. desember 2019 16:50
NVI-rapporteringsår: 2019
Resultat
Vitenskapelig artikkel
2019

Unsupervised Image Regression for Heterogeneous Change Detection

Bidragsytere:
  • Luigi Tommaso Luppino
  • Filippo Maria Bianchi
  • Gabriele Moser og
  • Stian Normann Anfinsen

Tidsskrift

IEEE Transactions on Geoscience and Remote Sensing
ISSN 0196-2892
e-ISSN 1558-0644
NVI-nivå 2

Om resultatet

Vitenskapelig artikkel
Publiseringsår: 2019
Publisert online: 2019
Trykket: 2019
Volum: 57
Hefte: 12
Sider: 9960 - 9975
Artikkelnummer: 19157193
Open Access

Importkilder

Scopus-ID: 2-s2.0-85075675875

Klassifisering

Vitenskapsdisipliner

Simulering, visualisering, signalbehandling, bildeanalyse

Emneord

Endringsdeteksjon • Jordobservasjon fra satellitter • Mønstergjenkjenning • Maskinlæring

Beskrivelse Beskrivelse

Tittel

Unsupervised Image Regression for Heterogeneous Change Detection

Sammendrag

Change detection (CD) in heterogeneous multitemporal satellite images is an emerging and challenging topic in remote sensing. In particular, one of the main challenges is to tackle the problem in an unsupervised manner. In this paper, we propose an unsupervised framework for bitemporal heterogeneous CD based on the comparison of affinity matrices and image regression. First, our method quantifies the similarity of affinity matrices computed from colocated image patches in the two images. This is done to automatically identify pixels that are likely to be unchanged. With the identified pixels as pseudotraining data, we learn a transformation to map the first image to the domain of the other image and vice versa. Four regression methods are selected to carry out the transformation: Gaussian process regression, support vector regression, random forest regression (RFR), and a recently proposed kernel regression method called homogeneous pixel transformation. To evaluate the potentials and limitations of our framework and also the benefits and disadvantages of each regression method, we perform experiments on two real data sets. The results indicate that the comparison of the affinity matrices can already be considered a CD method by itself. However, image regression is shown to improve the results obtained by the previous step alone and produces accurate CD maps despite of the heterogeneity of the multitemporal input data. Notably, the RFR approach excels by achieving similar accuracy as the other methods, but with a significantly lower computational cost and with fast and robust tuning of hyperparameters.

Bidragsytere

Luigi Tommaso Luppino

  • Tilknyttet:
    Forfatter
    ved Institutt for fysikk og teknologi ved UiT Norges arktiske universitet

Filippo Maria Bianchi

  • Tilknyttet:
    Forfatter
    ved Institutt for fysikk og teknologi ved UiT Norges arktiske universitet

Gabriele Moser

  • Tilknyttet:
    Forfatter
    ved Università degli Studi di Genova
Aktiv cristin-person

Stian Normann Anfinsen

  • Tilknyttet:
    Forfatter
    ved Institutt for fysikk og teknologi ved UiT Norges arktiske universitet
1 - 4 av 4