Trainee Project
Title:
Multimodal image fusion. Applications in geological imaging
Dates:
2025/03/01 - 2025/08/31
Supervisor(s):
Other supervisor(s):
DELCHINI Sylvain (s.delchini@brgm.fr)
, LOHIER Théophile (t.lohier@brgm.fr)
Description:
One of the main objectives of PEPR Sous-sol is to define the conditions for sustainable and optimal use of the subsoil [1,2]. This involves not only improving geological knowledge of the subsoil and the technical, economic, social and environmental aspects of its use, but also developing innovative interdisciplinary scientific approaches to promote its exploitation.
Information on the mineralogical composition of samples can be obtained using a number of techniques. One such technique is hyperspectral imaging. Mineral identification often requires the use of several acquisition modalities corresponding to different spectral bands, such as near infrared (NIR), short wavelength infrared (SWIR) and middle wavelength infrared (MWIR). Data from these different modalities need to be mapped into a common digital representation in order to be exploited. The aim of this internship is to propose fast fusion algorithms for processing images comprising up to 10^5 voxels, each made up of 10^3 wavelengths. The first step in this work is to establish a state-of-the-art of existing fusion methods [3], focusing on methods based on optimal transport [4,5]. A performance study of a few fusion approaches is also planned. This work will benefit from data obtained on a BRGM platform in Orléans.
[1] https://www.brgm.fr/en/programme/subsurface-common-good-research-programme-responsible-sustainable-use-exploitation
[2] https://www.brgm.fr/fr/programme/sous-sol-bien-commun-programme-recherche-utilisation-exploitation-responsable-durable
[3] D. Lahat, T. Adali, C. Jutten. Multimodal Data Fusion: An Overview of Methods, Challenges and Prospects. Proceedings of the IEEE, vol. 3, no. 9, 2015.
[4] N. Courty, R. Flamary, D. Tuia, T. Corpetti. Optimal Transport for Data Fusion in Remote Sensing. IGARSS, Jul 2016, Beijing, China.
[5] J. Mifdal, B. Coll, N. Courty, J. Froment, B. Vedel. Hyperspectral and multispectral Wasserstein barycenter for image fusion. IGARSS 2017, Jul 2017, Houston, United States.
Information on the mineralogical composition of samples can be obtained using a number of techniques. One such technique is hyperspectral imaging. Mineral identification often requires the use of several acquisition modalities corresponding to different spectral bands, such as near infrared (NIR), short wavelength infrared (SWIR) and middle wavelength infrared (MWIR). Data from these different modalities need to be mapped into a common digital representation in order to be exploited. The aim of this internship is to propose fast fusion algorithms for processing images comprising up to 10^5 voxels, each made up of 10^3 wavelengths. The first step in this work is to establish a state-of-the-art of existing fusion methods [3], focusing on methods based on optimal transport [4,5]. A performance study of a few fusion approaches is also planned. This work will benefit from data obtained on a BRGM platform in Orléans.
[1] https://www.brgm.fr/en/programme/subsurface-common-good-research-programme-responsible-sustainable-use-exploitation
[2] https://www.brgm.fr/fr/programme/sous-sol-bien-commun-programme-recherche-utilisation-exploitation-responsable-durable
[3] D. Lahat, T. Adali, C. Jutten. Multimodal Data Fusion: An Overview of Methods, Challenges and Prospects. Proceedings of the IEEE, vol. 3, no. 9, 2015.
[4] N. Courty, R. Flamary, D. Tuia, T. Corpetti. Optimal Transport for Data Fusion in Remote Sensing. IGARSS, Jul 2016, Beijing, China.
[5] J. Mifdal, B. Coll, N. Courty, J. Froment, B. Vedel. Hyperspectral and multispectral Wasserstein barycenter for image fusion. IGARSS 2017, Jul 2017, Houston, United States.
Keywords:
Data fusion, hyperspectral imaging, multimodal imaging, optimal transport
Conditions:
Profile: Master's internship in signal processing, computer science or applied mathematics.
Compensation: around ¬600/month.
The internship lasts 6 months. The trainee will be based at the Faculté des Science et Technologies (Bât. Henri Poincaré) in Vandoeuvre-lès-Nancy, France.
Compensation: around ¬600/month.
The internship lasts 6 months. The trainee will be based at the Faculté des Science et Technologies (Bât. Henri Poincaré) in Vandoeuvre-lès-Nancy, France.
Department(s):
| Biology, Signals and Systems in Cancer and Neuroscience |
Funds:
PEPR Sous-sol