Creating and sharing knowledge for telecommunications

Project: Plenoptic imaging for skin lesion assessment

Acronym: PlenoISLA
Main Objective: This project proposes the development and implementation of new imaging techniques for non-invasive skin surface characterization, based on recent light field imaging technology. The main objective is to obtain a new set of 3D-based quantitative markers associated to physiological processes of the skin, allowing the characterization of its morphological and functional structure, that can then be used as markers in evolution and prognosis in Dermatology studies. Based on the obtained skin 3D characteristics, new patterns of dermoholoscopy will be considered, leading to the definition of clusters of features and dermoholoscopic patterns to be used in the context of automatic learning algorithms, in particular within the scope of "Deep Learning". Besides the computational analysis, new techniques for acquisition and compression of the light field images will also be addressed, as well as a database with skin lesion plenoptic images that will be created for the scientific community.
Reference: PTDC/EEI-TEL/28325/2017
Funding: FCT/POCI, P2020
Start Date: 25-07-2018
End Date: 24-07-2021
Team: Sérgio Manuel Maciel Faria, Rui Manuel Fonseca Pinto, Pedro António Amado Assunção, Luís Miguel de Oliveira Pegado de Noronha e Távora, Lucas Arrabal Thomaz, Pedro Miguel Marques Pereira, João Miguel Pereira da Silva Santos, Rui Miguel Leonel Lourenço, Francisco Ferreira Cunha, Miguel Santos, Cristian Perra
Groups: Multimedia Signal Processing – Lr
Partners: Centro Hospitalar de Leiria
Local Coordinator: Sérgio Manuel Maciel Faria
...
Associated Publications
  • 2Thesis
  • Improved Image Rendering for Focused Plenoptic Cameras with Extended Depth-of-field, Instituto Politécnico de Leiria, Supervisor: S.M.M. Faria, out-2016 - jan-2019
  • Structure Tensor-based Depth Estimation from Light Field Images, Instituto Politécnico de Leiria, Supervisor: S.M.M. Faria, Co-supervisor: L.M. Távora, out-2016 - jan-2019