ANAP Seminar | Speaker: Rui Pereira (CMAT).

Venue: Sala de Seminários Ed.6-3.08, Campus de Gualtar

Online: Link  

Abstract: We study the problem of classification of various real-world objects using as input a database (DB) of laboratory polarimetric measures (Mueller Matrix elements - MMEs). It can be used in autonomous driving. 

To this end, we look for an algorithm using less input parameters without great loss of the quality of classification. Different sets of attributes are studied using an artificial neural network (ANN).

After that, an improved machine learning (ML) architecture is built using the K- nearest neighbors (KNN) classifier on each cluster generated by applying the pre-trained ANN to the training set. This is important if one cannot measure all MMEs, or it would be too expensive or challenging .

The results obtained for a reduced set of attributes using different ML architectures are quite good, especially for the proposed combined ANN-KNN which can help to avoid measuring all MMEs.

This is a joint work with Filipe Oliveira (CFIS-UM), Nazar Romanyshyn (CFIS-UM), Irene Estevez (DFIS – AUB), Joel Borges (CFIS-UM), Stéphane Clain (CMUC -UC), Mikhail I. Vasilevskiy (CFIS-UM).