Descripción
|
|
---|---|
In this paper, a study of the adaptation of a K Nearest Neighbors (KNN) algorithm on a system that integrates a host computer and an accelerator of 256 cores in 16 clusters - a Massively Parallel Processor Array (MPPA) - is presented. The goal of this research line is to exploit the possibilities the MPPA offers to speedup the execution of the KNN algorithm. The functional aim of the KNN algorithm is to filter the probability map provided by a Support Vector Machine (S VM) classifier with (1) the assistance of the spectral information obtained from a Principal Components Analysis (PCA) algorithm and (2) the inherent spatial information of hyperspectral images. The KNN algorithm is a key block of an algorithm employed to discriminate between cancer and normal tissues during neurosurgical procedures. A testbench has been built using hyperspectral medical brain images captured in two operating theaters. As a result, the execution time of the KNN algorithm in the MPPA developer workstation is in the order of the time required for the hyperspectral sensor to capture a new image. | |
Internacional
|
Si |
Nombre congreso
|
XXXI Design of Circuits and Integrated Systems Conference |
Tipo de participación
|
960 |
Lugar del congreso
|
Granada, Spain |
Revisores
|
Si |
ISBN o ISSN
|
978-1-5090-4565-5 |
DOI
|
|
Fecha inicio congreso
|
23/11/2016 |
Fecha fin congreso
|
25/11/2016 |
Desde la página
|
1 |
Hasta la página
|
5 |
Título de las actas
|
Proceedings of the XXXI Design of Circuits and Integrated Systems Conference |