Compressão de Sinais de EEG com JPEG2000 e H.264

Heitor J. Savino, Eddie B. Lima Filho, Waldir S. Silva Júnior

DOI: 10.14209/sbrt.2012.124

Evento: XXX Simpósio Brasileiro de Telecomunicações (SBrT2012)

Keywords: Electroencephalography Data Compression H.264 JPEG2000 Preprocessing

Abstract

Recently, some authors have addressed the encoding of biological signals as two-dimensional arrays, in such a way that intra and intersegment dependencies are exploited by off- the-shelf image encoders. However, the rearrangement in a 2-d array itself provides a poor correlation across signal segments, which has the potential to reduce the compression efficiency. In order to correct this behavior, preprocessing techniques have been applied to matrices generated from different biological signals, with the goal of increasing the correlation of the arrangement, which favors the exploitation of signal features. The present work addresses this issue, by using the H.264 and JPEG2000 as image compressors. The simulation results show that the proposed methodology is competitive and constitutes a viable alternative for electroencephalogram compression.

Download