Automatic Seizure Detection in Face Videos Based on CWT

TLDR: A novel Radon Transform based technique on Dual Tree Complex Wavelet DT-CWT coefficients that can give an indication of early seizure signs from videos by magnifying micro movements in a complete automatic manner, without any human interaction is introduced.

Abstract: Magnifying micro movements of natural videos that are undetectable by human eye have recently received considerable interests. This is due to its impact on numerous applications. Seizure has been classified as a dangerous symptom of a victim’s behavior. It is an indication of abnormality in the brain neuro activity which can lead to a damage of brain cells and victims gets worse rapidly. In this paper, we introduce a novel Radon Transform based technique on Dual Tree Complex Wavelet DT-CWT coefficients that can give an indication of early seizure signs from videos by magnifying micro movements in a complete automatic manner, without any human interaction. We modify the phases of the CWT wavelet coefficients of successive video frames, in order to detect any minor change in the object’s spatial position. We limited our experiments to baby video due to data availability and privacy conformity. We were able to detect all cases of true seizure in our limited database as will be illustrated in our simulation results. Our results show that our proposed system can be utilized for automatic seizure detection and analysis.