Face Localization, Color Images and Postural Degradation

The present research explores experimental evaluation of the face localization method using skin color based approach and works on the color images. This method first corrects the used bias by lighting compensation techniques which automatically estimate the references white pixels. The outcome difficulty of localizing the low and high luma skin tones is performed by applying a nonlinear transformation to the YCbCr color space. Proposed method is able to localize skin regions over the face images and thereby identifies the candidate based on the skin patches. The present study continues towards the identification of the most suitable unique class of model and the parameter reductions for elderly fall prevention.

Autorentext

Chandan Srivastava is doctoral research fellow at Department deEnginyeria Química, Universitat Rovira i Virgili,Spain. He hasobtained his master study and research at the Department ofSystem Optimization and Security, University of Technology ofTroyes,France. His area of research includes machinelearning, and biomedical signal processing.