Security Systems with Biometry based on Partial View Facial Images Using Geometrical Features
Piotr Milczarski, Zofia Stawska, Shane Dowdall · 2018
In the security systems with a biometrical support it is vital to check non-invasively some features e.g. iris color, gender. In the paper, we have shown the problem of choosing the best set of features in the task of gender classification/recognition. The choice of a minimum set of features, that can give satisfactory results is also important in the biometrical systems where only a part of the face is visible e.g. security check. The minimum set of features can simplify the classification process to make it really fast in IoT security systems, video analysis of faces in the crowd, surveillance video analysis as well as for mobile applications The proposed areas can be taken into account while having veiled or partially obscured faces.