Face Detection from Greyscale Images Using Details from Categorized Wavelet Coefficients as Features for a Dynamic Counterpropagation Network

Abstract

A dynamic counterpropagation network based on the forward only counterpropagation network (CPN) is applied to face detection in this paper. The network, called the dynamic supervised forward-propagation network (DSFPN) trains using a supervised algorithm and can grow dynamically during training allowing subclasses in the training data to be learnt. The network is trained using the categorized wavelet coefficients of the image as features of the image. The results suggests a 98% correct detection rate can be achieved with 4% false positives by increasing network complexity.