Face Detection Using Neural Network Based Boosting Algorithm And PCA
An ideal Face Detection system should be able to identify and locate all faces regardless of their positions, scale,
orientation, lightning conditions, and expressions and so on. Face Detection is the prior stage in any face processing system,
as it provides challenging research area in computer vision and is of great interest. Challenges resides in the fact that the
faces are non-rigid objects. The goal of face detection is to detect human faces in still images or in different situations. Some
parameters plays a crucial role while detecting the faces amongst still images such as false positive, false negative, true
positive, and detection rate. High detection rate with high speed and accuracy of detector is the prime goal of this system.
In this paper, we applied Boosting Algorithm  which is capable of processing images rapidly. Here, we applied AdaBoost
which is an aggressive learning algorithm for solving classification problems, which combines an ensemble of weak
classifiers into a strong classifier. Specifically, to increase the Speed and Accuracy of the system which is the important
feature in case of this face detection system is being done with these neural network based boosting algorithms. By taking
AdaBoost classifiers in cascaded manner, a new boosting algorithm i.e. MLPBoost  which we used as a strong classifier
for face detection. Basically, MLPBoost is hybridization between AdaBoost and multi-layer perceptron networks. PCA i.e.
Principal Component Analysis is a very popular unsupervised statistical method to find useful image representation . This
method find out set of basic images and represents all the faces as a linear combination of those images.
Keywords-AdaBoost, Face Detection, MLPBoost, PCA (Principal Component Analysis).