Automation of smart home for well being of individual using face detection and recognition.

This work designs Face detection and recognition system to provide security to an individual living alone at home. Enhancement and improvement of this complete system is done by joining two algorithms Principle Component Analysis and Independent Component Analysis and the combination of these algorithms i.e. PICA is used for feature extraction. Firstly, the image of a person is acquired using camera, then face detection is done based on template matching and face recognition is done by Feed Forward Neural Network. This improved version of Face Recognition is combined with improved version of Voice Recognition to identify a person and to provide more secure environment to an individual living alone at home. In Voice Recognition, RASTA-Perceptual Linear Prediction with Mel Frequency Cepstal Coefficient module is used and for Face Recognition, PCA with ICA algorithms is used. There are two steps in this: if Face is recognised successfully of a person, then Voice will be recognised, if both are recognised accurately for an individual then access will be granted otherwise if any one parameter is not recognised properly then we calculate the similarity index of that individual and if it shows the value greater than 75% then only permission will be granted else the person will be declared as unknown person. This combined approach of Face Recognition with Voice Recognition performs better than previous approaches and the experiments we perform give results with 98.99% accuracy by using number of images. All the information will be send to mobile phone of family members by using MATLAB tool and Internet Dongle. [ABSTRACT FROM AUTHOR]