Skip to Main Content
Faces represent complex, multidimensional, meaningful visual stimuli and developing a computational model for face recognition is difficult. We present an unsupervised neural network solution which compares favorably with other methods. The system combines local image sampling, a self-organizing map neural network. In this process, the images for the different persons will be scanned and it will be used as a data base. The scanned images will be resized according to the SOM architecture. The database consists of images which are used for the passports. The features obtained from the scanned images will be used in the training process. The Linear Architecture of SOM will be trained for the given data set for the different parameters such as Learning Rate, Neighborhood and Output Neurons. Several experiments will be conducted to test the recognition rate, and proper architecture for the said pattern classification task will be suggested.