Face recognition using fractional coefficients and discrete cosine transform tool

Face recognition is a computer vision application based on biometric information for automatic person identification or verification from image sequence video frame. In this context DCT the easy technique to determine significant parameters. Until now main object selection of coefficients obtain best recognition. Many techniques rely premasking windows discard high and low enhance performance. However, problem resides in shape size premask. To improve discriminator ability discrete cosine transform domain, we used fractional transformed images with limit area better performance system. Then selected bands, use discrimination power analysis search having highest discriminate different classes each other. Feature algorithm key issue all pattern system, fact utilized define features vector among several ones, where these are according specified criterion. classifiers evaluate our approach like, support machine random forests. The proposed validated Yale ORL databases. Experimental results prove sufficiency method face facial expression field.