STANDARD FACE GRAPH GENERATION METHOD WHICH INCREASES FACE RECOGNITION PERFORMANCE BY GENERATING A STANDARD FACE GRAPH BASED ON A GABOR JET

摘要

PURPOSE: A standard face graph generation method is provided to calculate an average Gabor jet and deviation between the average location information of a reference point of an exercised face image and the average location information.;CONSTITUTION: A reference point selection module selects a plurality of reference points in selected training face images(S20). A Gabor Jet extraction module calculates an average Gabor Jet(S40). A deviation extraction module calculates deviation of the average location information of the reference point and the average location information(S50). A deviation change type analysis module analyzes variation type of the deviation(S60). A standard face graph generation module generates a standard face graph(S70).;COPYRIGHT KIPO 2013;[Reference numerals] (AA) Start; (BB) End; (S10) Selecting image of training face; (S20) Setting reference point; (S30) Normalization; (S40) Average gabor jet extraction; (S50) Extracting average location information and deviation; (S60) Analyzing type of average deviation change; (S70) Generating standard face graph