Cubic Bézier approximation of a digitized curve
Pattern Recognition
| Hi-index | 0.00 |
This paper proposes an efficient face recognitionscheme, which incorporates certain methods. In thisscheme, edges are obtained from sobelled face images.The major feature areas are segmented and principalcurves are obtained by applying thinning algorithm onthe features of resulting face image. A modifiedthinning algorithm based on line sweep procedures isused. A line sweep is a process where the plane isdivided into parallel slabs by lines passing throughcertain "events" and then items are processedaccording to an order of the slabs [A Line Sweep thinning Algorithm]. A new methodfor Bezier contour point interpolation is used afterlocating the two end points of the curve on the primefacial features. The two control points of a Beziercurve are interpolated by this method. The Bezierpoints i.e. the two end points of a curve and the twocontrol points, form the line segments whose Hausdroffdistances are calculated as specified in the algorithmdiscussed in this paper. The recognition rate is 100%on normal faces and also on faces with some simpleexpressions.