Method for forgery recognition in fingerprint recognition by using a texture classification of gray scale differential images
View Patent ↗In a method for forgery recognition in the fingerprint recognition, at first an image sequence is captured, which comprises a plurality of images of the fingerprint to be recognized. From at least two images capturing a texture captures a differential image. The texture will be compared with a predetermined texture in order to determine whether the captured image sequence comes from a real finger or from an imitation.
1. A method for forgery recognition, comprising:
(a) capturing an image sequence comprising a plurality of images of a fingerprint to be recognized;
(b) forming of at least one differential image from two images of the captured image sequence;
(c) capturing a texture in the at least one differential image;
(d) comparing the texture with a predetermined texture to determine whether the captured image sequence comes from a real finger or from an imitation.
2. The method according to claim 1 , wherein the image sequence is generated during placing of a finger on a sensor unit and comprises images that are consecutive in time.
3. The method according to claim 1 , wherein the images of the captured image sequence are captured as gray scale images.
4. The method according to claim 1 , wherein a plurality of differential images is generated in step (b) by subtracting two consecutive images of the image sequence, respectively, images to be subtracted are those images that are arranged in the captured image sequence at the time of the first contact between finger and sensor unit.
5. The method according to claim 1 , wherein in step (d) the texture captured in step (c) is compared with a first texture for a real finger and with a second texture for an imitation in order to classify the captured texture.
6. The method according to claim 1 , wherein the step (d) further comprises:
(d1) filtering a gray scale differential image to obtain a gradient image,
(d2) calculating a histogram of the gradient image;
(d3) correlating the calculated histogram with a first given mask for a real finger and with a second given mask for an imitation, and
(d4) classifying the captured image sequence as coming from a real finger if the value of the correlation with the first mask is higher than the value of the correlation with the second mask.
7. The method according to claim 6 , wherein the steps (d1) and (d2) are carried out for a plurality of differential images wherein the method comprises the following steps before the steps (d3) and (d4):
calculating a center of gravity in each histogram,
comparing the center of gravity with a threshold, and
passing on those histograms, whose center of gravity exceeds the threshold, to the steps (d3) and (d4).