Method of validation of the use of a real finger as support of a fingerprint
View Patent ↗Validation of using a finger as support of a fingerprint may include: placing the support such that it bears; capturing a captured image of the print; transforming the captured image into a resulting image by passage through a low-pass filter; locating on the resulting image an origin point, the intensity of whose pixel is representative of the maximum pressure exerted on the support; verifying that, on the resulting image, for a plurality of radii issuing from the origin point, and for each of said radii, for a plurality of points M, the intensity of the pixel of each point M of said radius is representative of a decline in the pressure exerted on the support as the distance from the origin point to the point M increases, and a decision as regards the validity of the support is taken as a function of the results of the verification step.
1. Method of validation of the use of a real finger as support of a fingerprint, implemented by a validation device comprising a capture surface on which said support bears, a sensor intended to capture an image of the fingerprint, a processing module and a decision taking module, said method of validation comprising:
a positioning step during which the support is placed such that it bears against the capture surface,
a capture step during which the sensor captures a so-called captured image of the print,
a filtering step during which the processing module transforms the captured image into a resulting image by passage through a low-pass filter whose cutoff frequency is much lower than the frequency of the ridges of a print,
a locating step during which the processing module locates on the resulting image, an origin point O, the intensity of whose pixel is representative of the maximum pressure exerted on the support,
a verification step during which the processing module verifies that, on the resulting image, for a plurality of radii issuing from the origin point O, and for each of said radii, for a plurality of points M, the intensity of the pixel of each point M of said radius is representative of a decline in the pressure exerted on the support as the distance from the origin point O to the point M increases, and
a decision taking step during which the decision taking module takes a decision as regards the validity of the support as a function of the results of the verification step.
2. Method of validation according to claim 1 , wherein that said cutoff frequency is of the order of from 0.1 to 1 cycle per mm.
3. Method of validation according to claim 1 , wherein the verification step consists in verifying that starting from the origin point O and on each radius issuing from the origin point O, the intensity gradient of each point M is negative.
4. Method of validation according to claim 3 , wherein the decision taking step is based on a cumulative criterion denoted Dism(P,O) and representing the deviation of the intensity profile of the actual intensity profile P from the theoretical model of intensity centered at O of a real finger and on a comparison of this deviation Dism(P,O) with respect to a threshold.
5. Method of validation according to claim 4 , wherein the deviation Dism(P,O) is given by the formula:
Dism
(
P
,
O
)
=
1
Area
(
P
)
∫
0
2
π
∫
0
R
ma
x
L
(
∇
r
O
P
(
r
,
θ
)
)
·
ⅆ
r
·
ⅆ
θ
(
3
)
where:
∇ r O P(r,θ) is the projection on the radial local basis at M relating to O of the intensity gradient at the point M of the resulting image with polar coordinates (r,θ),
L is a zero function on − and an increasing function on + *,
R max is the maximum distance between the origin point O and any point M of the resulting image, and
Area(P) is the area of the region considered around the point M.
6. Method of validation according to claim 1 , wherein it comprises a preprocessing step intended to improve the rendition of the resulting image, between the capture step and the filtering step.
7. Method of validation according to claim 6 , wherein the preprocessing step consists in applying to the pixels of the captured image, a function F defined by the formula:
F
(
x
,
y
)
=
(
255
-
p
(
x
,
y
)
)
·
R
r
(
x
,
y
)
R
ir
(
x
,
y
)
(
7
)
where when the pixel (x, y) does not belong to the pixels of the ridges, R r is equal to the average value of the local widths of adjacent ridges on either side of the valley including the pixel, at the level of said pixel, and where when the pixel belongs to the pixels of the ridges, R r equals the local width of the ridge in this pixel,
where when the pixel does not belong to the pixels of the valleys, R ir is equal to the average value of the local widths of adjacent valleys on either side of the ridge including this pixel, at the level of said pixel, and where when the pixel belongs to the pixels of the valleys, R ir equals the local width of the valley in this pixel.
8. Method of validation according to claim 6 , wherein the preprocessing step consists in applying to the pixels of the captured image, a function F defined by the formula:
F
(
x
,
y
)
=
(
255
-
p
(
x
,
y
)
)
·
R
r
(
x
,
y
)
R
ir
(
x
,
y
)
(
7
)
where when the pixel does not belong to the skeleton of the ridges, R r is zero, and where when the pixel (x, y) belongs to the skeleton of the ridges, R r equals the local width of the ridge in this pixel,
where when the pixel does not belong to the skeleton of the valleys, R ir is zero, and where when the pixel belongs to the skeleton of the valleys, R ir equals the local width of the valley in this pixel.
9. Method of validation according to claim 6 , wherein the preprocessing step consists in applying to the pixels of the captured image, a function F defined by the formula:
F
(
x
,
y
)
=
(
255
-
p
(
x
,
y
)
)
·
R
r
(
x
,
y
)
R
ir
(
x
,
y
)
(
7
)
where R r is the number of pixels of ridges in a zone centered on the pixel, and
where R ir is the number of pixels of valleys in a zone centered on the pixel.
10. Method of validation according to claim 1 , wherein the locating step consists in choosing the origin point O as the point of the resulting image exhibiting the highest intensity.
11. Validation device comprising:
a capture surface on which said support bears,
a sensor intended to capture an image of the fingerprint,
a processor configured to:
transform the captured image into a resulting image by passage through a low-pass filter whose cutoff frequency is much lower than the frequency of the ridges of a print,
locate on the resulting image, an origin point O, the intensity of whose pixel is representative of the maximum pressure exerted on the support,
verify that, on the resulting image, for a plurality of radii issuing from the origin point O, and for each of said radii, for a plurality of points M, the intensity of the pixel of each point M of said radius is representative of a decline in the pressure exerted on the support as the distance from the origin point O to the point M increases, and
take a decision as regards the validity of the support as a function of the results transmitted by the processor.