Standoff and mobile fingerprint collection
View Patent ↗A method, a biometric identifier collection device, and a set of instructions are disclosed. A memory 208 may store a digital image having a biometric identifier. A processor 204 may execute a lighting compensation to remove a lighting effect from the biometric identifier. The processor 204 may process the biometric identifier to create an identification profile.
1. A method, comprising:
determining that a digital image contains a biometric identifier;
dividing the digital image into a grid of blocks;
computing a ridge orientation for each block containing a portion of the biometric identifier in the grid of blocks;
determining a bad block in the grid of blocks by applying a bad block heuristic, the bad block heuristic comparing the computed ridge orientation for each block containing the portion of the biometric identifier in the grid of blocks as a target block to the ridge orientation for each block directly adjacent to the target block and identifying the target block as a bad block if a deviation between the ridge orientation of the target block and ridge orientations of the directly adjacent blocks exceeds a certain threshold;
executing lighting compensation to remove a lighting effect from the biometric identifier, the lighting compensation being a piece-wise linear transformation of a light intensity for each block of the digital image that calculates an intensity height field of the each block in the grid of blocks and fits a separate plane to the calculated intensity height field for the each block;
processing the biometric identifier ignoring the identified bad blocks and after the lighting compensation is executed to create an identification profile; and
outputting the created identification profile in a format usable by a user.
2. The method of claim 1 , wherein the biometric identifier is a fingerprint.
3. The method of claim 1 , further comprising:
isolating the biometric identifier in the digital based on an analysis of the each block in the grid of blocks.
4. The method of claim 1 , further comprising:
applying a first iteration of a Gabor filter to the digital image.
5. The method of claim 4 , further comprising:
automatically computing a Gabor parameter controlling wavelength for use by the Gabor filter.
6. The method of claim 4 , further comprising:
applying multiple iterations of the Gabor filter to the digital image.
7. The method of claim 1 , the ridge orientation being computed by applying the following equations:
Δ
x
=
∑
i
=
1
W
∑
j
=
1
W
2
G
x
(
ⅈ
,
j
)
G
y
(
ⅈ
,
j
)
,
Δ
y
=
∑
i
=
1
W
∑
j
=
1
W
(
G
x
2
(
ⅈ
,
j
)
-
G
y
2
(
ⅈ
,
j
)
)
,
θ
=
1
2
tan
-
1
(
Δ
x
Δ
y
)
,
where G x and G y are horizontal (X) and vertical (Y) gradients respectively in block (i,j), W is a size of each block (i,j) along each axis, and θ is the computed ridge orientation for the each block (i,j).
8. The method of claim 7 , the bad block heuristic identifying the target block as the bad block by applying the following equation:
V
>
1
N
∑
(
i
,
j
)
∈
N
θ
(
i
,
j
)
-
θ
(
u
,
v
)
2
where N is a set of the directly adjacent blocks to the target block at location (u,v) and V is the certain threshold.
9. A biometric identifier collection device, comprising:
a memory that stores a digital image having a biometric identifier;
a processor that is programmed to:
divide the digital image into a grid of blocks,
compute a ridge orientation for each block containing a portion of the biometric identifier in the grid of blocks,
determine a bad block in the grid of blocks by applying a bad block heuristic, the bad block heuristic comparing the computed ridge orientation for each block containing the portion of the biometric identifier in the grid of blocks as a target block to the ridge orientation for each block directly adjacent to the target block and identifying the target block as a bad block if a deviation between the ridge orientation of the target block and ridge orientations of the directly adjacent blocks exceeds a certain threshold,
execute a lighting compensation to remove a lighting effect from the biometric identifier, the lighting compensation being a piece-wise linear transformation of a light intensity for each block of the digital image that calculates an intensity height field of the each block in the grid of blocks and fits a separate plane to the calculated intensity height field for the each block, and
process the biometric identifier ignoring the identified bad blocks and after the lighting compensation is executed to create an identification profile; and
an output device that outputs the created identification profile in a format usable by a user.
10. The biometric identifier collection device of claim 9 , further comprising:
a digital camera component that generates the digital image.
11. The biometric identifier collection device of claim 9 , the processor being further programmed to automatically compute a Gabor parameter controlling wavelength and to apply multiple iterations of a Gabor filter to the digital image.
12. The biometric identifier collection device of claim 9 , the ridge orientation being computed by applying the following equations:
Δ
x
=
∑
i
=
1
W
∑
j
=
1
W
2
G
x
(
ⅈ
,
j
)
G
y
(
ⅈ
,
j
)
,
Δ
y
=
∑
i
=
1
W
∑
j
=
1
W
(
G
x
2
(
ⅈ
,
j
)
-
G
y
2
(
ⅈ
,
j
)
)
,
θ
=
1
2
tan
-
1
(
Δ
x
Δ
y
)
,
where G x and G y are horizontal (X) and vertical (Y) gradients respectively in block (i,j), W is a size of each block (i,j) along each axis, and θ is the computed ridge orientation for the each block (i,j).
13. The biometric identifier collection device of claim 12 , the bad block heuristic identifying the target block as the bad block by applying the following equation:
V
>
1
N
∑
(
i
,
j
)
∈
N
θ
(
i
,
j
)
-
θ
(
u
,
v
)
2
where N is a set of the directly adjacent blocks to the target block at location (u,v) and V is the certain threshold.
14. A non-transitory storage medium on which is recorded a set of instructions executable by a processor to implement a method for processing digital images, the method comprising:
determining that a digital image contains a biometric identifier;
dividing the digital image into a grid of blocks;
computing a ridge orientation for each block containing a portion of the biometric identifier in the grid of blocks;
determining a bad block in the grid of blocks by applying a bad block heuristic, the bad block heuristic comparing the computed ridge orientation for each block containing the portion of the biometric identifier in the grid of blocks as a target block to the ridge orientation for each block directly adjacent to the target block and identifying the target block as a bad block if a deviation between the ridge orientation of the target block and ridge orientations of the directly adjacent blocks exceeds a certain threshold;
executing lighting compensation to remove a lighting effect from the biometric identifier, the lighting compensation being a piece-wise linear transformation of a light intensity for each block of the digital image that calculates an intensity height field of the each block in the grid of blocks and fits a separate plane to the calculated intensity height field for the each block; and
processing the biometric identifier ignoring the identified bad blocks and after the lighting compensation is executed to create an identification profile; and
outputting the created identification profile in a format usable by a user.
15. The non-transitory storage medium of claim 14 , the method further comprising:
automatically computing a Gabor parameter controlling wavelength; and
applying multiple iterations of a Gabor filter to the digital image.
16. The non-transitory storage medium of claim 14 , the ridge orientation being computed by applying the following equations:
Δ
x
=
∑
i
=
1
W
∑
j
=
1
W
2
G
x
(
ⅈ
,
j
)
G
y
(
ⅈ
,
j
)
,
Δ
y
=
∑
i
=
1
W
∑
j
=
1
W
(
G
x
2
(
ⅈ
,
j
)
-
G
y
2
(
ⅈ
,
j
)
)
,
θ
=
1
2
tan
-
1
(
Δ
x
Δ
y
)
,
where G x and G y are horizontal (X) and vertical (Y) gradients respectively in block (i,j), W is a size of each block (i,j) along each axis, and θ is the computed ridge orientation for the each block (i,j).
17. The non-transitory storage medium of claim 16 , the bad block heuristic identifying the target block as the bad block by applying the following equation:
V
>
1
N
∑
(
i
,
j
)
∈
N
θ
(
i
,
j
)
-
θ
(
u
,
v
)
2
where N is a set of the directly adjacent blocks to the target block at location (u,v) and V is the certain threshold.