Automatic image processing for document de-skewing and cropping
View Patent ↗An image of a rectangular target is resolved. First and second dimensions for the rectangular target are determined from an initial image. A cropped and de-skewed final image for the rectangular target is produced responsive to the first and second dimensions.
1. A method for providing an image of a rectangular target object, comprising the steps of:
determining only a first dimension of a rectangular target object in an initial image, wherein the determining the first dimension further includes displaying a scanned image of the target object on a user display on a user interface, and via the user interface, providing a user with an option to select a first and adjacent corner point of the target object in the initial image; and subsequently determining a line between the first and adjacent corner points, a length of said line indicating the first dimension;
subsequently determining only a second dimension of the target object;
subsequently providing a cropped and de-skewed final image of the target object responsive to the determined first and second dimension; and
determining a de-skew angle θ according to
θ
=
tan
-
1
(
Δ
y
Δ
x
)
=
tan
-
1
(
y
1
-
y
0
x
1
-
x
0
)
where y 1 is a y co-ordinate at the first corner point, y 0 is a y co-ordinate at an adjacent corner point, x 1 is an x co-ordinate at the first corner point and x 0 is an x co-ordinate at the adjacent corner point.
2. The method as claimed in claim 1 , further comprising:
subsequent to selection of the first corner point, as a user moves an indicator over the displayed image, displaying a visual cue comprising a line displayed on the user display between the first corner point and a position of the indicator.
3. The method as claimed in claim 1 wherein the step of determining the second dimension comprises:
via the user interface, providing the user with an option to move an indicator on the user display away from the adjacent corner point and determining a perpendicular distance from a final position of the indicator selected by the user to a line passing through the first and adjacent corner point, said perpendicular distance comprising said second dimension.
4. The method as claimed in claim 1 , further comprising:
scanning a check on a flatbed scanner thereby providing scanned image data; and
displaying the scanned image responsive to the scanned image data.
5. The method as claimed in claim 1 , further comprising:
selecting the first and adjacent corner points by a user touching a touch screen display displaying the initial image where a first and adjacent corner of the target object are displayed or by a user moving a computer mouse to locate a cursor displayed on the user display displaying the initial image where the first and adjacent corner of the target object are displayed.
6. A method for providing an image of a rectangular target object, comprising the steps of:
determining only a first dimension of a rectangular target object in an initial image, wherein the determining the first dimension further includes displaying a scanned image of the target object on a user display on a user interface, and via the user interface, providing a user with an option to select a first and adjacent corner point of the target object in the initial image; and subsequently determining a line between the first and adjacent corner points, a length of said line indicating the first dimension;
subsequently determining only a second dimension of the target object, via the user interface, providing the user with an option to move an indicator on the user display away from the adjacent corner point and determining a perpendicular distance from a final position of the indicator selected by the user to a line passing through the first and adjacent corner point, said perpendicular distance comprising said second dimension;
subsequently providing a cropped and de-skewed final image of the target object responsive to the determined first and second dimension;
said step of determining the second dimension comprises determining h according to
h
=
a
×
b
a
=
x
2
(
y
1
-
y
0
)
+
y
2
(
x
1
-
x
0
)
+
x
0
y
1
-
x
1
y
0
(
x
1
-
x
0
)
2
+
(
y
1
-
y
0
)
2
where x 2 is an x co-ordinate at the final position, x 1 is an x co-ordinate at a first corner point, x 0 is an x co-ordinate at an adjacent corner point, y 2 is a y co-ordinate at the final position, y 1 is a y co-ordinate at a first corner point and y 0 is a y co-ordinate at the adjacent corner point.
7. Apparatus for providing an image of a rectangular target object, comprising:
a user interface comprising a user display and at least one user response device for selecting a desired location on the user display;
a processor that determines only a first dimension of a rectangular target object in an initial image, subsequently determines only a second dimension of the target object, and subsequently displays a cropped and de-skewed final image of the target object on the user display responsive to the first and second dimension;
wherein the processor determines a first dimension of the target object by determining a distance between a first and adjacent corner point selected by a user and determines a second dimension of the target by determining a distance h where
h
=
a
×
b
a
=
x
2
(
y
1
-
y
0
)
+
y
2
(
x
1
-
x
0
)
+
x
0
y
1
-
x
1
y
0
(
x
1
-
x
0
)
2
+
(
y
1
-
y
0
)
2
where x 2 is an x co-ordinate at the final position, x 1 is an x co-ordinate at a first corner point, x 0 is an x co-ordinate at an adjacent corner point, y 2 is a y co-ordinate at the final position, y 1 is a v co-ordinate at a first corner point and y 0 is a y co-ordinate at the adjacent corner point.
8. The apparatus as claimed in claim 7 wherein the processor determines a de-skew angle θ according to
θ
=
tan
-
1
(
Δ
y
Δ
x
)
=
tan
-
1
(
y
1
-
y
0
x
1
-
x
0
)
where y 1 is a y co-ordinate at the first corner point, y 0 is a y co-ordinate at an adjacent corner point, x1 is an x co-ordinate at the first corner point and x 0 is an x co-ordinate at the adjacent corner point.