U.S. patent application number 16/551375 was filed with the patent office on 2019-12-12 for image processing method, device and photographic apparatus.
The applicant listed for this patent is SZ DJI TECHNOLOGY CO., LTD.. Invention is credited to Ketan TANG, Guyue ZHOU.
Application Number | 20190378250 16/551375 |
Document ID | / |
Family ID | 56106428 |
Filed Date | 2019-12-12 |
![](/patent/app/20190378250/US20190378250A1-20191212-D00000.png)
![](/patent/app/20190378250/US20190378250A1-20191212-D00001.png)
![](/patent/app/20190378250/US20190378250A1-20191212-D00002.png)
![](/patent/app/20190378250/US20190378250A1-20191212-D00003.png)
![](/patent/app/20190378250/US20190378250A1-20191212-D00004.png)
![](/patent/app/20190378250/US20190378250A1-20191212-D00005.png)
![](/patent/app/20190378250/US20190378250A1-20191212-D00006.png)
![](/patent/app/20190378250/US20190378250A1-20191212-M00001.png)
![](/patent/app/20190378250/US20190378250A1-20191212-M00002.png)
![](/patent/app/20190378250/US20190378250A1-20191212-M00003.png)
![](/patent/app/20190378250/US20190378250A1-20191212-M00004.png)
View All Diagrams
United States Patent
Application |
20190378250 |
Kind Code |
A1 |
ZHOU; Guyue ; et
al. |
December 12, 2019 |
IMAGE PROCESSING METHOD, DEVICE AND PHOTOGRAPHIC APPARATUS
Abstract
An image processing method includes correcting a target image
based on an initial distortion coefficient to obtain a first
corrected target image, performing straight-line fitting on a first
border line in the first corrected target image to calculate a
first distortion metric value and a correction distortion
coefficient, correcting the target image based on the correction
distortion coefficient to obtain a second corrected target image,
removing outlier points on a second border line in the second
corrected target image, performing straight-line fitting on the
second border line with the outlier points removed to calculate a
second distortion metric value, detecting whether a preset
correction condition is satisfied based on at least one of the
first distortion metric value or the second distortion metric
value, and, if the preset correction condition is satisfied,
applying the correction distortion coefficient to subsequent image
correction to obtain better corrected images.
Inventors: |
ZHOU; Guyue; (Shenzhen,
CN) ; TANG; Ketan; (Shenzhen, CN) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
SZ DJI TECHNOLOGY CO., LTD. |
Shenzhen |
|
CN |
|
|
Family ID: |
56106428 |
Appl. No.: |
16/551375 |
Filed: |
August 26, 2019 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
15617488 |
Jun 8, 2017 |
10417750 |
|
|
16551375 |
|
|
|
|
PCT/CN2014/093389 |
Dec 9, 2014 |
|
|
|
15617488 |
|
|
|
|
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
G06T 3/4007 20130101;
G06T 7/13 20170101; G06T 5/006 20130101; G06T 5/005 20130101; G06T
2207/10016 20130101 |
International
Class: |
G06T 5/00 20060101
G06T005/00; G06T 7/13 20060101 G06T007/13 |
Claims
1. An image processing method comprising: correcting a target image
based on an initial distortion coefficient to obtain a first
corrected target image; performing straight-line fitting on a first
border line in the first corrected target image to calculate a
first distortion metric value and a correction distortion
coefficient; correcting the target image based on the correction
distortion coefficient to obtain a second corrected target image;
removing outlier points on a second border line in the second
corrected target image; performing straight-line fitting on the
second border line with the outlier points removed to calculate a
second distortion metric value; detecting whether a preset
correction condition is satisfied based on at least one of the
first distortion metric value or the second distortion metric
value; and applying, if the preset correction condition is
satisfied, the correction distortion coefficient to subsequent
image correction to obtain better corrected images.
2. The method according to claim 1, further comprising, before
correcting the target image: capturing an image of an object
including straight line features; and using the captured image as
the target image or adjusting a size of the captured image to
obtain the target image.
3. The method according to claim 2, wherein capturing the image of
the object including straight line features includes: capturing a
plurality of images; and analyzing the plurality of images to
determine the image of the object including straight line
features.
4. The method according to claim 3, wherein analyzing the plurality
of images includes analyzing the plurality of images at a same time
using a same processing method.
5. The method according to claim 2, wherein adjusting the size of
the captured image includes: magnifying, if the size of the
captured image is smaller than a preset size threshold, the
captured image to a target size through interpolation; or scaling
down, if the size of the captured image is greater than the preset
size threshold, the captured image to the target size through
down-sampling.
6. The method according to claim 1, wherein performing
straight-line fitting on the first border line to calculate the
first distortion metric value and the correction distortion
coefficient includes: performing edge detection on the first
corrected target image to determine the first border line in the
first corrected target image; performing straight-line fitting on
the first border line based on polynomial straight-line fitting to
obtain a fitted straight line; and calculating the first distortion
metric value of the first border line relative to the fitted
straight line and the correction distortion coefficient
corresponding to the first distortion metric value.
7. The method according to claim 6, wherein calculating the first
distortion metric value and the correction distortion coefficient
includes: determining a straight line segment in the first border
line; calculating distances from corresponding points on the
straight line segment to the fitted straight line; obtaining the
first distortion metric value according to the distances; and
performing non-linear optimization on the first distortion metric
value to obtain the correction distortion coefficient.
8. The method according to claim 1, wherein: correcting the target
image according to the initial distortion coefficient includes
correcting a target border line in the target image based on the
initial distortion coefficient, and correcting the target image
based on the correction distortion coefficient includes correcting
the target border line in the target image based on the correction
distortion coefficient.
9. The method according to claim 1, wherein performing
straight-line fitting on the second border line with the outlier
points removed to calculate the second distortion metric value
includes: performing edge detection on the second corrected target
image to determine the second border line in the second corrected
target image; performing straight-line fitting on the second border
line based on polynomial straight-line fitting to obtain a fitted
straight line; and calculating the second distortion metric value
of the second border line relative to the fitted straight line.
10. The method according to claim 9, wherein calculating the second
distortion metric value includes: determining a straight line
segment in the second border line with the outlier points removed;
calculating distances from corresponding points on the straight
line segment to the fitted straight line; and obtaining the second
distortion metric value according to the distances.
11. The method according to claim 1, wherein detecting whether the
preset correction condition is satisfied includes: calculating a
relative variation amount between the first distortion metric value
and the second distortion metric value; and determining whether the
relative variation amount calculated is smaller than a preset
variation threshold to determine whether the preset correction
condition is satisfied.
12. The method according to claim 1, wherein detecting whether the
preset correction condition is satisfied includes determining
whether the second distortion metric value is smaller than a preset
metric threshold.
13. The method according to claim 1, further comprising:
configuring, if the preset correction condition is not satisfied,
the correction distortion coefficient as the initial distortion
coefficient.
14. A camera comprising: a camera lens; and an image processor
configured to: correct a target image based on an initial
distortion coefficient to obtain a first corrected target image;
perform straight-line fitting on a first border line in the first
corrected target image to calculate a first distortion metric value
and a correction distortion coefficient; correct the target image
based on the correction distortion coefficient to obtain a second
corrected target image; remove outlier points on a second border
line in the second corrected target image; perform straight-line
fitting on the second border line with the outlier points removed
to calculate a second distortion metric value; detect whether a
preset correction condition is satisfied based on at least one of
the first distortion metric value or the second distortion metric
value; and apply, if the preset correction condition is satisfied,
the correction distortion coefficient to subsequent image
correction to obtain better corrected images.
15. The camera according to claim 14, wherein the image processor
is further configured to: capture an image of an object including
straight line features through the camera lens; and determine the
captured image as the target image or adjust a size of the captured
image to obtain the target image.
16. The camera according to claim 15, wherein the image processor
is further configured to: capture a plurality of images; and
analyze the plurality of images to determine the image of the
object including straight line features.
17. The camera according to claim 15, wherein the image processor
is further configured to analyze the plurality of images at a same
time using a same processing method.
18. The camera according to claim 15, wherein the image processor
is further configured to: magnify, if the size of the captured
image is smaller than a preset size threshold, the captured image
to a target size through interpolation; or scale down, if the size
of the captured image is greater than the preset size threshold,
the captured image to the target size through down-sampling.
19. The camera according to claim 14, wherein the image processor
is further configured to: perform edge detection on the second
corrected target image to determine the second border line in the
second corrected target image; perform straight-line fitting on the
second border line based on polynomial straight-line fitting to
obtain a fitted straight line; and calculate the second distortion
metric value of the second border line relative to the fitted
straight line.
20. The camera according to claim 19, wherein the image processor
is further configured to: determine a straight line segment in the
second border line with the outlier points removed; calculate
distances from corresponding points on the straight line segment to
the fitted straight line; and obtain the second distortion metric
value according to the distances.
Description
CROSS-REFERENCE TO RELATED APPLICATIONS
[0001] This application is a continuation of application Ser. No.
15/617,488, filed on Jun. 8, 2017, which is a continuation
application of International Application No. PCT/CN2014/093389,
filed on Dec. 9, 2014, the entire contents of both of which are
incorporated herein by reference.
FIELD OF THE DISCLOSURE
[0002] The present disclosure relates to the technical field of
image processing, and in particular, to an image processing method,
device, and photographic apparatus.
BACKGROUND
[0003] As wide-angle lenses have been more and more widely used,
especially with their use on aerial shooting apparatuses, camera
distortion caused by the wide-angle lenses has attracted more and
more attention. If correction is not made, images or videos taken
by the wide-angle lenses may have a serious barrel distortion. For
example, when a sport court is shot, straight lines sprayed on the
court may be distorted and appear as curved lines. Therefore, a
camera may need to be calibrated to obtain a distortion coefficient
thereof to correct the images or videos taken by the camera.
[0004] Camera distortion generally includes radial distortion and
tangential distortion. For a wide-angle lens, a fourth-order
polynomial radial distortion model has been proved to be
sufficient. Distortion equations of a wide-angle lens are:
x.sup.d=x.sup.u(1+k.sub.1r.sup.2+k.sub.2r.sup.4) (1)
y.sup.d=y.sup.u(1+k.sub.1r.sup.2+k.sub.2r.sup.4), (2)
where (x.sup.u, y.sup.u) denote coordinates before distortion, also
referred to as "non-distorted coordinates," and (x.sup.u, y.sup.u)
denote corresponding coordinates after distortion, also referred to
as "distorted coordinates." A distortion center can be represented
by its coordinates (c.sub.x, c.sub.y), and the parameter r in
Equations (1) and (2) can be calculated using r= {square root over
((x.sup.u-c.sub.x).sup.2+(y.sup.u-c.sub.y).sup.2.)} The purpose of
the calibration is to determine distortion coefficients k
=(k.sub.1, k.sub.2) and the distortion center c=(c.sub.x,
c.sub.y).
[0005] Usually, the distortion center (c.sub.x, c.sub.y) in an
image is close to a central point
( w 2 , h 2 ) ##EQU00001##
of the image, where w and h are the width and the height of the
image, respectively. The central point of the image can be used as
an initial value of the distortion center, and an optimal solution
to the distortion center can be obtained by a small number of times
of iteration. However, it may be relatively more difficult to
calculate the distortion coefficients.
[0006] In conventional technologies, a lens can be calibrated using
a calibration board. In particular, a series of images or videos of
the calibration board can be taken, and then internal references
and distortion coefficients can be calculated according to
geometric constraints. After the distortion coefficients are
obtained, the shot videos or images can be corrected. Calibration
using the calibration board can be relatively accurate as long as a
high-accuracy calibration board is used. This is a barrier to
ordinary users when performing calibration.
SUMMARY OF THE DISCLOSURE
[0007] Embodiments of the present disclosure provide an image
processing method, device, and video camera, which can easily and
rapidly determine a distortion coefficient to process an image.
[0008] In accordance with the disclosure, there is provided an
image processing method including correcting a target image based
on an initial distortion coefficient to obtain a first corrected
target image, performing straight-line fitting on a first border
line in the first corrected target image to calculate a first
distortion metric value and a correction distortion coefficient,
correcting the target image based on the correction distortion
coefficient to obtain a second corrected target image, performing
straight-line fitting on a second border line in the second
corrected target image to calculate a second distortion metric
value, detecting whether a preset correction condition is satisfied
based on at least one of the first distortion metric value or the
second distortion metric value, and configuring the correction
distortion coefficient as the initial distortion coefficient if the
preset correction condition is not satisfied.
[0009] In some embodiments, the method further includes, before
correcting the target image, capturing an image of an object
including straight line features and adjusting a size of the
captured image to obtain the target image.
[0010] In some embodiments, adjusting the size of the captured
image includes magnifying, if the size of the captured image is
smaller than a preset size threshold, the captured image to a
target size through interpolation, or scaling down, if the size of
the captured image is greater than the preset size threshold, the
captured image to the target size through down-sampling.
[0011] In some embodiments, performing straight-line fitting on the
first border line to calculate the first distortion metric value
and the correction distortion coefficient includes performing edge
detection on the first corrected target image to determine the
first border line in the first corrected target image, performing
straight-line fitting on the first border line based on polynomial
straight-line fitting to obtain a fitted straight line, and
calculating the first distortion metric value of the first border
line relative to the fitted straight line and the correction
distortion coefficient corresponding to the first distortion metric
value.
[0012] In some embodiments, calculating the first distortion metric
value and the correction distortion coefficient includes
determining a straight line segment in the first border line,
calculating distances from corresponding points on the straight
line segment to the fitted straight line, obtaining the first
distortion metric value according to the distances, and performing
non-linear optimization on the first distortion metric value to
obtain the correction distortion coefficient.
[0013] In some embodiments, correcting the target image according
to the initial distortion coefficient includes correcting a target
border line in the target image based on the initial distortion
coefficient, and correcting the target image based on the
correction distortion coefficient includes correcting the target
border line in the target image based on the correction distortion
coefficient.
[0014] In some embodiments, performing straight-line fitting on the
second border line to calculate the second distortion metric value
includes performing edge detection on the second corrected target
image to determine the second border line in the second corrected
target image, performing straight-line fitting on the second border
line based on polynomial straight-line fitting to obtain a fitted
straight line, and calculating the second distortion metric value
of the second border line relative to the fitted straight line.
[0015] In some embodiments, calculating the second distortion
metric value includes, removing outliers, determining a straight
line segment in the second border line, calculating distances from
corresponding points on the straight line segment to the fitted
straight line, and obtaining the second distortion metric value
according to the distances.
[0016] In some embodiments, detecting whether the preset correction
condition is satisfied includes calculating a relative variation
amount between the first distortion metric value and the second
distortion metric value, and determining whether the relative
variation amount calculated is smaller than a preset variation
threshold to determine whether the preset correction condition is
satisfied.
[0017] In some embodiments, the method further including performing
image correction based on the correction distortion coefficient if
the preset correction condition is satisfied.
[0018] Also in accordance with the disclosure, there is provided a
camera including a camera lens and an image processor. The image
processor is configured to correct a target image based on an
initial distortion coefficient to obtain a first corrected target
image, perform straight-line fitting on a first border line in the
first corrected target image to calculate a first distortion metric
value and a correction distortion coefficient, correct the target
image based on the correction distortion coefficient to obtain a
second corrected target image, perform straight-line fitting on a
second border line in the second corrected target image to
calculate a second distortion metric value, detect whether a preset
correction condition is satisfied based on at least one of the
first distortion metric value or the second distortion metric
value, and configure the correction distortion coefficient as the
initial distortion coefficient if the preset correction condition
is not satisfied.
[0019] In some embodiments, the image processor is further
configured to capture an image of an object including straight line
features through the camera lens, and adjust a size of the captured
image to obtain the target image.
[0020] In some embodiments, the image processor is further
configured to magnify, if the size of the captured image is smaller
than a preset size threshold, the captured image to a target size
through interpolation, or scale down, if the size of the captured
image is greater than the preset size threshold, the captured image
to the target size through down-sampling.
[0021] In some embodiments, the image processor is further
configured to perform edge detection on the first corrected target
image to determine the first border line in the first corrected
target image, perform straight-line fitting on the first border
line based on polynomial straight-line fitting to obtain a fitted
straight line, and calculate the first distortion metric value of
the first border line relative to the fitted straight line and the
correction distortion coefficient corresponding to the first
distortion metric value.
[0022] In some embodiments, the image processor is further
configured to determine a straight line segment in the first border
line, calculate distances from corresponding points on the straight
line segment to the fitted straight line, obtain the first
distortion metric value according to the distances, and perform
non-linear optimization on the first distortion metric value to
obtain the correction distortion coefficient.
[0023] In some embodiments, the image processor is further
configured to correct the target image according to the initial
distortion coefficient by correcting a target border line in the
target image based on the initial distortion coefficient, and
correct the target image based on the correction distortion
coefficient by correcting the target border line in the target
image based on the correction distortion coefficient.
[0024] In some embodiments, the image processor is further
configured to perform edge detection on the second corrected target
image to determine the second border line in the second corrected
target image, perform straight-line fitting on the second border
line based on polynomial straight-line fitting to obtain a fitted
straight line, and calculate the second distortion metric value of
the second border line relative to the fitted straight line.
[0025] In some embodiments, the image processor is further
configured to remove outliers, determine a straight line segment in
the second border line, calculate distances from corresponding
points on the straight line segment to the fitted straight line,
and obtain the second distortion metric value according to the
distances.
[0026] In some embodiments, the image processor is further
configured to calculate a relative variation amount between the
first distortion metric value and the second distortion metric
value, and determine whether the relative variation amount
calculated is smaller than a preset variation threshold to
determine whether the preset correction condition is satisfied.
[0027] In some embodiments, the image processor is further
configured to perform image correction based on the correction
distortion coefficient if the preset correction condition is
satisfied
[0028] According to the embodiments of the present disclosure, a
distortion coefficient of an image can be determined
comprehensively based on straight-line fitting and a distortion
metric value. This optimizes a distortion coefficient calculation
and can obtain a more accurate distortion coefficient automatically
and intelligently. The methods and devices consistent with
embodiments of the disclosure also do not require an additional
calibration board, and have a low cost and are easy for users to
use.
BRIEF DESCRIPTION OF THE DRAWINGS
[0029] FIG. 1 is a schematic flow chart of one image processing
method according to an embodiment of the present disclosure;
[0030] FIG. 2 is a schematic flow chart of another image processing
method according to an embodiment of the present disclosure;
[0031] FIG. 3 is a schematic flow chart of processing an image to
obtain distortion coefficients according to an embodiment of the
present disclosure;
[0032] FIG. 4 is a schematic flow chart of processing a corrected
image to obtain distortion coefficients according to an embodiment
of the present disclosure;
[0033] FIG. 5 is a schematic structural diagram of one image
processing device according to an embodiment of the present
disclosure;
[0034] FIG. 6 is a schematic structural diagram of another image
processing device according to an embodiment of the present
disclosure;
[0035] FIG. 7 is a schematic structural diagram of a processing
module in FIG. 6;
[0036] FIG. 8 is a schematic structural diagram of a detection
module in FIG. 6; and
[0037] FIG. 9 is a schematic structural diagram of a video camera
according to an embodiment of the disclosure.
DETAILED DESCRIPTION OF THE EMBODIMENTS
[0038] The technical solution of the present disclosure will be
described in more detail below with reference to the accompanying
drawings. The described embodiments are merely some of the
embodiments of the present disclosure rather than all of the
embodiments. All other embodiments obtained by a person of ordinary
skill in the art based on the embodiments of the present disclosure
without creative efforts shall fall within the scope of the present
disclosure.
[0039] FIG. 1 is a schematic flow chart of an image processing
method according to an embodiment of the present disclosure. The
image processing method according to the embodiment of the present
disclosure may be performed by an image processor. As shown in FIG.
1, at S101, a target image is corrected based on initial distortion
coefficients to obtain a first corrected target image, and
straight-line fitting is performed on a border line in the first
corrected target image to calculate a first distortion metric value
and correction distortion coefficients.
[0040] The initial distortion coefficients may be pre-configured.
In some embodiments, the initial distortion coefficients can be
configured according to a model of a camera lens.
[0041] The border line may be determined from the target image
through edge detection. In the target image, the border line may be
a straight-line edge of a building, and distortion of the target
image can be corrected based on the border line that should be a
straight line.
[0042] The edge detection may be based on positions of pixel points
and amplitude variations of pixel values of the pixel points. In
some embodiments, the edge detection can include a detection method
having a sub-pixel accuracy.
[0043] Simple polynomial straight-line fitting may be employed for
the straight-line fitting of the border line. A series of discrete
pixel points that are supposed to be on a straight line, e.g., the
border line, may scatter around a straight line in the image due to
distortion. These discrete pixel points can be fitted into a
straight line. The fitted straight line is used to reflect a basic
trend of these discrete pixel points.
[0044] The first distortion metric value E.sub.01 can be obtained
according to distances from the discrete pixel points to the fitted
straight line. For example, E.sub.01 may be the sum of squares of
the distances from the discrete pixel points to the fitted straight
line. The smaller E.sub.01 is, the smaller the distortion of the
target image is, or vice versa. After E.sub.01 is obtained, by
performing a non-linear optimization on a function corresponding to
E.sub.01, a set of distortion coefficients that minimizes E.sub.01
can be determined. The determined distortion coefficients are thus
the correction distortion coefficients. In the following
embodiments, one exemplary expression of the function corresponding
to a distortion metric value will be further described in
detail.
[0045] At S102, the target image is corrected based on the
correction distortion coefficients to obtain a second corrected
target image, and straight-line fitting is performed on a border
line in the second corrected target image to calculate a second
distortion metric value, E.sub.02.
[0046] In order to save computing time and computing resources, the
correction of the target image in S102 may include only the
correction of the border line in the target image. When
straight-line fitting is carried out, outliers can be removed to
calculate the second distortion metric value more quickly.
[0047] The second distortion metric may be calculated in a manner
similar to that described above for the first distortion
metric.
[0048] At S103, whether a preset correction condition is satisfied
is detected.
[0049] Whether the first distortion metric value and the second
distortion metric value satisfy the preset correction condition can
be determined by judging whether a relative variation amount
between the first distortion metric value and the second distortion
metric value is smaller than a preset variation threshold. In some
embodiments, an equation for calculating the relative variation
amount may be: (E.sub.01-E.sub.02)/E.sub.02. If a calculated result
is smaller than the preset variation threshold, the correction
condition is satisfied. On the other hand, if the calculated result
is not smaller than the preset variation threshold, the correction
condition is not satisfied.
[0050] In some embodiments, whether the preset condition is
satisfied can be determined by judging whether the second
distortion metric value is smaller than a preset metric threshold.
If the second distortion metric value is smaller than the preset
metric threshold, it is determined that the preset correction
condition is satisfied.
[0051] At S5104, if the preset correction condition is not
satisfied, the correction distortion coefficients are configured as
the initial distortion coefficients, and the process returns to
S101. The above-described processes in S101-S103 can be repeated
until the preset correction condition is satisfied.
[0052] That is, if the preset correction condition is not
satisfied, e.g., if the relative variation amount is not smaller
than the preset threshold or if the second distortion metric value
is not smaller than the preset metric threshold, then the
correction to the target image is not sufficient enough to correct
the distortion, and the distortion of the target image is still
relatively large. Therefore, the target image is further corrected
according to the correction distortion coefficients, and a new
distortion metric value is calculated to determine whether the
corrected target image meets the requirement.
[0053] At S105, if the preset correction condition is satisfied,
image correction is performed based on the correction distortion
coefficients.
[0054] That is, if the preset correction condition is satisfied,
e.g., if the relative variation amount is smaller than the preset
variation threshold or if the second distortion metric value is
smaller than the preset metric threshold, then the correction to
the target image with the correction distortion coefficients has
met the requirement for distortion correction. The correction
distortion coefficients are outputted for performing subsequent
correction on other target images or other related processing.
[0055] The method according to the embodiment of the present
disclosure can be performed when a video camera is being
initialized, and the obtained correction distortion coefficients
can be stored in a memory in order to perform subsequent processing
based on the obtained correction distortion coefficients. In some
embodiments, the method according to the present disclosure can be
performed every time when shooting is performed to obtain
correction distortion coefficients.
[0056] According to the embodiment of the present disclosure,
distortion coefficients of an image can be determined based on
straight-line fitting and a distortion metric value. This optimizes
the calculation of the distortion coefficients and can obtain more
accurate distortion coefficients automatically and intelligently.
Further, additional calibration board is not required, which
reduces cost and is easy for users to use.
[0057] FIG. 2 is a schematic flow chart of another image processing
method according to an embodiment of the present disclosure. The
image processing method according to the embodiment of the present
disclosure may be performed by an image processor. As shown in FIG.
2, at S201, an image of an object that includes straight line
features is captured.
[0058] In some embodiments, multiple pictures can be captured and
analyzed, among which one or more images including an object with
straight line features, such as a building, a playground, or a
motorway, can be used as target images for subsequent distortion
analysis. In some embodiments, the multiple images can be processed
simultaneously or separately, and each image is processed in the
same manner.
[0059] At S202, a size of the captured image is adjusted to obtain
a target image.
[0060] In some embodiments, if the size of the captured image is
smaller than a preset size threshold, the captured image can be
magnified to a target size through interpolation. If the size of
the captured image is greater than the preset size threshold, the
captured image can be scaled down to the target size through
down-sampling.
[0061] Because the distortion coefficients are irrelevant to the
size of the image, the size of the target image can be adjusted to
balance computing time and accuracy. If the image is too small, the
image is magnified to the target size through interpolation to
improve computing accuracy. If the image is too large, the image is
reduced to the target size through down-sampling to improve
computing speed.
[0062] At S203, the target image is corrected based on initial
distortion coefficients to obtain a first corrected target
image.
[0063] In some embodiments, preset initial distortion coefficients
are acquired. In some embodiments, camera model information is
detected and distortion coefficients corresponding to the camera
model information is searched. The located distortion coefficients
are configured as the initial distortion coefficients.
[0064] That is, a user may directly enter the initial distortion
coefficients according to an actual condition of the camera in
order to shorten the computing time of the distortion coefficients.
The initial distortion coefficients can also be automatically
configured based on a model of the camera. Usually, distortion
coefficients of cameras or lenses of the same model are about the
same and are usually only slightly different. Thus, after the model
of the camera or the lens is detected, common distortion
coefficients of the model can be used as the initial distortion
coefficients.
[0065] In some embodiments, in the process of correcting the image
using the initial distortion coefficients, it is feasible to merely
correct the border line in the target image to reduce the
correction time, thereby shortening the entire computing time for
the distortion coefficients.
[0066] At S204, a straight-line fitting is performed on a border
line included in the first corrected target image to calculate a
first distortion metric value and correction distortion
coefficients. The border line in the first corrected target image
is also referred to as a "first border line." Simple polynomial
fitting may be employed for the straight-line fitting. Exemplary
methods for the straight-line fitting and calculating the
distortion metric value and the correction distortion coefficients
are described in more detail below.
[0067] At S205, the target image is corrected based on the
correction distortion coefficients to obtain a second corrected
target image, and a straight-line fitting is performed on a border
line included in the second corrected target image to calculate a
second distortion metric value. The border line in the second
corrected target image is also referred to as a "second border
line." In some embodiments, in the process of correcting the target
image using the correction distortion coefficients, it is feasible
to merely correct a border line in the target image to reduce the
correction time, thereby shortening the entire computing time for
the distortion coefficients.
[0068] At S206, whether a preset correction condition is satisfied
is detected. This can include, for example, whether a relative
variation amount between the first distortion metric value and the
second distortion metric value is smaller than a preset variation
threshold. If the relative variation amount is smaller than the
preset variation threshold, the correction condition is satisfied.
Otherwise, the correction condition is not satisfied. In some
embodiments, the relative variation amount can be calculated using:
(E.sub.01-E.sub.02)/E.sub.02. If a calculated result is smaller
than the preset threshold, the correction condition is satisfied.
If the calculated result is not smaller than the preset threshold,
the correction condition is not satisfied.
[0069] At S207, if the correction condition is not satisfied, the
correction distortion coefficients are configured as the initial
distortion coefficients, and S203 to S206 are repeated until the
preset correction condition is satisfied. That is, if the relative
variation amount is not smaller than the preset variation
threshold, a relationship between the first distortion metric value
and the second distortion metric value does not satisfy the preset
correction condition. This indicates that the correction to the
target image using the correction distortion coefficients still
does not meet the requirement for distortion correction, and the
distortion of the target image is still relatively large. The
target image is again corrected based on the correction distortion
coefficients, and a new distortion metric value is calculated to
determine whether the target image meets the requirement.
[0070] At S208, if the preset correction condition is satisfied,
image correction is performed based on the correction distortion
coefficients.
[0071] That is, if the relative variation amount is smaller than
the preset threshold and the correction condition is satisfied, or
a result obtained after the above processes are repeated is smaller
than the preset threshold and the correction condition is
satisfied, the relationship between the first distortion metric
value and the second distortion metric value satisfies the preset
correction condition. This indicates that correction to the target
image with the correction distortion coefficients has met the
requirement for distortion correction, and the correction
distortion coefficients can be outputted for performing subsequent
corrections on other target image and other related processing.
[0072] FIG. 3 is a schematic flow chart of processing an image to
obtain distortion coefficients according to an embodiment of the
present disclosure. The method shown in FIG. 3 corresponds to S204
in FIG. 2. As shown in FIG. 3, at S301, edge detection is performed
on the first corrected target image to determine the border line in
the first corrected target image. In some embodiments, a detection
method having a sub-pixel accuracy may be employed for the edge
detection, and can depend on software and hardware resource
conditions. For example, in the case that the computing resources
are limited, a general integer-pixel edge detection method can be
used to detect the border line.
[0073] At S302, straight-line fitting is performed on the
determined border line based on polynomial straight-line fitting.
In some embodiments, simple polynomial fitting may be employed for
the straight-line fitting. For example, for n points (x.sub.j,
y.sub.j) (j=1, 2, 3, . . . , n) on a straight line i,
l.sub.i=(a.sub.i,b.sub.i,c.sub.i) can be used to represent the
straight line i corresponding to these points, where x.sub.j,
y.sub.j, a.sub.i, b.sub.i, and c.sub.i satisfy the following
equation:
a.sub.ix.sub.j+b.sub.iy.sub.j+c.sub.i=0.
[0074] Different methods can be employed to estimate a.sub.i,
b.sub.i, and c.sub.i. For example, they can be calculated as
follows:
a i = sin .theta. , b i = cos .theta. , and ##EQU00002## c i = - x
_ sin .theta. - y _ cos .theta. , where : ##EQU00002.2## x _ = 1 n
j = 1 n x j , y _ = 1 n j = 1 n y j , and ##EQU00002.3## tan 2
.theta. = - 2 V xy V xx - V yy , where : ##EQU00002.4## V xx = 1 n
j = 1 n ( x j - x _ ) , V xy = 1 n j = 1 n ( x j - x _ ) ( y j - y
_ ) , and ##EQU00002.5## V yy = 1 n j = 1 n ( y j - y _ ) .
##EQU00002.6##
[0075] As such, a.sub.i, b.sub.i, and c.sub.i can be estimated
using the above equations.
[0076] Another method of estimating a.sub.i, b.sub.i, and c.sub.i
is described below. Assume matrix X is set as a matrix formed by
homogenous expressions of the above-mentioned set of n points,
i.e.,
X = [ x 1 y 1 1 x 2 y 2 1 x n y n 1 ] . ##EQU00003##
In an ideal scenario, all points (x.sub.j, y.sub.j) are on a same
line represented by l, i.e.,
Xl=0.
However, since the points are not ideal, Xl may only be
approximately equal to 0. That is, an optimal solution of l can be
used as the straight line i corresponding to the points (x.sub.j,
y.sub.j). The optimal solution of l satisfies
min l Xl 2 . ##EQU00004##
Various methods can be used to solve this optimization problem. For
example, the expression of l can be obtained by solving the
following optimization equation:
min l Xl 2 , s . t . l = 1. ##EQU00005##
The solution of this equation is a right singular vector
corresponding to the smallest singular value of X Suppose that the
singular value of X is decomposed as X=U.SIGMA.V.sup.T, then
l=V.sub.3, where V.sub.3 is the third column of V, i.e., the right
singular vector corresponding to the smallest singular value. In
order to facilitate calculation of a distance from a point to the
straight line, after the straight line coefficients are obtained,
the coefficients can be multiplied by a scaling factor such that
a.sub.i.sup.2+b.sub.i.sup.2=1.
[0077] In addition, in order to minimize the impact of outliers on
the straight-line fitting, fitting points can be selected by using
RANSAC (RANdom SAmple Consensus).
[0078] At S303, the first distortion metric value of the border
line relative to the straight line corresponding to the
above-described straight-line fitting and the correction distortion
coefficients corresponding to the first distortion metric value are
calculated. The straight line corresponding to the straight-line
fitting is also referred to as a "fitted straight line." In some
embodiments, a straight line segment in the border line can be
determined and distances from corresponding points on the straight
line segment to the fitted straight line can be calculated. The
first distortion metric value can be obtained according to the
calculated distances. The correction distortion coefficients can be
obtained by non-linearly optimizing the first distortion metric
value.
[0079] In some embodiments, multiple fitted straight lines can be
obtained by the straight-line fitting. Adjacent straight lines
having close coefficients can be connected to form one straight
line, and the first distortion metric value can be calculated for
all of the fitted straight lines, which can be the sum of squares
of distances from the points to the lines:
E.sub.01=.SIGMA..sub.i.SIGMA..sub.j.SIGMA.l.sub.i(a
.sub.ix.sub.j+b.sub.iy.sub.j+c.sub.i).sup.2.
where a.sub.i, b.sub.i, and c.sub.i are coefficients of the i-th
straight line among the multiple straight lines. The smaller
E.sub.01 is, the smaller the distortion is. The greater E.sub.01
is, the greater the distortion is.
[0080] Moreover, it can be seen from the above equation that the
distortion metric is a function of x.sub.j and y.sub.j. In the
distortion equation, x.sub.j and y.sub.j are equations for the
distortion coefficients k=(k.sub.1, k.sub.2). Therefore, the
distortion metric is a function of the distortion coefficients. A
set of k.sub.1, k.sub.2) values can be found by non-linearly
optimizing the function corresponding to the distortion metric that
minimizes the distortion metric value, and the set k=(k.sub.1,
k.sub.2) are the correction distortion coefficients.
[0081] FIG. 4 is a schematic flow chart showing a method of
processing a corrected image to obtain distortion coefficients
according to an embodiment of the present disclosure. The process
shown in FIG. 4 corresponds to S205 in FIG. 2. As shown in FIG. 4,
at S401, edge detection is performed on the second corrected target
image to determine the border line in the second corrected target
image.
[0082] At S402, straight-line fitting is performed on the
determined border line based on polynomial straight-line
fitting.
[0083] At S403, the second distortion metric value of the border
line relative to a straight line corresponding to the straight-line
fitting is calculated. In some embodiments, calculating the second
distortion metric value can include removing outliers, determining
a straight line segment in the border line, calculating distances
from corresponding points on the straight line segment to the
straight line corresponding to the straight-line fitting, and
obtaining the second distortion metric value according to the
calculated distances.
[0084] Reference can be made to the description of the
corresponding embodiment of FIG. 3 for the processes of the edge
detection, the straight-line fitting, and the calculation of the
second distortion metric value.
[0085] According to the embodiments of the present disclosure,
distortion coefficients of an image can be determined based on both
straight-line fitting and a distortion metric value. This optimizes
a distortion coefficient calculation and can obtain more accurate
distortion coefficients automatically and intelligently. The
methods consistent with embodiments of the disclosure do not
require an additional calibration board, and have a low cost and
are easy for users to use.
[0086] An image processing device and a video camera according to
the embodiments of the present disclosure are described below in
detail.
[0087] FIG. 5 is a schematic structural diagram of an image
processing device according to an embodiment of the present
disclosure. The device according to the embodiment of the present
disclosure can be configured in various kinds of video cameras. As
shown in FIG. 5, the image processing device includes a processing
module 1, a detection module 2, and a correction module 3.
[0088] The processing module 1 is configured to correct a target
image based on initial distortion coefficients to obtain a first
corrected target image and to perform straight-line fitting on a
border line included in the corrected target image to calculate a
first distortion metric value and correction distortion
coefficients. The processing module 1 is further configured to
correct the target image according to the correction distortion
coefficients to obtain a second corrected target image and to
perform straight-line fitting on a border line included in the
second corrected target image to calculate a second distortion
metric value.
[0089] The detection module 2 is configured to detect whether a
preset correction condition is satisfied. If the preset correction
condition is not satisfied, the detection module 2 configures the
correction distortion coefficient as the initial distortion
coefficient, and forwards the new initial distortion coefficient to
the processing module 1 such that the processing module 1 can
repeat the correction and straight-line fitting processes based on
the new initial distortion coefficient.
[0090] The correction module 3 is configured to, when a detection
result of the detection module 2 is that the preset correction
condition is satisfied, perform image correction based on the
correction distortion coefficients.
[0091] The initial distortion coefficients used in the image
correction performed for the first time by the processing module 1
may be pre-configured. In some embodiments, the initial distortion
coefficients can be configured according to a model of a camera
lens.
[0092] The processing module 1 may determine the border line from
the target image by way of edge detection. In the target image, the
border line may be a straight-line edge of a building, and
distortion of the target image can be corrected through the border
line that should be a straight line.
[0093] The edge detection used by the processing module 1 may be
based on positions of pixel points and amplitude variations of
pixel values of the pixel points. In some embodiments, the edge
detection can include a detection method having a sub-pixel
accuracy.
[0094] Simple polynomial straight-line fitting may be employed for
the straight-line fitting on the border lines by the processing
module 1. A series of discrete pixel points that are supposed to be
on a straight line, e.g., the border line, may scatter around a
straight line in the image due to distortion. These discrete pixel
points can be fitted into a straight line. The fitted straight line
is used to reflect a basic trend of these discrete pixel
points.
[0095] The processing module 1 can obtain the first distortion
metric value according to distances from the discrete pixel points
to the fitted straight line. After the first distortion metric
value is obtained, through a non-linear optimization on a function
corresponding to the first distortion metric value, a set of
distortion coefficients that minimizes the first distortion metric
value can be determined. The determined distortion coefficients are
thus the correction distortion coefficients.
[0096] In order to save computing time and computing resources,
when the processing module 1 corrects the target image according to
the first distortion coefficient, the processing module 1 may only
correct the border line in the target image. When straight-line
fitting is carried out, outliers can be removed to calculate the
second distortion metric value more quickly.
[0097] The detection module 2 may determine whether the first
distortion metric value and the second distortion metric value
satisfy the preset correction condition by judging whether a
relative variation amount between the first distortion metric value
and the second distortion metric value is smaller than a preset
variation threshold. In some embodiments, an equation for
calculating the relative variation amount may be:
(E.sub.01-E.sub.02)/E.sub.02, where E.sub.01 denotes the first
distortion metric value and E.sub.02 denotes the second distortion
metric value. If a calculation result is smaller than the preset
variation threshold, the correction condition is satisfied. On the
other hand, if the calculation result is not smaller than the
preset variation threshold, the correction condition is not
satisfied.
[0098] If the detection module 2 detects that the preset correction
condition is not satisfied, e.g., when the relationship between the
first distortion metric value and the second distortion metric
value is not smaller than the preset variation threshold, the
correction to the target image is not sufficient enough to correct
the distortion, and the distortion of the target image is still
relatively large. In this scenario, the detection module 2 can
notify the processing module 1 to correct the target image once
again based on the correction distortion coefficients and to
calculate a new distortion metric value to determine whether the
corrected target image meets the requirement.
[0099] If the correction condition is satisfied, e.g., if the above
relative variation amount is smaller than the preset threshold or
if the second distortion metric value is smaller than a preset
metric threshold, the detection module 2 determines that the preset
correction condition is satisfied, which indicates that the
correction to the target image with the correction distortion
coefficients has met the requirement for distortion correction. The
correction module 3 uses the correction distortion coefficients for
subsequent image correction and other related processing.
[0100] According to the embodiment of the present disclosure,
distortion coefficients of an image can be determined based on
straight-line fitting and a distortion metric value. This optimizes
a distortion coefficient calculation and can obtain more accurate
distortion coefficients automatically and intelligently. Further,
additional calibration board is not required, which reduces cost
and is easy for users to use.
[0101] FIG. 6 is a schematic structural diagram of another image
processing device according to an embodiment of the present
disclosure. The device according to the embodiment of the present
disclosure can be configured in various kinds of video cameras. As
shown in FIG. 6, the image processing device includes the
processing module 1, the detection module 2, the correction module
3, an image acquisition module 4, and a size adjustment module
5.
[0102] The image acquisition module 4 is configured to capture an
image of an object that includes straight line features. The size
adjustment module 5 is configured to adjust the size of the
captured image to obtain a target image.
[0103] In some embodiments, the size adjustment module 5 is
configured to, if the size of the captured image is smaller than a
preset size threshold, magnify the captured image to a target size
through interpolation; and, if the size of the captured image is
greater than the preset size threshold, scale down the captured
image to the target size through down-sampling.
[0104] The image acquisition module 4 may analyze multiple pictures
captured and use an image including an object with straight line
features, such as a building, a playground, or a motorway, as a
target image for subsequent distortion analysis. In the embodiment
of the present disclosure, the size adjustment module 5 can process
multiple images simultaneously or subsequently, and each image can
be processed in the same manner.
[0105] Because the distortion coefficients are irrelevant to the
size of the image, the size adjustment module 5 can adjust the size
of the target image in order to balance computing time and
accuracy. If the image is too small, the size adjustment module 5
magnifies the image to a target size through interpolation to
improve the computing accuracy; and if the image is too large, the
size adjustment module 5 reduces the image to the target size
through down-sampling to improve the computing speed.
[0106] In some embodiments, as shown in FIG. 7, the processing
module 1 includes a first processing unit 11, a first determination
unit 12, a correction unit 13, a second processing unit 14, and a
second determination unit 15.
[0107] The first processing unit 11 configured to perform edge
detection on the target image to determine a border line in the
target image, and to perform straight-line fitting on the
determined border line based on polynomial straight-line fitting to
obtain a fitted straight line.
[0108] The first determination unit 12 is configured to calculate a
first distortion metric value of the border line relative to the
fitted straight line and correction distortion coefficients
corresponding to the first distortion metric value.
[0109] In some embodiments, the first determination unit 12 is
configured to determine a straight line segment in the border line,
calculate distances from corresponding points on the straight line
segment to the fitted straight line, obtain the first distortion
metric value according to the calculated distances, and
non-linearly optimize the first distortion metric value to obtain
the correction distortion coefficients.
[0110] The correction unit 13 is configured to correct the border
line in the target image according to the initial distortion
coefficients or the correction distortion coefficients to complete
correction of the target image.
[0111] The second processing unit 14 is configured to perform edge
detection on the corrected target image to determine a border line
in the corrected target image, and perform straight-line fitting on
the determined border line based on polynomial straight-line
fitting to obtain a fitted straight line.
[0112] The second determination unit 15 is configured to calculate
a second distortion metric value of the border line corresponding
to the fitted straight line. In some embodiments, the second
determination unit 15 is further configured to remove outliers and
determine a straight line segment in the border line, and calculate
distances from corresponding points on the straight line segments
to the fitted straight line and obtain the second distortion metric
value based on the calculated distances.
[0113] In some embodiments, as shown in FIG. 8, the detection
module 2 includes a variation calculation unit 21 and a condition
determination unit 22. The variation calculation unit 21 is
configured to calculate a relative variation amount between the
first distortion metric value and the second distortion metric
value. The condition determination unit 22 is configured to, if the
relative variation amount calculated is smaller than a preset
variation threshold, determine that the correction condition is
satisfied, or otherwise, determine that the correction condition is
not satisfied.
[0114] Referring again to FIG. 6, in some embodiments, the device
according to the embodiment of the present disclosure further
includes an acquisition module 6 configured to acquire preset
initial distortion coefficients, or detect camera model
information, search for distortion coefficients corresponding to
the camera model information, and configure the located distortion
coefficients as the initial distortion coefficients.
[0115] Functions of various modules and units in the embodiments
described above in connection with FIGS. 5-8 are similar to the
methods in the embodiments described above in connection with FIGS.
1-4, and detailed description thereof is omitted.
[0116] According to embodiments of the present disclosure,
distortion coefficients of an image can be determined based on
straight-line fitting and a distortion metric value. This optimizes
a distortion coefficient calculation and can obtain more accurate
distortion coefficients automatically and intelligently. Therefore,
additional calibration board is not required, which reduces cost
and is easy for users to use.
[0117] FIG. 9 is a schematic structural diagram of a video camera
900 consistent with embodiments of the disclosure. The video camera
900 includes a camera lens 910 and an image processing device 920.
The image processing device 920 includes an image processor 922 and
a memory 924. The memory 924 stores an image processing program
containing instructions consistent with embodiments of the
disclosure. When the image processing program is executed by the
image processor 922, it causes the image processor 922 to execute a
method consistent with embodiments of the disclosure, such as one
of the exemplary methods described above.
[0118] In some embodiments, the image processor 922 is configured
to correct a target image according to initial distortion
coefficients to obtain a first corrected target image and perform
straight-line fitting on a border line included in the first
corrected target image to calculate a first distortion metric value
and correction distortion coefficients. The image processor 922 is
further configured to correct the target image according to the
correction distortion coefficients to obtain a second corrected
target image and perform straight-line fitting on a border line
included in the second corrected target image to calculate a second
distortion metric value. The image processor 922 can detect whether
a preset correction condition is satisfied. If the preset
correction condition is not satisfied, the image processor 922 can
configure the correction distortion coefficients as the initial
distortion coefficients, and perform processing again until the
preset correction condition is satisfied. If the preset correction
condition is satisfied, the processor 922 can perform image
correction according to the correction distortion coefficients.
[0119] In some embodiments, the image processor 922 is configured
to capture an image of an object including straight line features
through the camera lens 910, and adjust the size of the captured
image to obtain the target image. In some embodiments, the image
processor 922 is configured to, if the size of the captured image
is smaller than a preset size threshold, magnify the captured image
to a target size through interpolation, and if the size of the
captured image is greater than the preset size threshold, scale
down the captured image to the target size through
down-sampling.
[0120] In some embodiments, the image processor 922 is configured
to perform edge detection on the target image to determine a border
line in the target image, perform straight-line fitting on the
determined border line based on polynomial straight-line fitting to
obtain a fitted straight line, and calculate the first distortion
metric value of the border line corresponding to the fitted
straight line and the correction distortion coefficients
corresponding to the first distortion metric value.
[0121] In some embodiments, the image processor 922 determines a
straight line segment in the border line, calculates distance from
corresponding points on the straight line segment to the fitted
straight line and obtains the first distortion metric value
according to the calculated distances, and non-linearly optimizes
the first distortion metric value to obtain the correction
distortion coefficients.
[0122] In some embodiments, the image processor 922 is configured
to correct the border line in the target image according to the
initial distortion coefficients or the correction distortion
coefficients to complete correction of the target image.
[0123] In some embodiments, the image processor 922 is configured
to perform edge detection on the second corrected target image to
determine a border line in the second corrected target image,
perform straight-line fitting on the determined border line based
on polynomial straight-line fitting to obtain a fitted straight
line, and calculate the second distortion metric value of the
border line relative to the fitted straight line.
[0124] In some embodiments, the image processor 922 is configured
to remove outliers and determine a straight line segment in the
border line, and calculate distances from corresponding points on
the straight line segment to the fitted straight line and obtain
the second distortion metric value according to the calculated
distances.
[0125] In some embodiments, the image processor 922 is configured
to calculate a relative variation amount between the first
distortion metric value and the second distortion metric value. If
the relative variation amount calculated is smaller than a preset
variation threshold, the correction condition is satisfied.
Otherwise, the correction condition is not satisfied.
[0126] In some embodiments, the image processor 922 is further
configured to acquire preset initial distortion coefficients, or
detect camera model information and search for distortion
coefficients corresponding to the camera model information to
configure the located distortion coefficients as the initial
distortion coefficients.
[0127] According to the embodiment of the present disclosure, a
distortion coefficient of an image can be determined
comprehensively based on straight-line fitting and a distortion
metric value. This optimizes a distortion coefficient calculation
manner and can obtain more accurate distortion coefficients
automatically and intelligently. The embodiment also does not
require an additional calibration board, and has a low cost and is
easy for users to use.
[0128] In the several embodiments provided in the present
disclosure, it should be understood that the related devices and
methods disclosed may be implemented in another manner. For
example, the device embodiments described above are merely
illustrative. For example, division of the module or unit is merely
division of a logical function, and division in another manner may
exist in actual implementation. For example, a plurality of units
or assemblies may be combined or integrated to another system, or
some features may be omitted or not performed. In addition, the
mutual coupling or direct coupling or communication connections
displayed or discussed may be implemented by using some interfaces,
and the indirect coupling or communication connections between the
devices or units may be electrical, mechanical or in another
form.
[0129] The units described as separate components may be or may not
be physically separate, and components displayed as units may be or
may not be physical units, may be located in one position, or may
be distributed on a plurality of network units. Some or all of the
units may be selected according to actual needs to achieve the
objective of the solution of the embodiment.
[0130] In addition, functional units in the embodiments of the
present disclosure may be integrated into one processing unit, or
each of the units may exist alone physically, or two or more units
may be integrated into one unit. The aforementioned integrated unit
may be implemented in a form of hardware, or may be implemented in
a form of a software functional unit.
[0131] When the integrated unit is implemented in the form of a
software functional unit and sold or used as an independent
product, the integrated unit may be stored in a computer-readable
storage medium. The computer software product is stored in a
storage medium, and includes several instructions used for causing
a computer processor to perform all or a part of a method
consistent with embodiments of the present disclosure, such as one
of the exemplary methods described above. The foregoing storage
medium includes any medium that can store program codes, such as a
USB flash drive, a portable hard disk, a Read-Only Memory (ROM), a
Random Access Memory (RAM), a magnetic disk, or an optical
disc.
[0132] In some embodiments (for example, when only the distortion
in one dimension is of concern, or when the distortion in one
dimension is negligible such that only the distortion in the other
dimension needs to be considered), instead of calculating the two
distortion coefficients k.sub.1 and k.sub.2 as discussed above,
only one distortion coefficient k.sub.1 or k.sub.2 may need to be
calculated. The methods and apparatuses for calculating the one
distortion coefficient are similar to those described above for
calculating both distortion coefficients, and thus detailed
description thereof is omitted.
[0133] The above descriptions merely relate to embodiments of the
present disclosure, but are not intended to limit the scope of the
present disclosure. Any equivalent structure or equivalent process
variation made by using contents of the specification and the
drawings of the present disclosure, or directly or indirectly
applied to other related technical fields, should be likewise
included in the scope of the present disclosure.
* * * * *