U.S. patent application number 10/075300 was filed with the patent office on 2002-08-22 for digital cameras.
Invention is credited to Pilu, Maurizio, Pollard, Stephen B..
Application Number | 20020113882 10/075300 |
Document ID | / |
Family ID | 9908880 |
Filed Date | 2002-08-22 |
United States Patent
Application |
20020113882 |
Kind Code |
A1 |
Pollard, Stephen B. ; et
al. |
August 22, 2002 |
Digital cameras
Abstract
A digital image capture apparatus is provided which is arranged
to capture at least two images of a document and combine the images
to produce a final image in which the effects of specular
reflection and/or ambient light are removed. The apparatus includes
a light source and a first image is captured with the source
illuminating the document. A second ambient image is captured with
the light source switched off, and the images are combined either
by subtracting the ambient image from the first image or by
replacing areas of the first image by corresponding areas of the
ambient image.
Inventors: |
Pollard, Stephen B.; (Nr
Dursley, GB) ; Pilu, Maurizio; (Bristol, GB) |
Correspondence
Address: |
HEWLETT-PACKARD COMPANY
Intellectual Property Administration
P.O. Box 272400
Fort Collins
CO
80527-2400
US
|
Family ID: |
9908880 |
Appl. No.: |
10/075300 |
Filed: |
February 15, 2002 |
Current U.S.
Class: |
348/239 ;
348/246; 348/370; 348/552 |
Current CPC
Class: |
H04N 1/2116 20130101;
H04N 1/2112 20130101; H04N 1/00835 20130101 |
Class at
Publication: |
348/239 ;
348/370; 348/246; 348/552 |
International
Class: |
H04N 005/262; H04N
005/222; H04N 009/64 |
Foreign Application Data
Date |
Code |
Application Number |
Feb 16, 2001 |
GB |
0103828.0 |
Claims
1. A digital image capture apparatus configured to capture an image
of a document (102) at least partially illuminated by ambient
light, the apparatus comprising: an image capture device (101)
comprising a detector (107) adapted to capture an image of said
document; an illumination source (106) for illuminating said
document, a detector read-out circuit (108) for capturing image
data from said detector (107) in response to a control signal; a
controller for controlling said detector read-out circuit (108) and
said illumination source (106), wherein said controller provides a
first control signal to said read-out circuit (108) to capture a
first image of said document (102) when illuminated by said
illumination source (106) at a first intensity and by said ambient
light to produce a first image, and wherein said controller further
provides a second control signal to said read-out circuit to
capture a second image of said document (102) when said document
(102) is illuminated by said illumination source at a second,
lower, intensity and by said ambient light, said second, lower,
intensity having a value greater than or equal to zero intensity
but less than the first intensity; and a processor (112) arranged
to process said first captured image with said second captured
image to produce a final image of said document.
2. Apparatus according to claim 1 wherein said first image and said
second image are captured with substantially a same exposure and
said processor (112) is arranged to combine said first captured
image and said second captured image to form a final image by
subtracting said second image from said first image.
3. Apparatus according to claim 2 wherein said processor (112)
includes a low pass filter, and wherein at least one of said
captured images is passed through said low pass filter prior to
subtraction.
4. Apparatus according to claim 3 wherein only said second captured
image is low-pass filtered.
5. Apparatus according to claim 1 wherein said second image is
captured with a longer exposure than said first image, and wherein
said illumination source (116) has an intensity during capture of
said second image of zero, and wherein said processor (112) is
arranged to combine said two captured images to form a final image,
said processor (112) including an identifying stage for identifying
areas in said first captured image that correspond to glare spots,
and a replacement stage for replacing those points with values
dependent upon values of corresponding areas in said second
image.
6. Apparatus according to claim 5 wherein an exposure of said first
image and an exposure of said second image are selected to provide
a full exposure for each of said images.
7. Apparatus according to claim 5 wherein said processor
pre-processes said second image prior to combining with said first
image to compensate for differences in at least one of colour
temperature and exposure profile and distribution of light between
said ambient light and light from said illumination source.
8. Apparatus according to claim 2, wherein said controller provides
a third control signal to said read-out circuit (108) to capture a
third image of said document (102) when said document (102) is
illuminated by said ambient light without being illuminated by said
illumination source, said third image having a longer exposure than
said first image; and wherein said processor is arranged to combine
said three captured images to form a final image, said processor
(112) including a combination stage arranged to subtract said
second image from said first image to form a processed first
captured image, an identifying stage for identifying areas or
points in said processed first captured image that correspond to
glare spots, and a replacement stage for replacing those points
with values dependent upon the values at corresponding areas or
points in said third image.
9. Apparatus according to claim 8 wherein an exposure for said
first image and said third image are selected to provide a full
exposure for each of said first image and said second image.
10. Apparatus according to claim 8 wherein said processor
preprocesses said third image prior to combining with said first
image to compensate for differences in colour temperature between
said ambient light and said light from said illumination means.
11. Apparatus according to claim 10 wherein said processor computes
a transformation that maps colour intensity data obtained for said
third image onto said processed first captured image.
12. Apparatus according to claim 5 wherein said processor includes
a blending stage for blending together edges of areas of said first
image which have been replaced by corresponding areas from another
image.
13. Apparatus according to claim 1 wherein said processor spatially
matches areas or points in said captured images by identifying
similar features in each image and producing a mapping transform
which ensures that areas of each image that correspond to identical
areas in corresponding images are combined.
14. Apparatus according to claim 1 wherein a shield (415) is
provided around at least part of said illumination source (106) to
prevent ambient light from striking said document (102) from said
illumination source (106).
15. Apparatus according to claim 1 which further includes a second
illumination source for illuminating said document, and wherein
said controller provides said first control signal to said read-out
circuit to capture said first image of said document when
illuminated by said first illumination source means and by said
ambient light but not said second illumination source, and further
wherein said controller provides an additional control signal to
said read-out circuit to capture an additional image of said
document when said document is illuminated by said second
illumination source and by said ambient light without being
illuminated by said first illumination source; said processor
processing said first captured image with said additional captured
image to produce said final image of said document.
16. Apparatus according to claim 15 wherein said processor combines
said first image and said additional image to produce a final image
in which value of at least one data point in said final image is a
minimum value selected from corresponding points in said first
image and said additional image.
17. Apparatus according to claim 1 wherein said image capture
device comprises a digital camera.
18. A method of capturing an image of a document at least partially
illuminated by ambient light, said method comprising providing a
controllable light source; capturing a first image of said document
with said light source illuminating said document at a first
intensity; capturing a second image of said document when
illuminated by said light source at a second, lower, intensity; and
processing said first captured image with said second captured
image to produce a final image of said document, wherein said
second lower intensity is greater than or equal to zero intensity
but less than said first intensity.
19. The method of claim 18 wherein said first image and said second
image are captured using identical exposures and said processor
combines said images by subtracting said second image from said
first image.
20. The method of claim 18 wherein said first image and said second
image are captured at different exposures such that said first
image and said second image are fully exposed, wherein said light
source has an intensity during capture of said second image which
is zero, and wherein said method further comprises identifying
areas or points in said first image that correspond to glare spots
and replacing those areas or points with corresponding areas or
points from said second image.
21. A digital image capture apparatus configured to capture an
image of a document (102) at least partially illuminated by ambient
light, the apparatus comprising: a digital camera (101) comprising
a detector (107) arranged to capture an image of said document; a
light source (106) for illuminating said document, a detector
read-out circuit (108) for capturing image data from said detector
(107) in response to a control signal; a controller for controlling
said detector read-out circuit (108) and said light source (106),
wherein said controller provides a first control signal to said
read-out circuit (108) to capture a first image of said document
(102) when illuminated by said light source (106) at a first
intensity and by said ambient light to produce a first image
captured at a first exposure, and wherein said controller further
provides a second control signal to said read-out circuit to
capture a second image of said document (102) when said document
(102) is illuminated only by said ambient light; and a processor
(112) arranged to process said first captured image with said
second captured image to produce a final image of said document,
and wherein said processor is further arranged to identify areas or
points in said first captured image that correspond to glare spots,
and to replace said identified points or areas with corresponding
points or areas taken from said second image.
22. A digital image capture apparatus according to claim 21 wherein
said processor includes a blending stage for blending together
areas of said first image which have been replaced by corresponding
areas of said second image.
23. A digital image capture apparatus configured to capture an
image of a document (102) at least partially illuminated by ambient
light, the apparatus comprising: an image capture device (101)
comprising a detector (107) adapted to capture an image of said
document; an illumination source (106) for illuminating said
document, a detector read-out circuit (108) for capturing image
data from said detector (107) in response to a control signal; a
controller for controlling said detector read-out circuit (108) and
said illumination source (106), in which said controller provides a
first control signal to said read-out circuit (108) to capture a
first image of said document (102) when illuminated by said
illumination source (106) at a first intensity and by said ambient
light to produce a first image, and wherein said controller further
provides a second control signal to said read-out circuit to
capture a second image of said document (102) when said document
(102) is illuminated by said illumination source at a second, lower
intensity and by said ambient light; and processor (112) arranged
to process said first captured image with said second captured
image to produce a final image of said document.
Description
[0001] This invention relates to the field of digital cameras. It
in particular, but not exclusively relates to a digital image
capture apparatus which is configured to obtain an electronic image
of a document which is largely free of the effects of specular
reflection.
[0002] The most widely used devices at present for capturing an
electronic image of a document are flat-bed and sheet feed
scanners. Whilst these devices have proved extremely robust they do
suffer from several disadvantages in some circumstances. For
instance, flat-bed scanners occupy a significant area of desk space
whilst sheet feed devices are limited to operation with documents
of limited size and thickness.
[0003] An alternative device which can be employed for document
capture is the digital camera. Digital cameras are well known in
the art and in their simplest form comprise a detector and a lens
assembly. The detector most commonly comprises an array of light
sensitive elements. An example is the charge coupled device (CCD)
array. The lens assembly is adapted to direct light from within a
predetermined field of view onto the detector.
[0004] In use, a document is placed within the field of view of the
camera and at a distance such that it properly focused onto the
lens. This can be assured by supporting the camera on a fixed stand
over an area of desk space onto which document is placed. The lens
produces an image of the document on the detector. This image is
converted into a digital image, with the value of each point or
pixel in the image corresponding to the output of a corresponding
node of the array.
[0005] The digital camera also typically includes a read-out
circuit which interrogates each of the nodes of the array. The
read-out circuit may serve as a digital shutter to control the
exposure of each image. The output of the read-out circuit is
stored in a memory integrated into the camera or is fed through a
remote link to a computer or the like. To ensure that a controlled
and even illumination of the original document is obtained a flash
light is typically incorporated into the camera. The flash light
illuminates the original document with white light during image
capture. By choosing a high intensity flash light and a
correspondingly high shutter speed it is possible to swamp the
majority of light produced by unwanted ambient light sources.
[0006] A simple camera of this design has many limitations.
Firstly, the image that is captured will typically include areas of
the document which are obscured by the phenomena of specular
reflection. Typically, many documents consist of glossy material or
semi-glossy material. When the flashlight is illuminated high
energy specular reflected light is incident upon the light
sensitive nodes which swamps the lower level diffuse light
corresponding to the image content of the document. Also, some
specular reflection may be caused by ambient light incident upon
the document.
[0007] Several solutions to the problem of specular reflection
within an image have been proposed in the prior art. The simplest
of these solutions is to move the flash light out of the field of
view of the camera lens.
[0008] Unfortunately, this increases the size of the camera as the
distance from the lens to the flash light must be increased
compared with a system in which the light source lies within the
field of view.
[0009] The use of an obliquely positioned light source also
produces an uneven illumination of the document which is
undesirable and requires carefully tailored (and expensive)
reflector designs.
[0010] An alternative solution has been proposed in which two light
sources are provided on the camera positioned on a circle
diametrically opposite one another with the lens in the centre of
the circle. A first image is captured with the document illuminated
by the first source only. A second image is then captured using
only light from the second source. The two images are then combined
to remove specular reflection caused by each light source.
[0011] Whilst this second technique is reasonably effective in
removing specular reflection there may still be glare spots caused
by uncontrolled ambient light. This can only be prevented by
operating the system in an area of low ambient light levels.
[0012] It is an object of this invention to provide a digital image
capture apparatus which is suitable for use in a variety of
operating conditions.
[0013] In accordance with a first aspect the invention provides a
digital image capture apparatus configured to capture an image of a
document at least partially illuminated by ambient light, the
apparatus comprising:
[0014] an image capture means comprising a detector adapted to
capture an image of the document;
[0015] illumination means for illuminating the document,
[0016] a detector read-out means for capturing image data from the
detector in response to a control signal;
[0017] control means for controlling the detector read-out means
and the illumination means, in which the control means provides a
first control signal to the read-out means to capture a first image
of the document when illuminated by the illumination means at a
first intensity and the ambient light to produce a first image, and
in which the control means further provides a second control signal
to the read-out means to capture a second image of the document
when the document is illuminated by the illumination means at a
second, lower, intensity and the ambient light; and
[0018] processing means adapted to process the first captured image
with the second captured image to produce a final image of the
document.
[0019] The second lower intensity level of the illumination means
preferably comprises zero or substantially zero intensity of the
illumination means. In this case, the second image is captured
using only ambient illumination.
[0020] The apparatus therefore combines the data obtained from two
captured images together to produce a final image of the document.
One of the images is captured with the document illuminated by the
camera together with ambient light. The other is captured under
ambient light with a smaller amount of illumination from the
illumination means. By ambient light, we may mean sunlight, ceiling
lights and/or task lights which illuminate the document.
[0021] The read-out circuit may be adapted to capture first and
second images from the detector at substantially the same exposure,
and the processing means may be adapted to combine the two captured
images to form a final image by subtracting the data forming the
second image from the data forming the first image. This will
remove the ambient light contribution from the first captured
image. This is advantageous over prior art systems which have
chosen to use a very fast exposure in attempts to make the final
image insensitive to the contribution of ambient light.
[0022] In many cases, the captured images may contain noise. A
common source of such noise is the detectors themselves which may
suffer from KTC noise. To combine two images which contain noise by
subtracting the captured data will have the effect of doubling the
noise levels in the final image. In extreme case this may have a
noticeable effect on the quality of the final image.
[0023] It is therefore proposed that the processing means may be
configured to pass one or both of the captured images through a low
pass filter prior to subtraction. The high frequency noise can be
removed by such a filter to leave a higher quality final image.
[0024] The filter may smooth the or each captured image in the
spatial or frequency domain.
[0025] It is most preferred that only the second captured image
corresponding to the ambient light is low-pass filtered.
[0026] Conveniently, the illumination means may comprise one or
more sources of light which are located outside of the field of
view of the detector. A suitable source is a white flash bulb such
as is commonly used on digital cameras. A lens may be provided in
order to control the field of view. With this arrangement the
illumination means will not produce any glare spots in the final
image due to specular reflection of light onto the detector.
However, glare spots may be produced due to ambient light which
reflects onto the detector. As the source of ambient light cannot
be controlled this can not be avoided. By subtracting the second
image from the first image these specular reflections can be
substantially removed from the final image.
[0027] In some applications, however, it may be preferred that the
illumination means comprises at least one source of light that is
located within the field of view of the detector. In this case,
subtracting the second image from the first image can be used to
remove the contribution to glare made by ambient light (and also
the background effect of ambient light) but will not remove the
glare produced by the illumination means.
[0028] To deal with such a problem, the control means may provide
the second control signal to the read-out means to capture the
second image of the document when illuminated by the ambient light
substantially without any illumination from the illumination means
and with a longer exposure than the first captured image, the
processing means including identifying means for identifying areas
or points in the first captured image that correspond to glare
spots and replacement means configured to replace those parts of
the first image identified as glare spots with corresponding parts
taken from the second image.
[0029] In effect, areas of glare are pasted over by the
corresponding areas of the second image which are captured under
ambient light.
[0030] It will of course be appreciated that the apparatus may
capture two ambient images-one at the same exposure as the first
image and another at the longer exposure which is preferably a full
exposure. By full exposure we mean that the captured image is
neither under or over exposed.
[0031] Therefore in a refinement the control means may:
[0032] provide a third control signal to the read-out means to
capture a third image of the document when the document is
illuminated by the ambient light without being illuminated by the
illumination means, the third image having a longer exposure than
the first image; and
[0033] in which the processing means is arranged to combine the
three captured images to form a final image, the processing means
including combining means arranged to subtract the second image
from the first image to form a processed first captured image,
identifying means for identifying areas or points in the processed
first captured image that correspond to glare spots, and
replacement means for replacing those points with values dependent
upon the data at corresponding points in the third image.
[0034] Preferably the first and second images are produced using
substantially the same exposure whilst the third image is produced
at a different, longer, exposure. Conveniently, the exposure for
the first and the third images is selected to provide a
substantially full exposure of the captured image.
[0035] The technique of pasting in areas of the fully exposed
ambient image is most successful in cases where the colour
temperature of the illumination means and the ambient light are the
same. If the colour temperatures differ there will be a noticeable
mismatch in the final image.
[0036] The processing means may pre-process the third image prior
to combining with the other images to compensate for differences in
colour temperature and/or the exposure and/or distribution of light
between the ambient light and the illumination means.
[0037] The processing means may compute a transformation that maps
the colour intensity data for the ambient image onto the first
image. This mapping enables variations in colour temperature across
the image to be compensated. Of course, this will require a
detector which is configured to produce colour output data,
typically having an array of nodes sensitive to red, green and blue
light.
[0038] Suitable transforms include 3.times.3 or 3.times.4 colour
matrix (computed using a least squares method between corresponding
pixels in the first and the third images. Most preferably, the
colour compensation is performed after the second image has been
subtracted from the first image.
[0039] In areas of specular reflection extreme variations in colour
temperature may be present. The processing means may therefore
determine the colour transform for areas of specular reflection
using data obtained from one or more regions immediately
surrounding the specular data.
[0040] The matching can be further enhanced by blending together
the edges of areas of the final image which have been replaced by
data from the third image. A blending region may be defined around
each replaced area or point which radiates away from each
identified point in the first image which corresponds to a specular
reflection. In the blend region, the data from the first and third
images may be blended by amounts varying from 100 percent third
image to 100 percent first image moving away from the specular
point. The blend may vary linearly with distance across the blend
region.
[0041] To ensure that the ambient light conditions present during
the capture of the first and the second (or third) images remain
constant a high shutter speed should be selected and the time gap
between the capture of images should be as short as possible. The
time gap is typically dependent upon the speed at which the image
data can be read from the detector. Modern devices require around
0.1 seconds to capture each image. This delay can in some cases
cause a spatial mismatch between the images if the detector moves
relative to the document between images.
[0042] To allow for some movement of the detector the processing
means may be adapted to spatially match the captured images by
identifying similar features in each image and producing a mapping
transform which ensures that areas of each image that correspond to
identical features are combined.
[0043] The apparatus may conveniently include a shield which is
provided around at least part of the illumination means which
substantially prevents ambient light from striking the document
from the location of the illumination means.
[0044] By providing the shield it is possible to ensure that any
specular reflection due to the illumination means occupies a
different part of the captured image from specular reflection
caused by ambient light.
[0045] In some cases the ambient light may be insufficient to allow
a fully exposed ambient image to be captured. In this case, there
will be no data available to paste into the first image to remove
specular reflections when the illumination means lies within the
field of view of the detector.
[0046] In an alternative arrangement for removing glare produced by
the first illumination means, the apparatus may conveniently
include a second illumination means for illuminating the document,
and in which the control means provides a first control signal to
the read-out means to capture a first image of the document when
illuminated by the first illumination means and the ambient light
but not the second illumination means to produce a first image,
provide an additional control signal to the read-out means to
capture an additional image of the document when the document is
illuminated by the second illumination means and the ambient light
substantially without being illuminated by the first illumination
means; and processing means adapted to process the first captured
image with the additional captured image to produce a final image
of the document.
[0047] The combination of the images can be performed in a number
of ways. However, the simplest is to produce a final image in which
the value of each data point in the final image is the minimum
value of the corresponding point in the first and additional
images. This is possible since it is known that glare points will
take a higher than normal value (possibly even saturating the
detector).
[0048] The first illumination means and the additional illumination
means may respectively comprise flash lights. They may be supported
in a fixed position relative to the detector.
[0049] It is also envisaged that at least one further ambient image
may be captured (taken in ambient light) at the same exposure as
the first and/or additional images captured with the first and
second illumination means, the or each additional image being
subtracted from the respective first image or additional image
prior to producing the final image of the document.
[0050] The apparatus may further include a stand which is adapted
to support the camera above a worksurface or the like.
[0051] The processing means may form an integral part of the
camera. Alternatively, the camera may be connected to a computer
which forms the processing means.
[0052] The image capture means may comprise a digital camera. The
illumination means may form a part of the camera or it may be
located remotely to the camera. For example, the camera may be
supported by a stand which also supports the illumination
means.
[0053] The camera may include a housing and the processing means
may be located within the housing to form a part of the camera.
[0054] In accordance with a second aspect the invention provides a
method of capturing an image of a document at least partially
illuminated by ambient light, the method comprising the steps
of:
[0055] providing a controllable light source;
[0056] capturing a first image of the document with the light
source illuminating the document at a first intensity;
[0057] capturing a second image of the document when illuminated by
the light source at a second, low, intensity; and processing the
first captured image with the second captured image to produce a
final image of the document.
[0058] Preferably, the second, lower, intensity comprises zero
intensity such that the second image is captured with the light
source switch off or pointed away from the document.
[0059] The first and second images may be captured with the same
exposure and the step of processing may comprise subtracting the
second image from the first image.
[0060] Alternatively, the first and second images may be captured
at different exposures such that both images are correctly exposed
and with the intensity of the light source during capture of the
second image substantially zero. In this case the step of
processing the images may comprise identifying areas or points in
the first image that correspond to glare spots and replacing those
points with values dependent upon the data at corresponding points
in the second image.
[0061] There will now be described, by way of example only, three
embodiments of the present invention of which:
[0062] FIG. 1 is a schematic illustration of a first embodiment of
a digital camera apparatus in accordance with the present
invention;
[0063] FIG. 2 summarises the steps required to obtain a final image
of a document from the digital camera apparatus of FIG. 1;
[0064] FIG. 3 summarises the steps required to process the captured
images obtained from the apparatus of FIG. 1;
[0065] FIG. 4 is a schematic illustration of a second embodiment of
a digital camera apparatus in accordance with the present
invention;
[0066] FIG. 5 summarises the steps required to process the captured
images obtained from the apparatus of FIG. 4; and
[0067] FIG. 6 is a schematic illustration of a third embodiment of
a digital camera apparatus in accordance with the present
invention
[0068] As illustrated in FIG. 1 of the accompanying drawings, a
digital image capture apparatus in accordance with the invention
comprises a digital camera 101 placed directly above a document
102. The camera is maintained in fixed position by a stand (not
shown) vertically above the document, which is placed on a flat
surface 104 such as an area of deskspace.
[0069] The digital camera comprises a lens 105 located on a lower
face of the camera 101. Also positioned on the lower face of the
camera-to one side of the lens, is a light source 106. The light
source emits a controlled beam of light downwards towards the
document 102 which is then reflected upwards from the document. The
field of view of the lens 105 and the angle at which the light is
emitted ensures that any specularly reflected light from the light
source 106 is not collected by the lens.
[0070] The digital camera 101 further includes a detector 107
located behind the lens 105. This comprises an array of image
collecting nodes which each produce an output signal when
illuminated. The magnitude of the signal emitted by each node
depends on the amount of light incident upon that node and also the
wavelength (or wavelengths) of the light.
[0071] The digital camera 101 further includes a read-out circuit
108 which is adapted to produce a captured image corresponding to
the output from each node in the array. This captured image
comprises an array of data values which represents the image of the
document that is directed onto the detector by the lens. The
integration period of the read-out circuit also acts as an
electronic shutter for the camera to control the exposure. Of
course, a mechanical shutter may be provided if preferred. In fact,
it is envisaged that the invention will offer increased benefits in
applications which employ mechanical shutters as they are typically
slower and so allow a greater build up of ambient light.
[0072] The camera read-out circuit 108 is connected to a computer
109 by an appropriate electrical cable 110. This cable 110 carries
the captured image data to the remote computer 109 where it is
stored in an area of memory 111. The cable 110 also carries control
signals from the computer 109 to the camera. These signals turn the
light source ON or OFF as required, and also control the read-out
circuit to initiate the capture of an image.
[0073] The computer 109 includes a processor 112, a display 113 and
an input means 114 such as a keyboard or mouse. The processor 112
is connected to the input means, the display and the memory by a
communications bus (not shown) in a known manner and also to the
memory 111 in which the captured images are stored.
[0074] In use of the apparatus, the document 102 is positioned
below the camera 101 and operation is initiated by a user pressing
a button on the keyboard or mouse 114. Upon receipt of the user
input the processor 112 issues a sequence of control signals to the
digital camera 101 along the cable 109 to complete the capture
process as illustrated in FIG. 2 of the accompanying drawings.
[0075] After activating 200 the camera the processor initially
issues a control signal 201 to the camera to activate the light
source, the detector and the read-out circuit. The light source is
switched ON and the processor issues a control signal to the
read-out circuit to capture 202 a first image of the document. This
image is downloaded to the computer 109 along the cable 110 and
stored in the memory 111. The light source need not be constantly
switched on during the image capture.
[0076] The processor 112 next issues a second control signal to the
light source to turn it OFF 203 and issues a control signal to the
read-out circuit to capture 204 a second image. The only source of
illumination of the document is ambient light. Of course, the order
of capture of the two images could be reversed without any effect.
The second image is also downloaded to the processor and stored in
the memory.
[0077] Having captured the two images the processor combines 205
the images to produce a final captured image which represents an
electronic copy of the document placed below the camera.
[0078] The final image is stored in the memory and the user is
prompted 206 to initiate capture of a different image or to end the
capture process. The stored final image can be subsequently
displayed or printed or subjected to additional processing if
desired. If the user indicates that no more images are needed the
camera is deactivated and the process ends 207.
[0079] The camera may capture the second image at the same exposure
as the first image (i.e. using the same shutter speed and lens
settings). In this case the contribution to both the first and the
second images made by ambient light will be the same.
[0080] Where the first and second images are captured with the same
exposure settings, the processing means may combine the two images
in the manner illustrated in FIG. 3 of the accompanying drawings.
After capturing 301 the two images at the same exposure, the
processor subtracts 302 the value of each data point in the second
image from the value of the corresponding data point in the first
image. The resulting value is allocated 303 to the corresponding
data point in the final image. This is repeated 304 choosing
different pixels 305 until all data points have been processed 306
(i.e. all pixels in the final image have been determined). The
final image is then free to the effects of ambient light.
[0081] In an optional step, the captured images amy be passed
through a low pass filter 307 prior to combining. This reduces the
amount of noise present in the images.
[0082] An alternative arrangement is illustrated in FIG. 4 of the
accompanying drawings. This differs from FIG. 1 only in so far as
the light source is within the field of view of the lens, and in
that the processor combines the images in a different way. For
clarity, the same reference numerals as used for FIG. 1 have been
used but incremented by 300. An optional shield 415 is also shown
around the light source which prevents specular reflections due to
ambient light occupying the same region of the first image as the
specular reflections due to the light source.
[0083] In this arrangement, the second image may be captured at a
longer exposure than the first image so that both captured images
are fully exposed.
[0084] Obviously, a longer exposure is required for the second
image since the amount of light reflecting from the document in the
second image is less than the first image. By fully exposed we mean
neither under or over exposed.
[0085] In this case, the processor may combine the two images in
the manner indicated in FIG. 5 of the accompanying drawings,
[0086] After capturing 501 two images which are correctly exposed
(i.e. neither under nor over exposed), the processor identifies 502
pixels in the first or the second image which correspond to
specular reflected light from either the controlled light source or
the ambient source. This identification can be performed simply by
looking for bright spots in the images (which may even be saturated
in many cases).
[0087] Having identified the bright spots the final image is
constructed 503 by combining the two images such that any pixels
identified as glare spots in the first image are replaced by
corresponding pixels in the second image.
[0088] The replacement or pasting of pixels is continued for the
whole image until all the specular areas have been replaced. The
pixels around the saturated points may also be blended 505 together
to provide a smooth transition from the original areas to the
replaced areas. Once all have been replaced 505 the process is
ended 506 and the final image stored in the memory 111.
[0089] Several refinements can be applied in this case. Firstly, a
third image obtained using only ambient light and at the same
exposure as the first image can be captured. This image is then
processed with the first image as set out in FIG. 2 prior to
replacing the specular pixels in the (processed) first image with
data values from the second image.
[0090] Secondly, the processor may compare the data in the first
image and the second image to construct a map or transform which
compensates for differences in colour temperature between the
ambient light and the controlled light source. This colour
transform can be calculated for regions of the image where pixels
do not need to be replaced. Whenever a pixel is replaced in the
first image the new pixel taken from the correctly exposed ambient
image is passed through the colour transform before it is pasted
into the final image.
[0091] Furthermore, a transform may also be used to ensure that any
misalignment between the images is removed prior to pasting in the
replacement pixels. This is achieved by identifying the location of
features in each image and mapping the features in the second image
onto the features of the first image.
[0092] A third embodiment of the present invention is illustrated
in FIG. 6 of the accompanying drawings. This differs from FIG. 1
only in so far as the two identical light sources 606a and 606b are
provided within the field of view of the lens. For clarity, the
same reference numerals as used for FIG. 1 have been used but
incremented by 500.
[0093] In this embodiment, the read-out circuit captures a first
image of the document illuminated by the first light source and a
second image captured when the document is illuminated by the
second light source. An additional ambient image is also captured
in which the first and second light sources are extinguished so
that the only source of illumination is the ambient light. Notably,
the exposure times of the first and second images are chosen so
that the images are fully exposed. The ambient image is captured
with the same exposure as the first two images (and is consequently
underexposed).
[0094] As with the first embodiment, the ambient image is
subtracted from each of the first and second captured images to
produce respective first and second processed captured images. The
effect of ambient light will have been removed from these processed
images.
[0095] Subsequently, points of glare in the first image are
identified and are replaced by corresponding data values taken from
the second processed image to form the final image.
[0096] Of course, if the two light sources are not identical (i.e.
illuminate the document at different intensities) the exposure
required for each of the first and second images will be different.
In that case, two ambient images may be captured. A first one
should be taken at the same exposure as the first captured image
and the second at the same exposure as the second captured
image.
* * * * *