|
Digital Camera Patent Abstract
Smear detect circuitry within an analog front end (AFE) of a digital
camera determines when black area pixel values received from an
image sensor are indicative of smear leakage. Smear leakage can
cause a light vertical line in the resulting digital image. When
a sensor that is coupled to a storage element is exposed to a bright
light source, storage element overload can cause a leakage charge
to leak from the storage element to other storage elements along
a transfer line. Smear detect circuitry identifies the transfer
line exhibiting smear leakage and excludes pixel values from storage
elements along that transfer line from the calculation of a black
level value used to calibrate color pixel values. The digital camera
displays a smear icon indicating smear leakage in a digital image
that is to be taken. A digital file of the digital image includes
a header with a smear detect field.
Digital Camera Patent Claims
32. A method comprising: displaying a digital image and an icon
on a display of a digital camera, wherein the icon indicates smear
leakage.
33. The method of claim 32, wherein the icon indicates that the
digital image contains smear leakage, and wherein the digital image
is overexposed.
34. The method of claim 32, wherein the icon indicates that the
digital image contains smear leakage, and wherein the digital image
is underexposed.
35. The method of claim 32, wherein the digital image represents
a real-world image, wherein a related digital image also represents
the real-world image, wherein the icon indicates that the related
digital image contains smear leakage, and wherein the smear leakage
appears as a light line passing through a source of bright light
in the related digital image.
36. The method of claim 35, wherein the digital image is acquired
by the digital camera in a viewfind mode, and wherein the related
digital image is acquired by the digital camera in a frame readout
mode.
37. The method of claim 32, wherein the icon is superimposed onto
the digital image.
38. The method of claim 32, wherein the digital camera is taken
from the group consisting of: a digital still camera, a digital
video camera and a cell phone containing a digital camera.
39. A method comprising: storing a digital image as a digital file
on a digital camera, wherein the digital file has a header, and
wherein the header includes a smear detect field.
40. The method of claim 39, wherein the smear detect field contains
information indicating whether the digital image contains smear.
41. The method of claim 40, wherein the digital image contains
smear when a light line passes through a source of bright light
in the digital image.
42. The method of claim 39, wherein the smear detect field is one
bit Wide.
43. The method of claim 39, further comprising: assigning a filename
to the digital file that includes a code indicating smear.
44. The method of claim 39, wherein the digital file is a jpg file.
45. A device comprising: smear detect circuitry that outputs a
smear detect signal, wherein the smear detect signal indicates whether
a digital image contains smear; and means for storing the digital
image as a digital file with a header, wherein the header includes
a smear detect field.
46. The device of claim 45, wherein the smear detect field is one
bit wide, and wherein the means sets the one bit to a predetermined
digital state when the smear detect signal is asserted.
47. The device of claim 45, wherein the means includes an analog
front end (AFE) of a digital camera.
48. The device of claim 45, wherein the smear detect signal indicates
that the digital image contains smear, and wherein the digital image
is overexposed.
49. The method of claim 45, wherein the smear detect signal indicates
that the digital image contains smear, and wherein the digital image
is underexposed.
Digital Camera Patent Description
TECHNICAL FIELD
[0001] The present invention relates digital imaging and, in particular,
to detecting smear leakage that results when an image sensor is
exposed to a bright light source.
BACKGROUND
[0002] When a digital photograph is taken of an image that includes
a bright light source, a light vertical line often appears in the
digital image. The light vertical line results from "smear"
leakage caused by the bright light source. The bright light source
can cause smear leakage from an overloaded storage element to an
adjacent storage element of an image sensor in a digital camera.
FIG. 1 illustrates a digital image 10 that includes a light vertical
line 11 caused by smear leakage. In this example, the smear leakage
is due to the bright light source of the sun in the real-world image
that was photographed. In addition to light vertical line 11, the
colors in digital image 10 may also not accurately reflect the colors
in the real-world image because the bright light source affects
the black level calibration used to correlate digital pixel data
to specific colors. For example, the tree in the original photographed
image of FIG. 1 may appear in digital image 10 as blue instead of
green.
[0003] An apparatus is sought for detecting and indicating the
presence of smear leakage in an image sensor. An apparatus is also
sought that reduces the smear-induced deviation of colors in a digital
image from the true colors in the corresponding real-world image.
SUMMARY
[0004] The black level calibrator of an analog front end (AFE)
integrated circuit of a digital camera includes smear detect circuitry.
The smear detect circuitry determines when black area pixel values
received from an image sensor of the digital camera are indicative
of smear leakage. The black area pixel values are obtained from
storage elements in an optical black area of the image sensor that
is not exposed to light. Smear leakage causes a light vertical line
in the digital image output by the digital camera. Smear leakage
occurs in the image sensor when a sensor that is coupled to a storage
element is exposed to a bright light source. The bright light source
can result in storage element overload that causes a leakage charge
to leak from the storage element to other storage elements along
a transfer line. Smear leakage can even leak to storage elements
in the optical black area and hamper the calculation of the black
level value used to calibrate color pixel values. Using an incorrect
black level value to calibrate color pixel values can result in
a digital image with "crazy" colors.
[0005] A state machine in the smear detect circuitry distinguishes
multiple, consecutive black area pixel values that exceed a predetermined
threshold from other black area pixel values that occasionally exceed
the threshold. Multiple, consecutive pixel values from the optical
black area that exceed the threshold are indicative of smear leakage
along a transfer line into the optical black area. In one embodiment,
the smear detect circuitry identifies the transfer line that exhibits
smear leakage and excludes pixel values from storage elements along
that transfer line from the calculation of the black level value.
In another embodiment, only black area pixel values that exceed
the threshold are excluded from the calculation of the black level
value.
[0006] In another embodiment, the digital camera displays a smear
icon indicating storage element overload and smear leakage in a
digital image that is to be taken or that has been taken. In an
embodiment where the pixel data that is corrupted by smear leakage
is not used, the smear icon warns the photographer to take another
picture. Where the corrupted pixel data is used, the smear icon
indicates that the resulting digital image contains smear noise.
The digital image is then stored in the digital camera as a digital
file. The digital file includes a header with a smear detect field.
A bit in the smear detect field indicates whether the digital image
exhibits storage element overload. In addition, a code may be included
in the filename assigned to the digital file containing the digital
image that exhibits smear leakage.
[0007] Other embodiments and advantages are described in the detailed
description below. This summary does not purport to define the invention.
The invention is defined by the claims.
BRIEF DESCRIPTION OF THE DRAWINGS
[0008] The accompanying drawings, where like numerals indicate
like components, illustrate embodiments of the invention.
[0009] FIG. 1 is a digital image containing a light vertical line
caused by smear leakage.
[0010] FIG. 2 is a simplified, schematic diagram of an analog front
end of a digital camera with a black level calibrator according
to an embodiment of the invention.
[0011] FIG. 3 is a simplified, schematic diagram of an image sensor
with an optical black area.
[0012] FIG. 4 is a more detailed diagram of storage elements, sensors
and a vertical transfer line of the image sensor of FIG. 3.
[0013] FIG. 5 is a diagram of a vertical transfer line of the image
sensor of FIG. 3 in which charge coupled devices implement both
storage and switching functions.
[0014] FIG. 6 is a waveform diagram illustrating the pulse signals
used for switching along the transfer lines of FIG. 4.
[0015] FIG. 7 is a simplified, schematic diagram of the image sensor
of FIG. 3 being exposed to an image with a bright light source.
[0016] FIGS. 8A-B show a smear icon on an on-screen display of
the digital camera of FIG. 2.
[0017] FIG. 9 is a more detailed diagram of the black level calibrator
of FIG. 2 including smear detect circuitry.
[0018] FIG. 10 is a more detailed diagram of the smear detect circuitry
of FIG. 9 including a state machine.
[0019] FIG. 11 is a diagram illustrating the transitions between
states of the state machine of FIG. 10.
[0020] FIG. 12 is a waveform diagram illustrating the operation
of the smear detect circuitry of FIG. 9.
DETAILED DESCRIPTION
[0021] Reference will now be made in detail to some embodiments
of the invention, examples of which are illustrated in the accompanying
drawings.
[0022] FIG. 2 is a simplified diagram of a high-resolution digital
camera 12 that exhibits storage element overload and smear leakage.
In an example of the operation of digital camera 12, a photographer
points digital camera 12 at a real-world image 13 that is to be
photographed. Image 13 contains a source of bright light, the sun
in this example. Image 13 passes through a lens 14 and is captured
by an image sensor 15. Image sensor 15 outputs analog pixel data
16 that includes pixel values corresponding to charge in individual
storage elements of image sensor 15. An analog front end (AFE) integrated
circuit 17 receives the analog pixel data 16 from image sensor 15.
[0023] AFE integrated circuit 17 includes a timing generator portion
18, a correlated double sampling (CDS) mechanism 19, an analog-to-digital
converter (ADC) 20, a decimation circuit 21, a black level calibrator
22, a signal processing block 23, a digital image processing (DIP)
interface 24 and a clock generator 25. Timing generator portion
18 supplies vertical pulse signals 26 and horizontal pulse signals
27 to image sensor 15 in order to read out analog pixel data 16.
Image sensor 15 requires the voltage minimums and voltage maximums
of vertical pulse signals 26 to extend outside the voltage range
that can be supplied by AFE integrated circuit 17. Vertical pulse
signals 26 output from AFE integrated circuit 17 are therefore supplied
to a vertical driver 28 that performs level shifting to the voltage
levels required by image sensor 15.
[0024] CDS 19 receives analog pixel data 16 from image sensor 15.
Each pixel value of analog pixel data 16 is typically in the form
of a pair of analog level signals. The first analog level signal
indicates the unique reference voltage level of the particular pixel,
and the second analog level signal indicates the color brightness
level of the pixel. CDS 19 determines the analog signal magnitude
between the reference level and the brightness level. ADC 20 digitizes
analog signal magnitude and outputs the digital result, which is
received by decimation circuit 21. Decimation circuit 21 outputs
decimated, digitized pixel data 29, which is received by black level
calibrator 22. Black level calibrator 22 determines a black level
calibration value of decimated, digitized pixel data 29 using pixel
data from sensors of image sensor 15 that are not exposed to light.
Black level calibrator 22 then calibrates AFE 17 by subtracting
the calibration value from the pixel values of pixel data 29 to
generate calibrated, decimated and digitized pixel data 30. Black
level calibrator 22 then passes the calibrated, decimated and digitized
pixel data 30 to signal processing block 23 and on to DIP interface
24. DIP interface 24 then outputs digitized image data 31 to a digital
image processing (DIP) ASIC 32.
[0025] DIP ASIC 32 performs image processing on digitized image
data 31 and then typically causes a digital image 33 to be displayed
on a display 34 of digital camera 12. In the example of FIG. 2,
smear leakage occurs between storage elements of image sensor 15
as real-world image 13 is captured. Smear leakage within image sensor
15 is manifested as a light vertical line 35 in digital image 33.
DIP ASIC 32 also stores digital image 33 as a digital file 36 on
a storage medium 37 within digital camera 12. Digital file 36 may,
for example, be a jpg file. The presence of smear in digital image
33 is indicated by a smear detect field 38 in the header of digital
file 36. A microcontroller 39 provides overall key scanning, control
and configuration functions for digital camera 12. Microcontroller
39 is coupled to DIP ASIC 32 via a control bus 40. Microcontroller
39 controls lens 14 via motor driver circuitry 41.
[0026] FIG. 3 shows image sensor 15 of digital camera 12 in more
detail. Image sensor 15 may, for example, be a charge coupled device
(CCD) sensor, a CMOS sensor, another type of pixilated metal oxide
semiconductor sensor or another type of image sensor. In this example,
image sensor 15 is a CCD sensor with a two-dimensional array of
sensors. In the illustration, the sensors are denoted as squares,
where each square contains a letter. A square that contains a "G"
is a sensor for green. A square that contains an "R" is
a sensor for red. A square that contains a "B" is a sensor
for blue. A square that contains a "Y" is a sensor for
a fourth color, such as yellow. Reference numeral 43 identifies
one such sensor for green. In one embodiment, the sensors for all
of the colors have the same structure. The various sensors are covered
by filters that allow only the appropriately colored light to reach
each sensor. In this example, sensors in the bottom three rows are
not designated as colored. These bottom rows of sensors fall within
an optical black area 44 of image sensor 15. The bottom rows of
sensors are actually at the top of the captured image because lens
14 inverts the image. Sensors within optical black area 44 are typically
covered such that they are not exposed to light.
[0027] In response to a shutter signal, each of the sensors of
image sensor 15 takes a sample of light. The sample is retained
in the sensor in the form of a charge. The magnitude of the charge
indicates the sample value. The charge values are read out of image
sensor 15 in serial fashion as a sequence of pixel values by supplying
vertical pulse signals 26 and horizontal pulse signals 27 to switches
within image sensor 15. In the example of FIG. 3, each sensor has
an associated storage element located to its left. Reference numeral
45 identifies the storage element for sensor 43. At one time, the
sample charges from all the sensors are transferred right to left
into the associated storage elements. A vertical pulse signal is
then applied to switches associated with columns of storage elements.
This causes the sample charge in each storage element to be shifted
down to the storage element below it. Reference numeral 46 identifies
a column of sensors and associated storage elements, including sensor
43 and storage element 45. For example, the sample charge in storage
element 45 is shifted down to a storage element 47 below it in column
46. In a similar manner, the sample charge is shifted down the entire
column 46.
[0028] The sample charge in the bottom-most row of storage elements
passes into a readout row 48 of storage elements at the bottom of
image sensor 15. Readout row 48 is a horizontal transfer line. Once
readout row 48 contains a set of charges, a plurality of horizontal
pulse signals 27 is applied to switches associated with readout
row 48. These horizontal pulses cause the sample charges in the
storage elements of readout row 48 to be shifted out of image sensor
15 one-by-one. When the complete row of sample charges has been
shifted out of image sensor 15, then another vertical pulse is applied
in order to load readout row 48 with the next row of sample charges
to be read out. This process of supplying a vertical pulse, and
then shifting out the bottom row of sample charges is repeated until
all the sample charges are read out of image sensor 15.
[0029] FIG. 4 shows column 46 of image sensor 15 in more detail
and illustrates an operation of column 46. Column 46 includes a
vertical transfer line 49 with two alternating sets of switches.
In one embodiment, vertical transfer line 49 is an analog shift
register. To transfer a charge from a storage element 50 to a storage
element 51, switches 52 and 53 are kept open and a switch 54 is
closed. This allows charge from storage element 50 to pass through
conductive switch 54 along vertical transfer line 49 and into storage
element 51. It is therefore seen that adjacent switches in column
46 are opened and closed in alternating fashion to shift a sample
charge down vertical transfer line 49. In one embodiment, storage
element 50 is a semiconductor depletion capacitor formed from a
field effect transistor. Switch 54 is also formed from a field effect
transistor manufactured in the same process as is storage element
50. Although FIG. 4 is a very simplified diagram of a vertical transfer
bus, more complex configurations of vertical transfer busses operate
in an analogous manner. For example, in another embodiment, both
the storage and switching functions are implemented by charge coupled
devices (CCDs). Charge is transferred from a first CCD to a second
CCD in response to a pulse signal by lowering the bias voltage of
the second CCD lower than the bias voltage of the first CCD.
[0030] FIG. 5 shows column 46 of image sensor 15 in which both
the storage and switching functions are implemented by charge coupled
devices (CCDs). In the embodiment of FIG. 5, vertical transfer line
49 is a row of CCDs.
[0031] FIG. 6 is a waveform diagram that illustrates vertical pulse
signals 26 and horizontal pulse signals 27 used to read analog pixel
data 16 out of the sensor array of image sensor 15. FIG. 6 shows
the alternating fashion of pulses in two vertical pulse signals
VPULSE1A and VPULSE1B that control the two alternating sets of switches
of FIG. 4, including switches 52, 53 and 54. FIG. 6 also shows two
horizontal pulse signals HPULSE1A and HPULSE1B that control the
switches associated with readout row 48, including a switch 55 and
a switch 56. After vertical pulse signals 26 shift a row of sample
charges into readout row 48, a complete set 57 of horizontal shift
pulses of horizontal pulse signals HPUSEL1A and HPULSE1B shifts
the sample charges out of readout row 48. The process repeats with
each vertical shift being followed by a set 57 of horizontal shift
pulses.
[0032] The state of the art in CCD image sensors has advanced well
beyond the simple examples set forth in FIGS. 4-6. CCD image sensors
typically have multiple modes including, for example, a high frame
rate readout mode, a frame readout mode (also called the capture
mode), an autoexposure mode and an autofocus mode. As a result,
more complex timing signals are often required to drive contemporary
CCD sensors than the signals shown in FIG. 6. The high frame rate
readout mode may, for example, be used in a hybrid camera when the
hybrid camera is used to capture video, whereas the higher resolution
capture mode may be used when the hybrid camera is used to take
still pictures. For example, the higher resolution capture mode
typically allows the sensors to be exposed to the real-world image
longer than in the autofocus mode.
[0033] Smear leakage results when charge from one storage element
leaks to another storage element. For example, a leakage charge
can leak from one storage element to an adjacent storage element
along a vertical transfer line even though a pulse signal has not
closed the switch between the two storage elements. Returning to
FIG. 4, a leakage charge 58 leaks from storage element 50 along
vertical transfer line 49 into storage element 51 even though switch
54 has not been closed in response to vertical pulse signal VPULSE1B.
One cause of leakage charge 58 is an excessive charge buildup across
storage element 50 that results when a sensor 59 adjacent to storage
element 50 is exposed to a bright light source 60. When a large
charge builds up across the semiconductor depletion capacitor of
storage element 50, the depletion area around storage element 50
may push charge as far as switch 54, allowing switch 54 to become
conductive. Leakage charge 58 may then leak along vertical transfer
line 49 to adjacent storage elements in a cascading fashion. In
this manner, all of the storage elements coupled to a vertical transfer
line may become highly charged although only a few of the associated
sensors were exposed to the bright light source. Storage element
overload may also result in charge leaking from one storage element
directly to an adjacent storage element without passing through
a switch or along a transfer line.
[0034] FIG. 7 illustrates the bright light source of the sun in
image 13 being focused by lens 14 onto sensor 59 of image sensor
15. Excessive charge builds up across the capacitor of storage element
50 resulting in storage element overload. Leakage charge 58 leaks
onto adjacent storage elements and storage elements that are coupled
to vertical transfer line 49. Although a sensor 61 is within optical
black area 44 and is not exposed to any light, storage element 51,
which is associated with sensor 61, is highly charged. Similarly,
although the light source from image 13 is less intense (darker)
at a sensor 62, the storage element associated with sensor 62 is
also highly charged. Analog pixel data 16 output by image sensor
15 results in the digital image 33 of FIG. 2 if digital camera 12
does not correct for the storage element overload. Digital image
33 has light vertical line 35 running through the darker area of
the tree in image 13. Light vertical line 35 may be several vertical
transfer lines wide where the bright light source also overloads
the sensors to the right and left of sensor 59 and thereby charges
the storage elements coupled to those vertical transfer lines in
a cascading fashion.
[0035] Smear leakage can reduce the quality of digital image 33
in two ways: first, by producing light vertical line 35 and second,
by producing "crazy" colors. Smear leakage can incorrectly
increase the black level used to interpret color data in the decimated,
digitized pixel data 29. Where an incorrect average black level
is subtracted from pixel data 29, DIP ASIC 32 interprets the color
data incorrectly. Digital image 33 then appears with "crazy"
colors. For example, the sky in digital image 33 might be green,
and the tree might be orange.
[0036] Digital camera 12 uses black level calibrator 22 to correct
for these two problems. The photographer may not wish to have light
vertical line 35 in digital image 33 because the vertical line was
not in original image 13. Smear leakage may not be apparent to the
photographer looking at a digital image on display 34 in a faster
viewfind mode, such as the autofocus or autoexposure modes. The
exposure time in those modes is typically shorter, and there is
less time for a bright light source to overfill storage elements.
In modes with shorter exposure periods, it is less likely that leakage
charge will cascade to other storage elements along a vertical transfer
line. In the viewfind mode, for example, storage element overload
may result in a shorter and less pronounced smear line.
[0037] If black level calibrator 22 detects smear leakage, digital
camera 12 can reduce the aperture (F stop) to reduce smear leakage
in the next frame of analog pixel data 16. For example, where digital
camera 12 is in the autoexposure mode, black level calibrator 22
detects smear and transmits a smear detect signal 63 to an interrupt
generator 64 that interrupts microcontroller 39. Digital camera
12 then recaptures real-world image 13 a second time with a reduced
aperture. Storage element overload is less likely to occur in the
second exposure with a smaller aperture. Pixel values obtained from
the first exposure that caused storage element overload are not
used to generate digital image 33. This procedure can be repeated
iteratively until an aperture is used that does not result in smear
leakage.
[0038] When digital camera 12 is not in a viewfind mode, the photographer
is warned that digital image 33 contained smear leakage so that
the photographer can retake the picture. The photographer may then
point the camera away from the bright light source. For example,
even where a beach scene might result in an overexposed digital
image, the photographer can nevertheless avoid storage element overload
and the resulting light vertical line by not including the sun in
the picture. In some cases, the photographer may wish to retain
vertical line 35 as a visual effect. For example, an underexposed
candlelight dinner scene may have light vertical lines through the
flames of the candles. Digital images with vertical lines can be
given a smear indication in the filename of the jpg file under which
they are stored in storage medium 36. The photographer can then
later identify which digital images contain the smear visual effect.
In addition, digital files containing images with smear also include
a smear indication in their file headers. For example, a bit in
smear detect field 38 indicates that the digital image contained
in digital file 36 exhibits storage element overload.
[0039] FIGS. 8A-B show a smear icon 65 on display 34 of digital
camera 12. Digital camera 12 displays smear icon 65 when black level
calibrator 22 detects smear leakage. When microcontroller 39 is
interrupted in response to smear detect signal 63 being asserted,
microcontroller 39 activates on-screen display logic that causes
smear icon 65 to be superimposed on the image being displayed on
display 34. In FIG. 8A, for example, smear icon 65 is superimposed
onto digital image 33 that includes light vertical line 35. Smear
icon 65 indicates that light vertical line 35 resulted from smear
leakage and not, for example, from the sun being reflected at a
vertical angle from lens 14 of digital camera 12. In FIG. 8B, smear
icon 65 appears on display 34 in the viewfind mode before the photographer
captures digital image 33. The appearance of smear icon 65 in a
viewfind image 66 on display 34 warns the photographer that taking
a picture with the selected aperture and shutter settings will result
in a digital image exhibiting smear leakage.
[0040] FIG. 9 is a simplified block diagram of black level calibrator
22 that correctly calibrates the black level value even from analog
pixel data 16 that contains storage element overload. Black level
calibrator 22 includes smear detect circuitry 69, a black level
generator 70, calibration registers 71, a black area generator 72
and a smear area generator 73. Decimation circuit 21 outputs decimated,
digitized pixel data 29, which is received by smear detect circuitry
69 and by black level generator 70. In this embodiment, pixel data
29 is sixteen bits wide. Black level generator 70 calibrates AFE
integrated circuit 17 by outputting a black level value 74 that
is an average of black area pixel values not affected by smear leakage.
The averaging function is performed by registers 75 and an adder
76. In other embodiments, black level value 74 is a weighted average,
an interpolated value or some other value derived from black area
pixel values. Smear detect circuitry 69 determines which black area
pixel values of analog pixel data 16 correspond to storage elements
influenced by smear leakage. Upon detecting smear leakage, smear
detect circuitry 69 outputs smear detect signal 63 that disables
black level generator 70 such that some or all black area pixel
values influenced by smear leakage are not included in the running
average calculation of black level value 74. Reference values 77-80
that are based on black level value 74 are stored in calibration
registers 71. One of reference values 77-80 is derived for each
color of sensor in image sensor 15. For example, registers CAL0,
CAL1, CAL2 and CAL3 may contain reference values for red, green,
blue and yellow sensors, respectively. When black level calibrator
22 receives pixel values that are not black area pixel values, the
reference values 77-80 are subtracted from the pixel value from
the correspondingly colored sensor. calibration registers 71 receive
a color ID signal 81 that identifies the color to which each pixel
value of pixel data 29 corresponds. By excluding pixel values that
are affected by storage element overload from the black level calibration,
the reference values 77-80 are more accurate, and DIP ASIC 32 is
less likely to interpret a pixel value of calibrated pixel data
30 as an inaccurate color.
[0041] FIG. 10 shows smear detect circuitry 69 of black level calibrator
22 in more detail. Smear detect circuitry 69 includes a state machine
82, a comparator 83 and three registers 84-86. Comparator 83 receives
each 16-bit value of decimated, digitized pixel data 29 on sixteen
input leads. In another embodiment, decimation circuit 21 is disabled,
and comparator 83 receives digitized pixel data with the same sampling
point as used by ADC 20. In addition, comparator 83 receives a 16-bit
threshold value (THLD) on an additional set of sixteen input leads
from register 84. The threshold value (THLD) is written to register
84 by microcontroller 39 over a data bus 87. Comparator 83 also
receives a valid-data-in signal (DIN_VLD) that is deasserted when
a pixel value of pixel data 29 corresponds to a defective sensor
or storage element and to a storage element outside of optical black
area 44. Thus, comparator 83 outputs a logic signal 88 that is a
digital low for all pixel values corresponding to storage elements
outside of optical black area 44.
[0042] Logic signal 88 is a digital high when a pixel value of
pixel data 29 is greater than threshold value (THLD). Threshold
value (THLD) is programmable to correspond to a usual charge magnitude
from a storage element associated with a sensor that is not exposed
to light in optical black area 44. A pixel value from optical black
area 44 might nevertheless exceed threshold value (THLD) for a number
of reasons. For example, a defective sensor might overcharge a storage
element and result in a pixel value that is too high. Heat may also
increase a pixel value. A pixel value from a storage element in
optical black area 44, however, may also be increased by a leakage
charge from a storage element outside optical black area 44. To
distinguish high pixel values that result from storage element overload
from other high pixel values that result from defective pixels and
other causes, smear detect circuitry 69 employs state machine 82.
[0043] State machine 82 transitions from a normal condition to
a smear condition when pixel data 29 exceeds threshold value (THLD)
for longer than a first time period. State machine 82 asserts smear
detect signal 63 in the smear condition. The state machine 82 transitions
back to the normal condition when pixel data 29 falls below threshold
value (THLD) for longer than a second time period. Two 4-bit reference
values that are written to registers 85 and 86 define the first
time period and the second time period, respectively. A reset signal
(RST_FLG) returns state machine 82 to the normal condition before
pixel values from each subsequent transfer line are analyzed.
[0044] FIG. 11 illustrates the possible transitions between states
of state machine 82. State machine 82 is in the normal condition
in states 0, 1, 2 and 3 and in the smear condition in states 4,
5 and 6. Reset signal (RST_FLG) returns state machine 82 to state
0 before smear detect circuitry 69 analyzes a sequence of pixel
values associated with each additional transfer line of image sensor
15. In this example, state machine 82 transitions from state 0 to
state 4, and from the normal condition to the smear condition, when
logic signal 88 remains high for four consecutive pixel values of
pixel data 29. Thus, the 4-bit reference value (L2H_TIME) that is
written to register 85 is 0100. If logic signal 88 goes low before
it remains high for four consecutive pixel values, then state machine
82 is returned to state 0. State machine 82 is returned from the
smear condition to state 0 when logic signal 88 remains low for
three consecutive pixel values. Thus, the 4-bit reference value
(H2L_TIME) that is written to register 86 is 0011.
[0045] FIG. 12 is a waveform diagram illustrating the operation
of state machine 82. FIG. 12 shows that state machine 82 does not
assert smear detect signal 63 when a sequence of black area pixel
values 89 of pixel data 29 exceeds threshold value (THLD) over a
period 90 of two pixel values. Smear detect signal 63 is, however,
asserted when sequence of black area pixel values 89 exceeds threshold
value (THLD) over a period 91 that extends over at least four pixel
values. Smear detect signal 63 is then deasserted when sequence
of black area pixel values 89 falls below threshold value (THLD)
over three consecutive pixel values. FIG. 12 also shows an optical
black area ID signal (OB_AREA_ID) 92.
[0046] Black area generator 72 generates optical black area ID
signal 92, which is asserted for those pixel values that correspond
to storage elements within optical black area 44. Returning to FIG.
9, a register 93 within black area generator 72 is programmable
to identify the storage elements of each transfer line that lie
within optical black area 44. For example, optical black area 44
in FIG. 7 is the first three storage elements of each transfer line
after readout row 48. In other embodiments, the optical black area
can be the last N storage elements at the top of the image sensor.
The black area can even be at the side of the image sensor if the
readout line runs vertically along one side of the image sensor.
Black level generator 70 is enabled and includes pixel values in
the calibration calculation only when black area ID signal 92 is
asserted and smear detect signal 63 is deasserted.
[0047] FIG. 12 shows that smear detect signal 63 is asserted only
after four consecutive pixel values of sequence of black area pixel
values 89 have exceeded threshold value (THLD). Although the subsequent
pixel values that exceed threshold value (THLD) are excluded from
the calculation to determine black level value 74, those four pixel
values may nevertheless also skew the calculation of black level
Value 74. A buffer 94 (as shown in FIG. 9) in black level generator
70 stores several pixel values of sequence of black area pixel values
89 and allows the determination of black level value 74 to be performed
with a delay of several pixel values. In this manner, several previous
pixel values (for example, four) can be excluded from the calculation
of black level value 74 after smear detect signal 63 is asserted.
[0048] In another embodiment, black level value 74 is recalculated
with pixel values from a subsequent exposure of image sensor 15.
Smear area generator 73 determines a smear area based on the pixel
values of the previous exposure that resulted in the assertion of
smear detect signal 63. When smear area generator 73 identifies
pixel values from a subsequent exposure as being within a smear
area, those pixel values can be immediately excluded from the recalculation
of black level value 74 without delaying the input of pixel values
using buffer 94. A register 95 in smear area generator 73 is programmable
with a parameter that defines a band of transfer lines on either
side of a transfer line with detected storage element overload.
All pixel values from transfer lines within the band of transfer
lines are then characterized as within the smear area and are excluded
from the recalculation of black level value 74.
[0049] Although the present invention has been described in connection
with certain specific embodiments for instructional purposes, the
present invention is not limited thereto. The smear detect circuitry
disclosed above detects storage element overload in a digital still
camera. In other embodiments, however, the smear detect circuitry
detects storage element overload in digital video cameras. Smear
detect circuitry is described above as detecting smear in pixel
data from an image sensor that senses four colors. In other embodiments,
smear detect circuitry detects smear in pixel data from multiple
image sensors, wherein each image sensor senses light of a different
color. Accordingly, various modifications, adaptations, and combinations
of various features of the described embodiments can be practiced
without departing from the scope of the invention as set forth in
the claims. |