U.S. patent application number 13/181666 was filed with the patent office on 2011-11-03 for imaging apparatus and processing method thereof.
This patent application is currently assigned to CANON KABUSHIKI KAISHA. Invention is credited to Kazuhiro Sonoda, Shintaro Takenaka.
Application Number | 20110267513 13/181666 |
Document ID | / |
Family ID | 40072014 |
Filed Date | 2011-11-03 |
United States Patent
Application |
20110267513 |
Kind Code |
A1 |
Sonoda; Kazuhiro ; et
al. |
November 3, 2011 |
IMAGING APPARATUS AND PROCESSING METHOD THEREOF
Abstract
An imaging apparatus is provided having: a pixel unit including
an effective pixel region of multiple pixels accumulating charges
generated according to incident light to output signals and an
ineffective pixel region of multiple pixels outputting signals not
dependent on incident light; a plurality of vertical signal lines
provided for each column of pixels of pixel unit; a vertical
scanning circuit that scans and selects pixels of pixel unit in row
units to output signals of selected pixels of same row to plurality
of vertical signal lines; and a horizontal scanning circuit that
scans and selects signals of plurality of vertical signal lines to
output signals of selected vertical signal lines; wherein the
vertical scanning circuit selects pixels of same row of effective
pixel region in row units once during one frame and selects pixels
of same row of ineffective pixel region in row units multiple times
during one frame.
Inventors: |
Sonoda; Kazuhiro;
(Yokohama-shi, JP) ; Takenaka; Shintaro;
(Yokohama-shi, JP) |
Assignee: |
CANON KABUSHIKI KAISHA
Tokyo
JP
|
Family ID: |
40072014 |
Appl. No.: |
13/181666 |
Filed: |
July 13, 2011 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
12121349 |
May 15, 2008 |
7999866 |
|
|
13181666 |
|
|
|
|
Current U.S.
Class: |
348/294 ;
348/E5.091 |
Current CPC
Class: |
H04N 5/361 20130101;
H04N 5/374 20130101 |
Class at
Publication: |
348/294 ;
348/E05.091 |
International
Class: |
H04N 5/335 20110101
H04N005/335 |
Foreign Application Data
Date |
Code |
Application Number |
May 21, 2007 |
JP |
2007-134495 |
Dec 27, 2007 |
JP |
2007-337461 |
Claims
1. An imaging apparatus comprising: a pixel unit including an
effective pixel region of a plurality of pixels accumulating a
carrier generated based on an incident light and outputting a
signal, and an ineffective pixel region of a plurality of pixels
outputting a signal independent of the incident light; a plurality
of vertical signal lines arranged each for each of columns of the
pixels of the pixel unit; a vertical scanning circuit for scanning
and selecting row by row, or rows by rows, the pixels of the pixel
unit, such that the signal of the pixels of the same selected row
is outputted to the plurality of vertical signal lines; and a
horizontal scanning circuit for scanning and selecting the signal
of the plurality of vertical signal lines such that the signal of
the selected signal line is outputted, wherein the vertical
scanning circuit selects at a single time row by row, or rows by
rows, the pixels of the same row of the effective pixel region
during a one frame time period, and selects at plural times row by
row, or rows by rows, the pixels of the same row of the ineffective
pixel region during the one frame time period.
Description
RELATED APPLICATIONS
[0001] This application is a divisional of application Ser. No.
12/121,349, filed May 15, 2008, claims benefit of that application
under 35 U.S.C. .sctn.120, and claims benefit under 35 U.S.C.
.sctn.119 of Japanese patent applications nos. 2007/134495, filed
May 21, 2007, and 2007/337461, filed Dec. 27, 2007. The entire
contents of each of those three earlier applications are
incorporated herein by reference.
BACKGROUND OF THE INVENTION
[0002] 1. Field of the Invention
[0003] The present invention relates to an imaging apparatus such
as a CMOS image sensor and a processing method thereof.
[0004] 2. Description of the Related Art
[0005] Imaging apparatuses such as electronic cameras that employ a
memory card having a solid-state memory device as a recording
medium are already known as imaging apparatus that image, record,
and play back still images or moving images. These electronic
cameras record and play back still images or moving images that are
imaged with a solid-state imaging device such as a CCD or a
CMOS.
[0006] When imaging using an imaging device, processing is carried
out using dark pixel signals that are read out after performing
charge accumulation in a similar manner to actual imaging in a
state in which the imaging device is not exposed and actual imaging
pixel signals that are read out after performing charge
accumulation in a state in which the imaging device is exposed. It
is thereby possible to perform dark noise correction processing. As
a result, a high quality image can be obtained by correcting imaged
pixel signals with respect to image degradation such as pixel
defects caused by unique minute flaws of the imaging device, fixed
pattern noise caused by a readout circuit, or dark current noise
generated in the imaging device.
[0007] In a correction method that uses dark pixel signals, since a
large-capacity memory for one frame is necessary for storing dark
pixel signals, the cost of the imaging apparatus is increased.
Also, since it is necessary to previously acquire dark pixel
signals for one frame prior to actual imaging, a shutter release
time lag increases and hinders optimum imaging. Further, since
random components that are referred to as random noise are included
in dark pixel signals, fixed pattern noise can not be completely
suppressed by only performing processing to subtract dark pixel
signals from actual imaging pixel signals.
[0008] To solve this problem, a method has been proposed that
divides a pixel region into an effective pixel region and an
ineffective pixel region, averages the pixel signals of the
ineffective pixel region, and performs processing to subtract the
obtained average value from the pixel signals of the effective
pixel region. The effective pixel region outputs signals that
accumulate charges generated according to incident light on the
pixel region of the imaging device. The ineffective pixel region is
a pixel region in which, for example, a pixel surface of the same
circuit configuration as the effective pixel region is covered with
a light-shielding film such as aluminum, and which outputs signals
that are not dependent on incident light. According to this method,
averaging parameters are increased by subjecting pixel signals of
the ineffective pixel region to averaging processing, and random
noise included in the correction signal is decreased.
[0009] An object of Japanese Patent Application Laid-Open No.
2002-016841 is to provide a solid-state imaging apparatus which
generates a correction signal based on pixel signals of an
ineffective pixel region having a plurality of rows to increase the
reliability of the correction signal and suppress fixed pattern
noise. To achieve the above object, a solid-state imaging apparatus
is proposed in which a correction signal for one row portion is
generated from light shielded pixels of a plurality of row portions
and pixel signals of effective pixels are corrected.
[0010] Japanese Patent Application Laid-Open No. 2004-015712
proposes a solid-state imaging apparatus in which pixel signals
obtained from pixel signals of an ineffective pixel region are
totaled across a plurality of frames, and an average value is
determined from that total value to generate a correction signal
and suppress fixed pattern noise.
[0011] Japanese Patent Application Laid-Open No. 2000-138864
proposes an imaging apparatus in which reading of each
photoelectric conversion element is performed a plurality of times
within a single-field period, and charges of elements for detecting
a vertical black reference are read per every reading that is
conducted a plurality of times within a single-field period.
[0012] However, there are the following problems with the above
described methods of suppressing fixed pattern noise according to
the prior art.
[0013] According to the solid-state imaging apparatus proposed in
Japanese Patent Application Laid-Open No. 2002-016841, there is the
problem that when an ineffective pixel region with a large number
of rows is provided to enhance the reliability of a correction
signal, the chip area increases. More specifically, the influence
of random noise included in a correction signal is reduced by
providing a large number of rows in the ineffective pixel region
used for averaging.
[0014] According to the solid-state imaging apparatus proposed in
Japanese Patent Application Laid-Open No. 2004-15712, there is the
problem that since pixel signals of an ineffective pixel region
extending across a plurality of frames are required to generate a
correction signal, the invention can not be used in an imaging mode
that captures only a single frame such as when a single image is
captured using a digital still camera. More specifically, the
influence of random noise included in a correction signal is
reduced by increasing the number of frames of pixel signals of an
ineffective pixel region used for averaging.
[0015] According to the solid-state imaging apparatus proposed in
Japanese Patent Application Laid-Open No. 2000-138864, the
influence of random noise included in a correction signal is
reduced by increasing the number of frames of pixel signals of an
ineffective pixel region used for averaging. More specifically,
there is the problem that since pixel signals of an ineffective
pixel region extending across a plurality of frames are required to
generate a correction signal, the invention can not be used in an
imaging mode that captures only a single frame such as when a
single image is captured using a digital still camera. Also, even
in a case in which pixel signals of an ineffective pixel region
that extends over a plurality of frames can be acquired as in a
case of imaging a moving image, when the ineffective pixel region
is small there is the problem that the response until the start of
imaging worsens since a large number of frames are required.
Further, there is the problem that when the number of frames to be
integrated is decreased by increasing the ineffective pixel region,
the chip area is increased.
[0016] To solve the above problems, an object of the present
invention is to provide an imaging apparatus that reduces the
influence of random noise included in a correction signal based on
pixel signals of an ineffective pixel region of only one frame
without increasing a chip area, as well as a processing method
thereof.
SUMMARY OF THE INVENTION
[0017] An imaging apparatus according to the present invention has
a pixel unit including an effective pixel region of a plurality of
pixels that accumulate charges generated according to incident
light and output a signal, and an ineffective pixel region of a
plurality of pixels that output a signal that does not depend on
incident light; a plurality of vertical signal lines provided for
each column of pixels of the pixel unit; a vertical scanning
circuit that, by scanning and selecting pixels of the pixel unit in
row units, outputs signals of pixels of the same row that is
selected to the plurality of vertical signal lines; and a
horizontal scanning circuit that, by scanning and selecting signals
of the plurality of vertical signal lines, outputs a signal of the
selected vertical signal line; wherein the vertical scanning
circuit selects pixels of a same row of the effective pixel region
in row units one time during one frame, and selects pixels of a
same row of the ineffective pixel region in row units a plurality
of times during one frame.
[0018] Other features and advantages of the present invention will
be apparent from the following description taken in conjunction
with the accompanying drawings, in which like reference characters
designate the same or similar parts throughout the figures
thereof.
BRIEF DESCRIPTION OF THE DRAWINGS
[0019] FIG. 1 is a plan view of a pixel unit of a solid-state
imaging apparatus relating to a first embodiment of the present
invention.
[0020] FIG. 2 is a circuit diagram of a pixel, a readout circuit,
and a common output circuit of a CMOS-type solid-state imaging
apparatus.
[0021] FIG. 3 is a block diagram illustrating a configuration
example of a CMOS-type solid-state imaging apparatus.
[0022] FIGS. 4A and 4B are timing charts of the solid-state imaging
apparatus relating to the first embodiment of the present
invention.
[0023] FIG. 5 is a circuit diagram of a vertical scanning circuit
of a CMOS-type solid-state imaging apparatus.
[0024] FIG. 6 is a timing chart of the solid-state imaging
apparatus relating to the first embodiment of the present
invention.
[0025] FIG. 7 is a conceptual diagram that describes signal levels
of pixel signals of the solid-state imaging apparatus relating to
the first embodiment of the present invention.
[0026] FIG. 8 is a block diagram that includes a pixel signal
correction circuit of the solid-state imaging apparatus relating to
the first embodiment of the present invention.
[0027] FIG. 9 is a timing chart of a solid-state imaging apparatus
relating to a second embodiment of the present invention.
[0028] FIG. 10 is a conceptual diagram that describes signal levels
of pixel signals of the solid-state imaging apparatus relating to
the second embodiment of the present invention.
[0029] FIG. 11 is a conceptual diagram illustrating the
correspondence between an average value calculation region and
average values of pixel signals of a solid-state imaging apparatus
relating to a second embodiment of the present invention.
[0030] FIG. 12 is a block diagram including a pixel signal
correction circuit of the solid-state imaging apparatus relating to
the second embodiment of the present invention.
[0031] FIG. 13 is a timing chart of a solid-state imaging apparatus
relating to a third embodiment of the present invention.
[0032] FIG. 14 is a circuit diagram of a pixel, a readout circuit,
and a common output circuit of a solid-state imaging apparatus
relating to a fourth embodiment of the present invention.
[0033] FIG. 15 is a timing chart of a solid-state imaging apparatus
relating to the fourth embodiment of the present invention.
[0034] FIG. 16 is a timing chart of a solid-state imaging apparatus
relating to a fifth embodiment of the present invention.
[0035] FIG. 17 is a conceptual diagram that describes signal levels
of pixel signals of the solid-state imaging apparatus relating to
the fifth embodiment of the present invention.
[0036] FIG. 18 includes a plan view of a solid-state imaging
apparatus relating to a sixth embodiment of the present invention
and a conceptual diagram that describes signal levels of read-out
pixel signals.
[0037] FIG. 19 is a timing chart of a solid-state imaging apparatus
relating to a sixth embodiment of the present invention.
[0038] FIG. 20 is a conceptual diagram that describes signal levels
of pixel signals of a solid-state imaging apparatus relating to a
seventh embodiment of the present invention.
[0039] FIG. 21 is a conceptual diagram that represents a method of
reading out pixel signals of the solid-state imaging apparatus
relating to the seventh embodiment of the present invention.
[0040] FIG. 22 is a conceptual diagram that illustrates the
relation between average value calculation regions and average
values of read-out signals in the solid-state imaging apparatus
relating to the seventh embodiment of the present invention.
[0041] FIG. 23 is a configuration diagram of an imaging system that
uses the solid-state imaging apparatus relating to the seventh
embodiment of the present invention.
DESCRIPTION OF THE EMBODIMENTS
[0042] The accompanying drawings, which are incorporated in and
constitute a part of the specification, illustrate embodiments of
the invention and, together with the description, serve to explain
the principles of the invention.
First Embodiment
[0043] Hereunder, the first embodiment of the present invention is
described in detail using the drawings.
[0044] FIG. 1 is a plan view of a pixel unit of a solid-state
imaging apparatus relating to a first embodiment of the present
invention. A solid-state imaging apparatus according to the present
embodiment has a pixel unit 100 that includes an effective pixel
region 101 having a plurality of pixels that accumulates charges
generated in response to incident light and output signals and an
ineffective pixel region 102 having a plurality of pixels that
output signals that are not dependent on incident light. The
ineffective pixel region 102 is disposed in a row unit at the top
in the vertical direction of the pixel unit 100. For example, a
semiconductor impurity region for accumulating a charge is formed
in the pixels of the effective pixel region 101, and a
semiconductor impurity region for accumulating a charge is not
formed in the pixels of the ineffective pixel region 102.
[0045] FIG. 2 is a circuit diagram of a pixel, a readout circuit,
and a common output circuit of a CMOS-type solid-state imaging
apparatus. A pixel 200 is configured as follows. A photodiode 201
converts received light into a charge. In this example, the anode
side of the photodiode 201 that generates a light signal charge is
grounded. The cathode side of the photodiode 201 is connected to a
gate of an amplification MOS transistor 203 through a transfer MOS
transistor 202. To the gate of the amplification MOS transistor 203
is connected a source of a reset MOS transistor 204 for resetting
the amplification MOS transistor 203. The drain of the reset MOS
transistor 204 is connected to a power supply voltage Vcc. The
drain of the amplification MOS transistor 203 is connected to the
power supply voltage Vcc, and the source is connected to the drain
of a selection MOS transistor 205.
[0046] FIG. 3 is a block diagram that illustrates a configuration
example of a CMOS-type solid-state imaging apparatus. A pixel unit
300 is composed of 3.times.3 pixels, in which three unit pixels 301
are disposed in a row direction, and three unit pixels 301 are
disposed in a column direction. The CMOS-type solid-state imaging
apparatus includes a vertical scanning circuit 302 that selects a
row to be read out, a horizontal scanning circuit 303 that selects
a column to be read out, and a readout circuit 304 that performs
amplification and noise removal with respect to pixel signals. The
unit pixel 301 corresponds to the pixel 200 shown in FIG. 2. An
operation to read out a pixel signal is described hereafter. A
plurality of vertical signal lines 206 shown in FIG. 2 are provided
for each column of pixels 301 of the pixel unit 300. The vertical
scanning circuit 302 outputs signals of the same row of pixels
selected by scanning and selecting pixels 301 of the pixel unit 300
in row units to the plurality of vertical signal lines 206. The
vertical scanning circuit may select one row by one row the pixels
301 of the pixel unit 300, and may select plural rows by plural
rows the pixels 301 of the pixel unit 300. The horizontal scanning
circuit 303 outputs a signal of the vertical signal line 206
selected by scanning and selecting signals of the plurality of
vertical signal lines 206.
[0047] The gate of the transfer MOS transistor 202 of a pixel 11 is
connected to a first row selection line (vertical scanning line)
Ptx1 that extends in the lateral direction. Gates of a similar
transfer MOS transistor 202 of a pixel 12 and a pixel 13 that are
disposed in the same row are also commonly connected to the above
described first row selection line Ptx1. The gate of a reset MOS
transistor 204 of the pixel 11 is connected to a second row
selection line (vertical scanning line) Pres1 that is extends in
the lateral direction. Gates of a similar reset MOS transistor 204
of the pixel 12 and the pixel 13 that are disposed in the same row
are also commonly connected to the second row selection line Pres1.
The gate of the selection MOS transistor 205 of the pixel 11 is
connected to a third row selection line (vertical scanning line)
Psel1 that extends in the lateral direction. Gates of a similar
selection MOS transistor 205 of the pixel 12 and the pixel 13 that
are disposed in the same row are also commonly connected to the
third row selection line Psel1. The first to third row selection
lines Ptx1, Pres1, and Psel1 are connected to the vertical scanning
circuit 302 and a signal voltage is supplied thereto. Pixels of the
same configuration and row selection lines are also provided in the
remaining rows illustrated in FIG. 3. Row selection signals Ptx1,
Pres1, and Psel1; Ptx2, Pres2, and Psel2; and Ptx3, Pres3, and
Psel3 that are formed by the vertical scanning circuit 302 are
supplied to these row selection lines. The sources of the selection
MOS transistors 205 are connected to a vertical signal line 206
that extends in the longitudinal direction. Sources of similar
selection MOS transistors 205 of pixels disposed in the same column
are also connected to the vertical signal line 206. The section
enclosed by a dashed line in FIG. 2 is a circuit example for one
column block of the readout circuit 304 shown in FIG. 3. A pixel
output Vout is connected to the vertical signal line 206. A
constant current source 207 is connected to the vertical signal
line 206. A feedback capacitor 210 is connected between the input
and output of an amplifier 211.
[0048] FIG. 4A is a timing chart that illustrates a processing
method of the CMOS-type solid-state imaging apparatus illustrated
in FIG. 1 to FIG. 3. Prior to reading out a light signal charge
from the photodiode 201, the gate Pres1 of the reset MOS transistor
204 changes to high level. As a result, the gate of the
amplification MOS transistor 203 is reset to a reset power supply
voltage. After a gate Pc0r of a clamp switch 209 changes to high
level at the same time as the gate Pres1 of the reset MOS
transistor 204 returns to low level, the gate Psel1 of the
selection MOS transistor 205 changes to high level. Thereby, a
reset signal (noise signal) in which reset noise is superposed is
read out to the vertical signal line 206 and clamped by a clamp
capacitor 208 (C0) of each column. Next, after a gate Pc0r of a
clamp switch 209 returns to low level, a gate Pctn of an N-side
transfer switch 213 changes to high level and a reset signal is
held by a noise holding capacitor 215 (Ctn) provided in each
column. Subsequently, a gate Pcts of an S-side transfer switch 212
changes to high level. Next, a gate Ptx1 of the transfer MOS
transistor 202 changes to high level, and simultaneously to
transfer of a light signal charge of the photodiode 201 to the gate
of the amplification MOS transistor 203, a light signal is read out
to the vertical signal line 206. Subsequently, after a gate Ptx1 of
the transfer MOS transistor 202 returns to low level, a gate Pcts
of the S-side transfer switch 212 changes to low level. Thereby,
the amount of change (light signal) from the reset signal is read
out to a signal holding capacitor 214 (Cts) provided in each
column. By the operations up to this point, pixel signals connected
to the first row are held by signal holding capacitors 214 and 215
(Ctn and Cts) that are connected to the respective columns. After
pixel signals are held by the signal holding capacitors 214 and
215, the gate Pres1 of the reset MOS transistor 204 and the gate
Ptx1 of the transfer MOS transistor 202 change to high level to
thereby reset the photodiode 201 to the reset power supply
voltage.
[0049] After pixel signals are held by the signal holding
capacitors 214 and 215, gates of horizontal transfer switches 216
and 217 of each column are sequentially changed to high level by a
signal Ph (Ph1) supplied from the horizontal scanning circuit 303.
Voltages that are held at the signal holding capacitors 214 and 215
(Cts and Ctn) are read out sequentially to the horizontal output
line capacitors 218 and 219 (Chs and Chn), subjected to
differential processing at an output amplifier 222 and sequentially
output to an output terminal OUT. More specifically, the output
amplifier 222 subtracts the signal of the horizontal output line
capacitor 219 from the signal of the horizontal output line
capacitor 218 to remove a noise component. During an interval
between reading out signals of each column, the horizontal output
line capacitors 218 and 219 (Chs and Chn) are reset to reset
voltages Vchrs and Vchrn by reset switches 220 and 221. Thus, read
out of pixels connected in the first row is completed.
[0050] FIG. 5 is a view that illustrates one example of the
vertical scanning circuit 302 shown in FIG. 3. The vertical
scanning circuit 302 comprises a shift register circuit that
sequentially transfers a vertical scanning start signal to
subsequent flip-flops 500 in synchrony with the rising edge of a
vertical scanning signal PV shown in FIG. 4A. Row selection signals
Ptx, Pres, and Pse1 and output signals Pv1 to Pv5 of flip-flops 500
are connected to inputs of AND gates 501 provided in each column.
Thus, row selection signals that are independent for each row are
generated by the vertical scanning signal PV.
[0051] FIG. 6 is a timing chart that illustrates operations of the
solid-state imaging apparatus relating to the present embodiment.
According to the present embodiment, the ineffective pixel region
102 shown in FIG. 1 is disposed in one row at the top in the
vertical direction of the pixel unit 100. In the present
embodiment, in order to output pixel signals of the ineffective
pixel region consisting of one row disposed at the top of the pixel
unit over four horizontal scanning time periods, the row selection
signal Pv1 that selects the first row is output over four
horizontal scanning time periods. In order to generate the row
selection signal Pv1, as shown in FIG. 4B, the vertical scanning
signal PV is set to low level during the 2 to 4 horizontal scanning
time periods and shift operations of the shift register circuit
shown in FIG. 5 are stopped. As a result, pixel signals of the
first row are repeatedly output over four horizontal scanning time
periods.
[0052] In the present specification, the term "one horizontal
scanning time period" refers to a length of time in which pixel
signals of one row are read out that is obtained by adding a time
period in which a light signal is converted to a voltage and
transferred to an internal capacitor (horizontal blanking time
period) and a time period in which pixel signals are scanned are
read out (horizontal transfer period). In FIG. 4, a time period in
which horizontal selection signals ph1, ph2 . . . are input is the
"horizontal transfer period", and the time period before that is
the "horizontal blanking time period". One horizontal scanning time
period is the interval of a horizontal synchronizing signal.
[0053] By thereafter repeating the operations of the timing chart
shown in FIG. 4A once more, pixel signals connected to the second
and subsequent rows are sequentially read out by means of a signal
from the vertical scanning circuit 302 to complete read out of all
pixels. For the second and subsequent rows, pixel signals of the
effective pixel region 101 of each row are output per one
horizontal scanning time period (one time).
[0054] A period from the start of an operation to read out pixel
signals of the first row comprising the pixel unit 100 until
completing read out of pixel signals of the last row in this manner
is defined as one frame period.
[0055] FIG. 7 is a conceptual diagram that describes signal levels
of read-out pixel signals of the solid-state imaging apparatus
relating to the present embodiment. According to the present
embodiment, pixels of an ineffective pixel region are composed of
null pixels that do not form a semiconductor impurity region for
accumulating a charge. The ineffective pixel region consists of one
row. Since the pixel signals 702 of the ineffective pixel region
700 are null pixel signals, they constitute a signal level that
serves as a standard for the black level of a captured image. By
reading out the pixel signals of the ineffective pixel region 700
that has only one row at the timings shown in FIG. 6, the number of
pixel signals can be increased to that of four rows and read out.
After reading out the null pixel signals 702 in an amount
equivalent to four rows, output of a row selection signal starts
once more, and effective pixel signals 703 of the effective pixel
region 701 that outputs signals which accumulate charges that are
generated according to incident light are output for each row.
[0056] FIG. 8 is a block diagram that includes a pixel signal
correction circuit of a solid-state imaging apparatus relating to
the present embodiment. The configuration illustrated in FIG. 8
includes a pixel unit 800, an ineffective pixel region 801, an
effective pixel region 802, a vertical scanning circuit 803, a
readout circuit 804, and a horizontal scanning circuit 805. Analog
pixel signals output from the readout circuit 804 by the above
described operations are digitized by an AD converter 806. An
averaging circuit 807 calculates an average value for each column
with respect to pixel signals of the ineffective pixel region 801
that are increased to an amount equivalent to four rows. The
calculated average value for each column is stored in a memory 808.
By subtracting the stored average value for each column from the
pixel signals of the effective pixel region 802 using a subtracter
809, fixed pattern noise occurring in each column can be accurately
suppressed.
[0057] The vertical scanning circuit 302 selects pixels of the same
row of the pixel unit 800 in row units a plurality of times during
one frame. More specifically, the vertical scanning circuit 802
selects pixels of the same row of the effective pixel region 802 in
row units once during one frame and selects pixels of the same row
of the ineffective pixel region 801 in row units a plurality of
times during one frame. For example, the vertical scanning circuit
802 selects pixels of the first row four times. The correction
circuit has the averaging circuit 807, the memory 808, and the
subtracter 809 shown in FIG. 8. The correction circuit corrects
pixel signals of the effective pixel region 802 using the pixel
signals that are selected a plurality of times and output by the
horizontal scanning circuit 805. More specifically, the correction
circuit averages the pixel signals that are selected a plurality of
times and corrects the pixel signals of the effective pixel region
802.
[0058] According to the present embodiment, pixels of the
ineffective pixel region are composed of null pixels that do not
form a semiconductor impurity region for accumulating a charge.
However, in a case in which a period for accumulating a charge is
short or a dark current of a photodiode is itself small, the effect
of the present embodiment does not change even when a configuration
is adopted in which pixels of the ineffective pixel region are
composed of light shielded pixels in which the pixel surface of a
circuit configuration that is the same as the effective pixel
region is covered with a light-shielding film such as aluminum.
[0059] According to the present embodiment, although line memories
of a number equivalent to the number of horizontal pixels are
required to store average values for each column in memory, the
effect of the present embodiment does not change even when a
configuration is adopted in which the horizontal pixels are
separated into several regions to reduce the number of memories
storing.
[0060] According to the present embodiment, although the number of
pixels of the ineffective pixel region that is averaged is one row,
the effect of the present embodiment does not change even when the
number of rows is increased to a degree that does not exert
space-related pressure on the chip area. By adopting a
configuration in which the rows of an ineffective pixel region
having a plurality of rows are each read out a plurality of times
to increase the read-out averaging parameters, an average value of
an even higher accuracy can be calculated.
Second Embodiment
[0061] Although the second embodiment of the present invention
resembles the first embodiment, a timing chart illustrating the
operations of the solid-state imaging apparatus and a method of
suppressing fixed pattern noise using pixel signals of an
ineffective pixel region are different from the first embodiment.
Hereunder, the present embodiment is described in detail using the
drawings.
[0062] FIG. 9 is a timing chart that illustrates an example of
operations of the solid-state imaging apparatus relating to the
second embodiment of the present invention. According to the
present embodiment, the ineffective pixel region 102 shown in FIG.
1 is provided as two rows at the top in the vertical direction of
the pixel unit 100. In order to output pixel signals of the second
row of the ineffective pixel region having two rows provided at the
top of the pixel unit in an amount equivalent to five rows, a row
selection signal Pv2 that selects the second row is a signal that
outputs the amount of five rows. To generate the row selection
signal Pv2, as shown in FIG. 4B, the vertical scanning signal PV is
set at low level during 3 to 6 horizontal scanning time periods and
shift operations of the shift register circuit shown in FIG. 5 are
stopped. Thus, pixel signals of the second row are repeatedly
output over five rows.
[0063] Thereafter, by repeating the operations of the timing chart
shown in FIG. 4A once more, pixel signals connected to the third
and subsequent rows are sequentially read out by means of a signal
from the vertical scanning circuit to complete read out of all
pixels. Rows other than the second row are output per one
horizontal scanning time period (one time).
[0064] A period from the start of an operation to read out pixel
signals of the first row that comprises the pixel unit until
completing read out of pixel signals of the last row in this manner
is defined as one frame period.
[0065] FIG. 10 is a conceptual diagram that describes signal levels
of read-out pixel signals of the solid-state imaging apparatus
relating to the present embodiment. According to the present
embodiment, pixels of an ineffective pixel region are composed of
light shielded pixels in which the pixel surface of a circuit
configuration that is the same as the effective pixel region is
covered with a light-shielding film such as aluminum. The
ineffective pixel region consists of two rows. The first row is a
first ineffective pixel region 1000, and the second row is a second
ineffective pixel region 1001. Since pixel signals 1003 of the
first ineffective pixel region 1000 are light-shielded pixel
signal, a pixel signal level that is influenced by a dark current
that depends on a temperature or an accumulation time is output.
With respect to pixel signals 1004 of the second ineffective pixel
region 1001 that is subsequently read out, a pixel signal level
that is influenced by a dark current similarly to the pixel signal
1003 of the first ineffective pixel region 1000 is output. By
reading out pixel signals according to the timings illustrated in
FIG. 9, pixel signals of the second ineffective pixel region 1001
having only one row can be increased to a number equivalent to five
rows and read out. With respect to pixel signals read out the
second and subsequent times, since the photodiode and signal
holding capacitor are reset at an immediately preceding horizontal
scanning time period as shown in FIG. 4B, the pixel signals are
pixel signals 1005 that are influenced by a dark current for which
the accumulation time is one horizontal scanning time period. The
length of a general single horizontal scanning time period is
approximately several tens of .mu. seconds, and the influence of
dark current for an accumulation period of this level can be mostly
ignored. After reading out the pixel signals of these ineffective
pixel regions, output of row selection signals is started once more
to output effective pixel signals 1006 of the effective pixel
region 1002 that outputs signals that accumulate charges generated
according to incident light. Using the timings shown in FIG. 9,
with the ineffective pixel regions 1000 and 1001 that consist of
only two rows it is possible to acquire data 1003 and 1004 for two
rows of pixel signals that are influenced by dark current, and data
1005 for four rows of pixel signals that receive almost no dark
current influence during one frame.
[0066] FIG. 11 is a conceptual diagram illustrating the
correspondence between an average value calculation region and
average values of pixel signals read out by a solid-state imaging
apparatus relating to the present embodiment. The average value
calculation region illustrated in FIG. 11 includes a first
ineffective pixel region 1100, a second ineffective pixel region
1101, and an effective pixel region 1102.
[0067] FIG. 12 is a block diagram that includes a pixel signal
correction circuit of a solid-state imaging apparatus relating to
the present embodiment. The configuration illustrated in FIG. 12
includes a pixel unit 1200, a first ineffective pixel region 1201,
a second ineffective pixel region 1202, an effective pixel region
1203, a vertical scanning circuit 1204, a readout circuit 1205, and
a horizontal scanning circuit 1206. Analog pixel signals output
from the readout circuit 1205 by the aforementioned operations are
digitized by an AD convertor 1207.
[0068] An average value of pixel signals of the first ineffective
pixel region 1201 (pixel signals of first row=pixel signals of one
row) and pixel signals of the second ineffective pixel region 1202
(pixel signals obtained by first read out of second row=pixel
signals of second row) is calculated by a first averaging circuit
1208. As shown in FIG. 11, the first averaging circuit 1208
calculates an average value for each region obtained by
horizontally dividing the first and second rows into three regions
and stores the calculated values in a first memory 1209. Unlike a
method that performs averaging for each column, since a large
number of averaging parameters are obtained by calculating average
values for respective regions consisting of a plurality of pixels,
highly accurate average values can be obtained based on even the
pixel signals of two rows. Next, the average value of pixel signals
of the second ineffective pixel region 1202 (pixel signals for
second read to pixel signals for fifth read out of second row=pixel
signals of third row to sixth row) is calculated by the second
averaging circuit 1210. As shown in FIG. 11, the second averaging
circuit 1210 calculates an average value for each column and stores
the calculated values in a second memory 1211. Since averaging is
performed for each column, pixel signals for four rows are used
according to the present embodiment to increase the averaging
parameters. The two average values calculated in this manner are
subtracted from the pixel signals of the effective pixel region
1203 by respective subtracting circuits 1212 and 1213. As
illustrated in FIG. 12, for the purpose of cancelling out the
effect of subtracting two average values, an adding circuit 1215
may add an offset 1214 for black level adjustment to the output
signal of the subtracting circuit 1213. According to the above
described method, by using the present embodiment it is possible to
accurately suppress influences due to dark current of a solid-state
imaging apparatus and fixed pattern noise occurring in each
column.
[0069] The ineffective pixel region includes the first ineffective
pixel region 1201 and the second ineffective pixel region 1202 that
have mutually different rows. The vertical scanning circuit 1204
selects pixels of the same row of the second ineffective pixel
region 1202 in row units a plurality of times during one frame. The
correction circuit has a first correction circuit and a second
correction circuit. More specifically, the correction circuit has
averaging circuits 1208 and 1210, memories 1209 and 1211,
subtracters 1212 and 1213, and an adder 1215. The first correction
circuit has the averaging circuit 1208, the memory 1209, and the
subtracter 1212. The first correction circuit averages the pixel
signals of the first ineffective pixel region 1201 and the second
ineffective pixel region 1202 that are selected the first time and
corrects the pixel signals of the effective pixel region 1203. The
second correction circuit has the averaging circuit 1210, the
memory 1211, and the subtracter 1213. The second correction circuit
averages the pixel signals of the second ineffective pixel region
1202 that are selected the second and subsequent times and corrects
the pixel signals of the effective pixel region 1203. The first
correction circuit averages pixel signals of the first ineffective
pixel region 1201 and the second ineffective pixel region 1202 for
each of a plurality of regions (for example, three regions) into
which the ineffective pixel regions are horizontally divided. The
second correction circuit averages pixel signals of the second
ineffective pixel region 1202 for each of a plurality of regions
into which the second ineffective pixel region 1202 is horizontally
divided by a dividing number that is larger than the number of
divisions of the first correction circuit.
[0070] According to the present embodiment, although the number of
pixels of the ineffective pixel region that is averaged is two
rows, the effect of the present embodiment does not change even
when the number of rows is increased to a degree that does not
exert space-related pressure on the chip area. By adopting a
configuration in which the rows of an ineffective pixel region
having a plurality of rows are each read out a plurality of times
to increase the read-out averaging parameters, an average value of
an even higher accuracy can be calculated.
[0071] According to the present embodiment, the ineffective pixel
region is composed of light shielded pixels in which the pixel
surface of a circuit configuration that is the same as the
effective pixel region is covered with a light-shielding film such
as aluminum. However, with respect to the pixels of the second
ineffective pixel region, even when an average value is calculated
by taking pixel signals that receive the influence of dark current
as null pixels that do not form a semiconductor impurity region for
accumulating a charge as one row, and taking pixel signals that
receive almost no influence from dark current as five rows, the
effect of the present embodiment does not change.
Third Embodiment
[0072] Although the third embodiment of the present invention
resembles the first embodiment, a timing chart illustrating the
operations of the solid-state imaging apparatus with respect to
ineffective pixel regions that are repeatedly read out is different
from the first embodiment. Hereunder, the present embodiment is
described in detail using the drawings.
[0073] FIG. 9 is a timing chart that illustrates the operations of
the solid-state imaging apparatus relating to the third embodiment
of the present invention. According to the present embodiment, the
ineffective pixel region 102 shown in FIG. 1 is provided as two
rows at the top in the vertical direction of the pixel unit 100. In
order to output the pixel signals of the second row of the
ineffective pixel region having two rows disposed at the top of the
pixel unit over five horizontal scanning time periods, a row
selection signal Pv2 that selects the second row is a signal that
outputs five horizontal scanning time periods. To generate this row
selection signal Pv2, as shown in FIG. 13, the vertical scanning
signal PV is set at low level during 3 to 6 horizontal scanning
time periods and shift operations of the shift register circuit
shown in FIG. 5 are stopped. Thus, pixel signals of the second row
are repeatedly output over five horizontal scanning time periods.
The operations for the second row that is repeatedly read out
differ only in the respect that a light signal charge is not
transferred to the amplification MOS transistor 203 by the
selection signal Ptx1. The pixel signals that are read out at the
timings shown in FIG. 13 are black level output that do not depend
on dark current generated at the photodiode.
[0074] Thereafter, by repeating the operations of the timing chart
shown in FIG. 4A once more, pixel signals connected to the third
and subsequent rows are sequentially read out by means of a signal
from the vertical scanning circuit to complete read out of all
pixels.
[0075] A period from the start of an operation to read out pixel
signals of the first row that comprises the pixel unit until
completing read out of pixel signals of the last row in this manner
is defined as one frame period.
[0076] Since a method that suppresses fixed pattern noise of the
effective pixel region using pixel signals of the ineffective pixel
region is the same as the method described in the first embodiment,
a description thereof is omitted here.
[0077] According to the present embodiment, pixels of an
ineffective pixel region may include null pixels that do not form a
semiconductor impurity region for accumulating a charge, light
shielded pixels in which the pixel surface of a circuit
configuration that is the same as the effective pixel region is
covered with a light-shielding film such as aluminum, and pixels of
a circuit configuration that is the same as the effective pixel
region.
Fourth Embodiment
[0078] Although the fourth embodiment of the present invention
resembles the first embodiment, a timing chart illustrating the
operations of the solid-state imaging apparatus with respect to
ineffective pixel regions that are repeatedly read out is different
from the first embodiment. Hereunder, the present embodiment is
described in detail using the drawings.
[0079] FIG. 14 is a circuit diagram of a pixel, a readout circuit,
and a common output circuit of a solid-state imaging apparatus
relating to the fourth embodiment of the present invention. The
difference between the present embodiment and the first to third
embodiments is that a MOS transistor switch 1423 is provided for
resetting the voltage level of a vertical signal line 1406 to a
reference voltage Voutr. A MOS transistor switch 1423 is provided
for each vertical signal line 1406, and is a switch for connecting
the vertical signal line 1406 to the reference voltage Voutr. FIG.
15 illustrates the operation timing for a row of an ineffective
pixel region that is repeatedly read out.
[0080] Since it is not necessary to read out a charge that is
generated at a photodiode 1401, a gate Pres1 of a reset MOS
transistor 1404 is maintained at high level. As a result, the gate
of the amplification MOS transistor 1403 is continually reset to
the reset power supply voltage. The vertical signal line 1406 is
reset to the reset voltage Voutr by setting a gate Pvoutr1 of the
MOS transistor switch 1423 to high level. At this time, after a
gate Pc0r of a clamp switch 1409 changes to high level, a gate
Psel1 of a selection MOS transistor 1405 changes to high level such
that the reset signal Voutr is clamped by a clamp capacitor 1408
(C0) of each column. Next, after the gate Pc0r of the clamp switch
1409 returns to low level, a gate Pctn of the N-side transfer
switch 1413 changes to high level and reset signals are held in
noise holding capacitors 1415 (Ctn) provided in each column.
Subsequently, the gate Pctn of the N-side transfer switch 1413
changes to low level and the gate Pcts of the S-side transfer
switch 1412 changes to high level. Next, the gate Pcts of the
S-side transfer switch 1412 changes to low level. As a result, a
reset signal is once more read out to the signal holding capacitor
1414 (Cts) provided in each column. By the operations up to this
point, a reset signal connected to a first row is held in signal
holding capacitors 1414 and 1415 (Ctn and Cts) connected to the
respective columns. A constant current source 1407 is connected to
the vertical signal line 1406. A feedback capacitor 1410 is
connected between the input and output of an amplifier 1411.
[0081] After pixel signals are held at the signal holding
capacitors 1414 and 1415, gates of horizontal transfer switches
1416 and 1417 of each column are sequentially changed to a high
level by a signal Ph1 supplied from the horizontal scanning
circuit. Voltages being held at the signal holding capacitors 1414
and 1415 (Cts and Ctn) are sequentially read out to horizontal
output line capacitors 1418 and 1419 (Chs and Chn), subjected to
differential processing at an output amplifier 1422, and output in
sequence to an output terminal OUT. During an interval between read
out of signals of each column the horizontal output line capacitors
1418 and 1419 (Chs and Chn) are reset to reset voltages Vchrs and
Vchrn by reset switches 1420 and 1421. Thus, read out of pixels
connected in the first row is completed.
[0082] According to the present embodiment, since the same charge
in accordance with reset voltage Voutr of the vertical signal line
1406 is held in the two signal holding capacitors 1414 and 1415, a
black level signal that undergoes differential processing at the
output amplifier 1422 is output to the output terminal OUT.
[0083] Thereafter, by repeating the operations of the timing chart
shown in FIG. 4A once more, pixel signals connected to the third
and subsequent rows are sequentially read out by means of a signal
from the vertical scanning circuit to complete read out of all
pixels.
[0084] A period from the start of an operation to read out pixel
signals of the first row that comprises the pixel unit until
completing read out of pixel signals of the last row in this manner
is defined as one frame period.
[0085] Since a method that suppresses fixed pattern noise of the
effective pixel region using pixel signals of the ineffective pixel
region is the same as the method described in the first embodiment,
a description thereof is omitted here.
Fifth Embodiment
[0086] Although the fifth embodiment of the present invention
resembles the first embodiment, a timing chart illustrating the
operations of the solid-state imaging apparatus and a method of
suppressing fixed pattern noise using pixel signals of an
ineffective pixel region are different from the first embodiment.
Hereunder, the present embodiment is described in detail using the
drawings.
[0087] FIG. 16 is a timing chart that illustrates operations of the
solid-state imaging apparatus relating to the fifth embodiment of
the present invention. According to the present embodiment, the
ineffective pixel region 102 shown in FIG. 1 is provided as two
rows at the top in the vertical direction of the pixel unit 100. In
order to output pixel signals of the ineffective pixel region
having two rows at the top of the pixel unit two rows at a time, a
row selection signal Pv1 that selects the first row and a row
selection signal Pv2 that selects the second row are signals that
output the amount of two rows. According to the present embodiment,
by inputting a vertical scanning start signal PVST as a start
signal of the vertical scanning circuit shown in FIG. 5 once more
after output of a row selection signal for the second row, the row
selection signal is returned to the first row.
[0088] FIG. 17 is a conceptual diagram that describes a signal
level of read-out pixel signals of the solid-state imaging
apparatus relating to the present embodiment. According to the
present embodiment, pixels of an ineffective pixel region 1700 are
composed of null pixels that do not form a semiconductor impurity
region for accumulating a charge. The ineffective pixel region 1700
consists of two rows. Since pixel signals 1702 of the ineffective
pixel region 1700 are null pixel signals, they constitute a signal
level that serves as a standard for the black level of a captured
image. By reading out the pixel signals of the ineffective pixel
region that originally has only two rows at the timings shown in
FIG. 16, the number of pixel signals can be increased to that of
four rows and read out. After reading out null pixel signals 1702
of an amount equivalent to four rows, output of row selection
signals starts once more to thereby output effective pixel signals
1703 of the effective pixel region 1701 that outputs signals in
which charges generated according to incident light are
accumulated.
[0089] Thereafter, by repeating the operations of the timing chart
shown in FIG. 4A once more, pixel signals connected to the second
and subsequent rows are sequentially read out by means of a signal
from the vertical scanning circuit to complete read out of all
pixels.
[0090] A period from the start of an operation to read out pixel
signals of the first row that comprises the pixel unit until
completing read out of pixel signals of the last row in this manner
is defined as one frame period.
[0091] Since a method that suppresses fixed pattern noise of the
effective pixel region 1701 using pixel signals of the ineffective
pixel region 1700 is the same as the method described in the first
embodiment, a description thereof is omitted here.
Sixth Embodiment
[0092] Hereunder, a sixth embodiment of the present invention is
described in detail using the drawings.
[0093] FIG. 18 includes a plan view of a pixel unit of a
solid-state imaging apparatus relating to the sixth embodiment of
the present invention and a conceptual diagram that describes
signal levels of read out pixel signals. The solid-state imaging
apparatus of the present embodiment has a pixel unit 1800 that
includes an effective pixel region 1801 having a plurality of
pixels that accumulates charges generated according to incident
light to output a signal, and an ineffective pixel region 1802
having a plurality of pixels that outputs a signal that is not
dependent on incident light. According to the present embodiment,
pixel signals of the ineffective pixel region 1802 disposed in one
column at the start in the horizontal direction of the pixel unit
1800 are repeatedly read out four times in one horizontal scanning
time period.
[0094] Since the block diagram and the circuit configuration of the
solid-state imaging apparatus relating to the present embodiment
are the same as in the first embodiment, a detailed description is
omitted here.
[0095] FIG. 19 is a timing chart for one horizontal scanning time
period of the solid-state imaging apparatus relating to the present
embodiment. By repeatedly selecting a total of four times a
horizontal selection signal Ph1 that selects the first column and
sequentially selecting the second and subsequent columns with
horizontal selection signals Ph2 to Ph5, it is possible to read out
only the pixel signals of the first column four times. At this
time, by fixing a reset signal Pchres of the signal holding
capacitor to low level while selecting the first column, it is
possible to execute read out a plurality of times while holding
pixel signals of the first column in the signal holding
capacitor.
[0096] In FIG. 8, a block diagram is shown that includes a pixel
signal correction circuit of a solid-state imaging apparatus
relating to the present embodiment, with the exception of the pixel
unit 800. In FIG. 18, a pixel unit 1800 is provided instead of the
pixel unit 800. Analog pixel signals output from the readout
circuit 804 by the aforementioned operations are digitized by an AD
convertor 806. The averaging circuit 807 calculates a row average
value for pixel signals of ineffective pixels that are increased to
an amount equivalent to four columns. The calculated row average
value is stored in the memory 808. The stored row average value is
subtracted from pixel signals of the effective pixel region 802 at
the subtracter 809. By performing the above described row average
value calculating operation and subtraction processing in a similar
manner in sequential row units, fixed pattern noise having a
horizontal stripe shape that occurs in each column can be
accurately suppressed.
[0097] As described above, according to the present embodiment the
horizontal scanning circuit 805 selects the same pixels of the
pixel unit 1800 a plurality of times during one horizontal scanning
time period. More specifically, the horizontal scanning circuit 805
selects the same pixels of the ineffective pixel region 1802 a
plurality of times during one horizontal scanning time period. For
example, the horizontal scanning circuit 805 selects the same
pixels four times. A semiconductor impurity region for accumulating
a charge is formed in pixels of the effective pixel region 1801. In
contrast, a semiconductor impurity region for accumulating a charge
is not formed in pixels of the ineffective pixel region 1802. The
correction circuit has the averaging circuit 807, the memory 808,
and the subtracter 809. The correction circuit corrects pixel
signals of the effective pixel region 1801 using the pixel signals
that are selected a plurality of times and output by the horizontal
scanning circuit 805. More specifically, the correction circuit
averages the pixel signals that are selected a plurality of times
and corrects the pixel signals of the effective pixel region
1801.
[0098] According to the present embodiment, pixels of the
ineffective pixel region may include light shielded pixels in which
the pixel surface of a circuit configuration that is the same as
the effective pixel region is covered with a light-shielding film
such as aluminum or may include null pixel that do not form a
semiconductor impurity region for accumulating a charge.
[0099] According to the present embodiment, although the number of
pixels of the ineffective pixel region that is averaged is one
column, the effect of the present embodiment does not change even
when the number of columns is increased to a degree that does not
exert space-related pressure on the chip area. By adopting a
configuration in which the columns of an ineffective pixel region
having a plurality of columns are each read out a plurality of
times to increase the averaging parameters that are read out, an
average value of an even higher accuracy can be calculated.
Seventh Embodiment
[0100] FIG. 20 includes a plan view of a pixel unit of a
solid-state imaging apparatus relating to the seventh embodiment of
the present invention and a conceptual diagram that describes
signal levels of read out pixel signals. The solid-state imaging
apparatus of the present embodiment includes three pixel regions:
an effective pixel region 2002 that accumulates charges generated
according to incident light to output signals, a first ineffective
pixel region 2000, and a second ineffective pixel region 2001. The
first ineffective pixel region 2000 is composed of light shielded
pixels in which the pixel surface of a circuit configuration that
is the same as the effective pixel region 2002 is covered with a
light-shielding film such as aluminum. The second ineffective pixel
region 2001 is composed of null pixels that do not form a
semiconductor impurity region for accumulating a charge.
[0101] Since the pixel signals of the first ineffective pixel
region 2000 having two rows are light-shielded pixel signals, a
pixel signal level 2003 that receives the influence of a dark
current that depends on temperature or accumulation time is output.
Since the pixel signals of the second ineffective pixel region 2001
having two rows are null pixel signals, a signal level 2005 that
serves as a standard for the black level of a captured image is
output. FIG. 20 also illustrates pixel signal levels 2004 of the
effective pixel region 2002.
[0102] FIG. 21 is a conceptual diagram that represents a driving
method of the solid-state imaging apparatus relating to the present
embodiment, that performs a so-called focal-plane shutter
operation. The time of one frame is, for example, 1/60 seconds.
Although in FIG. 21 an accumulation time for the first ineffective
pixel region 2000 and the effective pixel region 2002 is one frame
( 1/60 seconds), the effect of the present embodiment does not
change even when an accumulation time of equal to or less than one
frame is realized by separately performing a read-out scan and a
reset scan. According to this embodiment, pixel signals are read
out in the order of first ineffective pixel region
2000.fwdarw.effective pixel region 2002.fwdarw.second ineffective
pixel region 2001. In a case in which the period of one frame is
fixed, such as when imaging a moving image, a large number of null
pixel signals can be acquired by repeatedly reading out the second
ineffective pixel region 2001 a plurality of times in surplus time
after reading the effective pixel region 2002. According to FIG.
21, for example, null pixel signals equivalent to the amount for 12
rows can be acquired by repeatedly scanning the second ineffective
pixel region 2001 that has only two rows six times during one
frame.
[0103] According to the present embodiment, since the second
ineffective pixel region 2001 is repeatedly read out a plurality of
times in surplus time after reading out the effective pixel region
2002, it is possible to increase the null pixel signal parameters
to be averaged while maintaining the frame rate to thereby acquire
highly accurate correction data.
[0104] Further, the effect of the present embodiment does not
change even when null pixel signal parameters to be averaged are
further increased to acquire highly accurate correction data by
averaging acquired null pixel signals over a plurality of
frames.
[0105] FIG. 22 is a conceptual diagram that shows the correlation
between an average value calculation region and an average value of
pixel signals that are read out by the solid-state imaging
apparatus relating to the present embodiment. To simplify the
description, the first ineffective pixel region 2200 is set as two
rows, the effective pixel region 2202 is 98 rows, and the second
ineffective pixel region 2201 is two rows. For the first
ineffective pixel region 2200, an average value is calculated for
each of three regions obtained by horizontally dividing the rows
into three. For the second ineffective pixel region 2201, an
average value is calculated for each column.
[0106] FIG. 23 is a block diagram that includes a pixel signal
correction circuit of the solid-state imaging apparatus relating to
the present embodiment. Similarly to FIG. 12, the configuration
illustrated in FIG. 23 includes a pixel unit 2300, a first
ineffective pixel region 2301, a second ineffective pixel region
2302, an effective pixel region 2303, a vertical scanning circuit
2304, a readout circuit 2305, and a horizontal scanning circuit
2306.
[0107] As described above, after selecting the last row of the
second ineffective pixel region 2302, the vertical scanning circuit
2304 selects the first row of the second ineffective pixel region
2302 once again to sequentially scan the rows of the second
ineffective pixel region 2302. As a result, the vertical scanning
circuit 2304 repeatedly scans the second ineffective pixel region
2302. As illustrated in FIG. 21, the vertical scanning circuit 2304
repeatedly scans the second ineffective pixel region 2302 during
the period from completing scanning of the effective pixel region
2303 of the Nth frame until the start of scanning of the effective
pixel region 2303 of the N+1 frame.
[0108] Pixel signals output from the readout circuit 2305 by the
aforementioned operations are digitized by an AD convertor 2307.
The digitized pixel signals are stored in a frame memory 2316 and
at the appropriate time are read out by a first averaging circuit
2308, a second averaging circuit 2310, and a subtracting circuit
2312.
[0109] Average values of pixel signals of the first ineffective
pixel region 2301 (pixel signals of first row to second row=pixel
signals written in first row to second row of frame memory) that
are read out from the frame memory 2316 are calculated by the first
averaging circuit 2308. At the first averaging circuit 2308, as
shown in FIG. 22, an average value is calculated for each of three
regions obtained by horizontally dividing the first ineffective
pixel region 2301 in three, and the calculated average values are
stored in a first memory 2309. Unlike a method that performs
averaging for each column, since a large number of averaging
parameters are obtained by calculating average values for each of a
plurality of regions that include a plurality of pixels, a highly
accurate average value can be acquired even from a pixel region
that includes as few as two rows.
[0110] The average values of pixel signals of the second
ineffective pixel region 2302 (pixel signals of 100th row to 101st
row.times.6 times=pixel signals written in 100th row to 111th row
of frame memory) that are read out from the frame memory 2316 are
calculated by the second averaging circuit 2310. As shown in FIG.
22, the second averaging circuit 2310 calculates an average value
for each column and stores the calculated values in a second memory
2311. Since averaging is performed for each column, pixel signals
for 12 rows are used according to the present embodiment to
increase the averaging parameters.
[0111] Next, the two average values that are calculated are
respectively subtracted at subtracting circuits 2312 and 2313 from
the pixel signals of the effective pixel region 2303 (pixel signals
of 3rd row to 99th row=pixel signals written in 3rd row to 99th row
of frame memory) of the frame memory 2316. As illustrated in FIG.
23, for the purpose of cancelling out the effect of subtracting two
average values, an adding circuit 2315 may add an offset 2314 for
black level adjustment to the output signal of the subtracting
circuit 2313.
[0112] The ineffective pixel region includes the first ineffective
pixel region 2301 and the second ineffective pixel region 2302
which consist of mutually different rows. The pixels of the first
ineffective pixel region 2301 are light shielded pixels in which
the surface of pixels in which a semiconductor impurity region for
accumulating a charge is formed is covered with a light-shielding
film. The pixels of the second ineffective pixel region 2302 are
pixels in which a semiconductor impurity region for accumulating a
charge is not formed. The vertical scanning circuit 2304 selects
pixels of the same row of the effective pixel region 2303 in row
units one time during one frame and selects pixels of the same row
of the second ineffective pixel region 2302 in row units a
plurality of times during one frame. The correction circuit has a
first correction circuit and a second correction circuit. The first
correction circuit has the first averaging circuit 2308 and the
subtracting circuit 2312. The first correction circuit averages
pixel signals of the first ineffective pixel region 2301 to correct
the pixel signals of the effective pixel region 2303. The second
correction circuit has the second averaging circuit 2310 and the
subtracting circuit 2313. The second correction circuit averages
pixel signals of the second ineffective pixel region 2302 to
correct the pixel signals of the effective pixel region 2303.
[0113] According to the above described method, it is possible to
accurately suppress fixed pattern noise occurring in each column as
well as the influence of dark current of the solid-state imaging
apparatus.
[0114] According to the present embodiment, the effects produced by
dark current of the solid-state imaging apparatus are suppressed by
performing subtraction processing for pixel signals of the
effective pixel region 2303 using pixel signals of the first
ineffective pixel region 2301 that are digitized by the AD
convertor 2307. Instead, a configuration may be adopted in which
dark current components included in pixel signals after A/D
conversion by the AD convertor 2307 are suppressed using pixel
signals of the first ineffective pixel region 2301 for black level
adjustment of the AD convertor 2307.
[0115] Although according to the present embodiment pixel signals
that are digitized by the AD convertor 2307 are temporarily stored
in the frame memory 2316, a configuration may be adopted in which
the frame memory 2316 is eliminated. For example, according to a
focal plane readout operation as illustrated in FIG. 21, pixel
signals of the second ineffective pixel region 2302 of the nth
frame are averaged by the second averaging circuit 2310 and stored
in the first memory 2309. Subsequently, pixel signals of the first
ineffective pixel region 2301 of the N+1 frame are averaged by the
first averaging circuit 2308 and stored in the second memory 2311.
If the two average values that are obtained are subtracted from the
pixel signals of the effective pixel region 2303 of the N+1 frame,
the frame memory 2316 can be omitted.
[0116] As described in the foregoing, according to the first to
fifth embodiments, even when there is only a small number of rows
in an ineffective pixel region for averaging, by selecting a
vertical selection signal over a plurality of row periods during
one frame, pixel signals thereof can be output for a plurality of
rows. By averaging pixel signals that are output over a plurality
of row periods to calculate a correction signal, the influence of
random noise included in the correction signal can be reduced. The
same applies to the sixth and seventh embodiments. Further, since
it is not necessary to have a large number of rows in an
ineffective pixel region as in the prior art, the chip area can be
reduced.
[0117] Further, since it is not necessary to average pixel signals
of an ineffective pixel region over a plurality of frames as in the
prior art, fixed pattern noise can be suppressed to a high degree
of accuracy even in an imaging mode that captures only a single
frame such as the case of a single image captured with a digital
still camera.
[0118] It should be understood that the above described embodiments
merely describe specific examples for implementing the present
invention and are not intended to limit the technical scope of the
present invention. That is, the present invention may be
implemented in various forms without departing from the technical
concept and principal features thereof.
[0119] This application claims the benefit of Japanese Patent
Application No. 2007-134495, filed May 21, 2007, and Japanese
Patent Application No. 2007-337461, filed Dec. 27, 2007, which are
hereby incorporated by reference herein in their entirety.
* * * * *