Digital Camera Patent AbstractA digital camera includes a shutter button. An optimal compression
ratio of an image signal fetched in response to a photographing
instruction from the shutter button is predicted prior to the photographing
instruction. More specifically, an image signal fetched before application
of the photographing instruction is compressed by a JPEG codec,
and the optimal compression ratio is predicted based on a size of
a compressed image signal thus obtained. An image signal fetched
in response to the photographing instruction is compressed on the
basis of the optimal compression ratio, and a compressed image signal
thus obtained is recorded onto a recording medium.
Digital Camera Patent ClaimsWhat is claimed is:
1. A digital camera that periodically creates, until a recording
Instruction is issued, a first still image corresponding to an object
scene in such a manner as a real time display is omitted, and creates,
when the recording instruction is issued, a second still image corresponding
to the object scene so as to record the second still image into
a recording medium in a compressed state, comprising: a calculator
for calculating a specific compression ratio coefficient capable
of compressing the first still image to a specific size; a compressor
for compressing the second still image by use of the specific compression
ratio coefficient; a corrector for correcting the specific compression
ratio coefficient when a compressed second still image created by
said compressor fails to satisfy a size condition including the
specific size; and a recorder for recording the compressed second
still image satisfying the size condition into said recording medium,
wherein the first still image and the second still image have the
same resolution with each other.
2. A digital camera according to claim 1, wherein said calculator
includes a first still image compressor for compressing the first
still image, and a calculation executor for calculating the specific
compression ratio coefficient on the basis of a compressed first
still image created by said first still image compressor.
3. A digital camera according to claim 1, wherein said recorder
records a newest compressed second still image into said recording
medium when the number of compressing operations directed to the
second still image reaches a threshold value.
4. A digital camera according to claim 1, further comprising a
shutter button to issue the recording instruction. Digital Camera Patent DescriptionBACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a digital camera applied to a
camera used by the consuming public as a hobby and a surveillance
camera set in a bank for security. More specifically, the present
invention relates to a digital camera for recording a photographed
image signal generated in response to a photographing instruction
onto a recording medium in a compressed-state.
2. Description of the Prior Art
In a conventional digital camera (1) for recording a motion image
formed by a plurality of still images, a still image signal of a
first frame is compressed at an initial compression ratio, and a
compression ratio after a second frame and later is determined on
a size of a compressed still image signal at the previous frame.
In a conventional digital camera (2) for photographing a still image,
a photographed still image signal is compressed at each different
compression ratio so as to detect a compression ratio capable of
obtaining a compressed still image signal of a target size.
However, there is a problem in the digital camera (1) that as to
the several frames from the top, the compression ratio is largely
deviated from an optimal compression ratio depending on the object,
and therefore, a compressed image size exceeds a target size where
a compression ratio is too small, and a noise occurs in a decompressed
image where a compression ratio is too large. Furthermore, there
is a problem in the digital camera (2) that the more compression
processing is repeated, the longer it takes a time from obtaining
of a photographed image signal to end of the recording of a compressed
image signal, and therefore, a shutter interval becomes longer.
SUMMARY OF THE INVENTION
Therefore, it is a primary object of the present invention to provide
a novel digital camera.
Another object of the present invention is to provide a digital
camera capable of performing proper recording processing from the
start when recording a motion image.
Further object of the present invention is to provide a digital
camera capable of shortening a shutter interval when recording a
still image.
According to the present invention, a digital camera which records
a photographed image signal generated in response to a photographing
instruction into a recording medium in a compressed state, comprises:
a predictor for predicting an optimal compression ratio capable
of compressing the photographed image signal up to a target size
prior to the photographing instruction; a compressor for compressing
the photographed image signal on the basis of the optimal compression
ratio; and a recorder for recording a compressed photographed image
signal generated by the compressor into the recording medium.
When the photographed image signal generated in response to the
photographing instruction is recorded onto the recording medium
in a compressed state, the predictor predicts the optimal compression
ratio capable of compressing the photographed image signal up to
the target size prior to the photographing instruction. The photographed
image signal is compressed by the compressor on the basis of the
optimal compression ratio, and the compressed photographed image
signal thus generated is recorded onto the recording medium by the
recorder.
The optimal compression ratio of the photographed image signal
is predicted prior to the photographing instruction. Accordingly,
as to a digital camera for recording a motion image, a proper motion
image recording can be performed immediately after application of
the photographing instruction. As to a digital camera for recording
a still image, a shutter interval can be shortened.
In a preferable embodiment, it is determined whether or not a first
size of the compressed photographed image signal satisfies the size
condition including the target size by a determiner. A corrector
corrects the optimal compression ratio in accordance with the determination
result of the determiner, and a recorder records the compressed
photographed image signal according to the corrected optimal compression
ratio.
In another preferable embodiment, an object image is photographed
by a photographer prior to the photographing instruction. An image
signal outputted by the photographer is compressed by an image compressor,
and the optimal compression ratio is predicted on the basis of a
second size thus obtained of the compressed image signal. The image
signal and the photographed image signal have the same resolution
with each other, and in a case the photographer outputs the image
signal at predetermined interval, an image based on each image signal
may be displayed.
Furthermore, in a case a memory having a plurality of memory areas
is prepared, and the outputted image signal at a predetermined interval
from the photographer is selectively written to the plurality of
memory areas, the image compressor should read the image signal
from a memory area to which no writing is performed.
In another preferable embodiment, the digital camera has a motion
image recording function, and the photographed image signal is a
still image signal of a top frame forming a recording motion image
signal.
In further preferable embodiment, the photographed image signal
is a still image signal of one frame.
The above described objects and other objects, features, aspects
and advantages of the present invention will become more apparent
from the following detailed description of the present invention
when taken in conjunction with the accompanying drawings.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 is a block diagram showing one embodiment of the present
invention;
FIG. 2 is an illustrative view showing a mapping state of a display
image area formed on an SDRAM;
FIG. 3 is a flowchart showing a part of operation in FIG. 1 embodiment;
FIG. 4 is a block diagram showing another embodiment of the present
invention;
FIG. 5 is a flowchart showing a part of operation in FIG. 4 embodiment.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
Referring to FIG. 1, a digital camera (surveillance camera) 10
of this embodiment has a motion image recording function and includes
an optical lens 12. An optical image of an object is incidented
to a light-receiving surface of an image sensor 14 through the optical
lens 12. The light-receiving surface is provided with a color filter
(not shown) of a primary color Bayer array, and a charge amount
generated in each light-receiving element reflects light quantity
corresponding to each primary color component.
When a power switch 46 is turned on, a system controller 44 applies
to a CPU 42 a state signal indicative of a state of the power switch
46 and a resolution change-over switch 50. The CPU 42 sets a selected
resolution to a TG/SG 16, a signal processing circuit 22 and a video
encoder 30, and activates a signal processing block including the
image sensor 14, the TG/SG 16, the signal processing circuit 22,
etc. and an encode block including the video encoder 30, a monitor
32, etc. It is noted that selectable resolutions are two kinds of
QVGA (320 pixels.times.240 lines) and VGA (640 pixels.times.480
lines).
When a resolution of "QVGA" is selected, the TG/SG 16
generates a vertical synchronization signal every 1/30 second, and
performs thin-out reading of 30 fps on the image sensor 14 in response
to the vertical synchronization signal. An RGB signal having a vertical
number of pixels of "240" is outputted every 1/30 second
from the image sensor 14. On the other hand, when a resolution of
"VGA" is selected, a vertical synchronization signal is
generated every 1/15 second, and in response to the vertical synchronization
signal, thin-out reading of 15 fps is performed on the image sensor
14. An RGB signal having a vertical number of pixels of "480"
is outputted from the image sensor 14 every 1/15 second.
The outputted RGB signal of each frame is subjected to noise removal
and level adjustment in a CDS/AGC circuit 18, and the RGB signal
thus processed is applied to the signal processing circuit 22 through
an A/D converter 20. The signal processing circuit 22 performs a
horizontal thining-out, color separation, white balance control,
YUV conversion, etc. on the applied RGB signal so as to generate
a YUV signal. When a resolution of "QVGA" is selected,
a horizontal number of pixels is thinned out to "320",
and a YUV signal of a QVGA resolution is outputted every 1/15 second
from the signal processing circuit 22. On the other hand, when a
resolution of "VGA" is selected, the horizontal number
of pixels is thinned out to "640", and a YUV signal of
a VGA resolution is outputted every 1/30 second from the signal
processing circuit 22. The outputted YUV signal is written to a
display image area 28a of an SDRAM 28 through a memory control circuit
24.
The video encoder 30 reads out the YUV signal stored in the display
image area 28a through the memory control circuit 24, and converts
the read YUV signal into a composite image signal. The YUV signal
is read out in an interlace-scan scheme every 1/60 second, and the
read YUV signal of each field is converted to the composite image
signal. The converted composite image signal is applied to the monitor
32, and therefore, a real-time motion image (through image) of the
object is displayed on a screen. It is noted that since a resolution
of the monitor 32 is lower than that of QVGA, thin-out processing
is performed in a corresponding manner to a selected resolution
in the video encoder 30, and whereby, a composite image signal having
the same resolution as the monitor 32 is generated.
A bank switch circuit 26 generates a bank switch signal in response
to the vertical synchronization signal outputted from the TG/SG
16 and applies the generated bank switch signal to the memory control
circuit 24. When the resolution of "QVGA" is selected,
since the vertical synchronization signal is generated every 1/30
second, a level of the bank switch signal is switched between a
high-level and a low-level every 1/30 second. When the resolution
of "VGA" is selected, since the vertical synchronization
signal is generated every 1/15 second, the level of the bank switch
signal is switched between the high-level and the low-level every
1/15 second.
The display image area 28a is formed with banks A and B as shown
in FIG. 2. The memory control circuit 24, when the bank switch signal
indicates the high-level, considers the bank A as a writing destination
of the YUV signal and the bank B as a reading destination of the
YUV signal. When the bank switch signal indicates the low-level,
the bank B is considered to be a writing destination of the YUV
signal, and the bank A is considered to be a reading destination
of the YUV signal.
Accordingly, as to the frame in which the bank switch signal indicates
high in level, the YUV signal outputted from the signal processing
circuit 22 is written to the bank A, and the YUV signal read from
the bank B is applied to the video encoder 30. As to the frame in
which the bank switch signal indicates low in level, the YUV signal
outputted from the signal processing circuit 22 is written to the
bank B, and the YUV signal read from the bank A is applied to the
video encoder 30. It is noted that when the resolution of "QVGA"
is selected, the signal processing circuit 22 writes the YUV signal
of 15 fps to the SDRAM 28 through the memory control circuit 24.
Accordingly, the video encoder 30 reads out the YUV signal of the
same frame twice.
When a through image is displayed on the monitor 32, the CPU 42
applies a compression instruction to a JPEG codec 34 in response
to the vertical synchronization signal. The JPEG codec 34 reads
the YUV signal of each frame from the display image area 28a through
the memory control circuit 24. The reading destination is described
above. The YUV signal is read out of the bank B where the bank switch
signal is high in level, and the YUV signal is read out of the bank
A where the bank switch signal is low in level.
The JPEG codec 34 actually subjects the read YUV signal to JPEG
compression so as to determine a size (compressed size) of a compressed
YUV signal. The determined compressed size value is applied to the
CPU 42. The CPU 42 obtains a compression ratio of the next frame
Q f next according to a equation 1 based on the applied compressed
size value, a target size value and a compression ratio of the current
frame Q f crnt. Q f next=Q f crnt+100.times.(target size-compressed
size)/target size [Equation 1]
The JPEC compression is performed by a series of processing of
DCT conversion, quantization, and Huffman coding. Herein, the quantization
is performed by utilizing a quantization table, and the compressed
size is determined depending on the amount of a high-frequency component
included in the YUV signal and a table value of the quantization
table. The Qf (Qf crnt or Qf next) is a coefficient for correcting
the table value. For example, the table value is a value of 1.2
times of an initial value for Qf=1.2, and the table value is a value
of 0.9 times of the initial value for Qf=0.9.
In this embodiment, such the Qf is defined as the compression ratio
and the next compression ratio Qf next is calculated according to
the equation 1. Accordingly, a compression ratio set to the JPEG
codec 34 is renewed every one frame even while the shutter button
48 is not operated, and each compression ratio is a value approximate
to an optimal compression ratio capable of compressing the YUV signal
of each frame to the target size. That is, the optimal compression
ratio is predicted according to the calculation according to the
equation 1. It is noted that the target size depends on the selected
resolution.
When the shutter button 48 is pressed, a corresponding state signal
is applied to the CPU 42 from the system controller 44. The CPU
42 instructs the JPEG codec 34 to perform the compression processing
as described above. The JPEG codec 34 reads the YUV signal from
the display image area 28a through the memory control circuit 24,
performs the JPEG compression on the read YUV signal so as to determine
the compressed size, and applies the determined compressed size
vale to the CPU 42. The CPU 42 determines if the applied compressed
size value satisfies a size condition shown in an equation 2 or
not. target size value.times.0.9<compressed size value.ltoreq.target
size value [Equation 2]
When the determined compressed size vale satisfies the size condition,
the CPU 42 shifts to motion image recording processing. On the other
hand, when the determined compressed size value does not satisfy
the size condition, renew processing of the optimal compression
ratio according to the equation 1 and compression processing of
the same YUV signal by the renewed optimal compression ratio are
repeated. Since the compression ratio is periodically renewed prior
to an operation of the shutter button 48, the compressed size value
determined after the operation of the shutter button 48 satisfies
the size condition shown in the equation 2 by performing compression
ratio renew processing, at most, 3 times. Accordingly, the motion
image recording processing is started in response to a second vertical
synchronization signal generated after the shutter button 48 is
pressed.
Taking the fact into consideration that reading out of the YUV
signal from the display image area 28a is late for writing of the
YUV signal to the display image area 28 by one frame, and that the
motion image recording processing is started in response to the
second vertical synchronization signal after the shutter button
48 is operated, the compressed YUV signal of a top frame to be recorded
corresponds to a object image photographed in response to the vertical
synchronization signal immediately after the shutter button 48 is
operated.
Shifting to the motion image recording, the JPEG codec 34 stores
the compressed YUV signal of each frame generated by the JPEG compression
to the compressed image area 28b of the SDRAM 28 through the memory
control circuit 24. Herein, the compression ratio of the next frame
set to the JPEG codec 34 is determined depending on the compressed
size value of the current frame, the target size value and the compression
ratio of the current frame. That is, after shifting to the motion
image recording, the compression ratio is renewed according to the
equation 1. The CPU 42 reads the compressed YUV signal stored in
the compressed image area 28b through the memory control circuit
24, and records the read compressed YUV signal onto the recording
medium 40 through an I/F circuit 36.
In this manner, where the selected resolution is "QVGA",
a motion image signal of the QVGA resolution having a frame rate
of 30 fps is recorded on the recording medium 40 in a compressed
state. On the other hand, where the selected resolution is "VGA",
a motion image signal of the VGA resolution having a frame rate
of 15 fps is recorded on the recording medium 40 in a compressed
state.
It is noted that the CPU 42 is a multitask CPU installed with a
real-time OS such as .mu.i TRON, and the compressed YUV signal accumulated
in the compressed image area 28b is recorded on the recording medium
40 by BG (Back Ground) processing. Furthermore, the recording medium
40 is detachable and is connected to the I/F circuit 36 when attached
to a slot 38.
Before and after an operation of the shutter button 48, the compression
ratio of the next frame is calculated according to the equation
1 on the basis of the compressed size of the current frame. Since
the calculated compression ratio is a compression ratio most suitable
for the current frame, even if the YUV signal of the next frame
is compressed at such the compression ratio, the compressed YUV
signal does not necessarily satisfy the size condition. However,
since the YUV signal is generated every 1/30 second or 1/15 second,
the time period to this extent never drastically change a object
image between the current frame and the next frame. Accordingly,
an appropriate compression ratio is calculated by performing an
arithmetic operation of the equation 1.
When the power switch 46 is turned on, the CPU 42 processes a flowchart
shown in FIG. 3 according to a program stored in a ROM 52. First,
in a step S1, the selected resolution is set to the TG/SG 16, the
signal processing circuit 22 and the video encoder 30, and an initial
compression ratio corresponding to the selected resolution is set
to the JPEG codec 34. When the initial setting is completed, the
signal processing block and the encode block are activated in respective
steps S3 and S5. Thus, the YUV signal of the QVGA resolution or
the VGA resolution is generated at a frame rate of 30 fps or 15
fps, and the through image based on the signal is displayed on the
monitor 32.
It is determined whether or not the shutter button 48 is operated
in a step S7, and it is determined whether or not the vertical synchronization
signal is generated in a step S9. When the vertical synchronization
signal is generated in a state that the shutter button 48 is not
operated, the process proceeds from the step S9 to a step S11 so
as to instruct the JPEG codec 34 to perform the JPEG compression.
The JPEG codec 34 reads the YUV signal from the display image area
28a through the memory control circuit 24, and actually performs
the JPEG compression on the read YUV signal so as to determine the
compressed size. The determined compressed size value is taken from
the JPEG codec 34 in a step S13, and the compression ratio of the
JPEG codec 34 is renewed according to the equation 1 in a following
step S15. Thus, the optimal compression ratio is predicted. When
the resolution of "QVGA" is selected, the optimal compression
ratio is predicted every 1/30 second, and the resolution of "VGA"
is selected, the optimal compression ratio is predicted every 1/15
second.
When the shutter button 48 is operated, the process proceeds from
the step S7 to a step S17 so as to determine whether the vertical
synchronization signal is generated or not. When the vertical synchronization
signal is generated, a counted value rty of a counter 42a is set
to "0" in a step S19, and the JPEG codec 34 is instructed
to perform the JPEG compression in a step S21. The JPEG codec 34
reads, as described above, the YUV signal from the display image
area 28a through the memory control circuit 24 and performs the
JPEG compression on the read YUV signal so as to determine the compressed
size.
The determined compressed size value is taken from the JPEG codec
34 in a step S23, and it is determined whether or not the taken
compressed size value satisfies the size condition of the above-described
equation 2 in a following step S25. Herein, where the size condition
is satisfied, the process proceeds to the motion image recording
processing of a step S27. On the other hand, where the size condition
is not satisfied, a compression ratio is renewed according to the
equation 1 in a step S29, and the counted value rty is identified
in a step S31. If the counted value doesn't reach "10",
the counted value is incremented in a step S33, and then, the process
returns to the step S21. On the other hand, if the counted value
reaches "10", the process proceeds to the motion image
recording processing of the step S27.
In the motion image recording process, the compressed YUV signal
of each frame is accumulated every 1/30 a second or 1/15 a second
in the compressed image area 28b, and the accumulated compressed
YUV signal is recorded on the recoding medium 40 through the I/F
circuit 36. After the motion image recording process is completed
by canceling an operation of the shutter button 48, the process
returns to the step S7.
It is noted that the identification processing of the counted value
rty in the step S31 is processing to avoid infinite execution of
a series of the steps S21 to S33 when the compression ratio satisfying
the size condition shown in the equation 2 is not specified.
According to this embodiment, the optimal compression ratio of
the YUV signal generated in response to an operation of the shutter
button 48 (photographing instruction) is predicted prior to the
operation of the shutter button 48. More specifically, the YUV signal
of the object image photographed prior to the photographing instruction
is compressed by the JPEG codec 34, and the optimal compression
ratio is obtained on the basis of a size of the compressed YUV signal
thus obtained. The YUV signal obtained in response to the photographing
instruction is compressed on the basis of the optimal compression
ratio, and the compressed YUV signal obtained by this compression
processing is recorded on the recording medium 40. It is noted when
the size of the compressed YUV signal according to the photographing
instruction does not satisfy the size condition including the target
size, the predicted optimal compression ratio is corrected, and
the compression processing according to the corrected optimal compression
ratio is executed again.
Thus, since the optimal compression ratio is predicted prior to
the photographing instruction, it is possible to perform an appropriate
motion image recording immediately after the photographing instruction.
Furthermore, a resolution of the YUV signal is the same before and
after the photographing instruction, and it is possible to identify
a object image with ease by displaying on the monitor 32 the through
image based on the YUV signal generated before application of the
photographing instruction.
In addition, the display image area 28a is formed with the banks
A and B, and the YUV signal outputted from the signal processing
circuit 22 every 1/30 second is alternately written to the banks
A and B. The video encoder 30 reads the YUV signal from a bank to
which no writing is performed, and the through image based on the
read YUV signal is displayed on the monitor 32. The JPEG codec 34
also reads the YUV signal from a bank to which no writing is performed,
and performs the JPEG compression on the read YUV signal. Accordingly,
noise doesn't occur on the displayed through image with due to overwriting
of the YUV signal, and furthermore, the JPEG compression can be
executed appropriately.
Furthermore, the problem of image deterioration which occurs at
the several frames from the top when the motion image recording
is started using the initial compression ratio as a reference is
conspicuous in proportion to the height of the resolution. However,
in this embodiment, since the optimal compression ratio is predicted
when any resolution is selected, it is possible to perform appropriate
motion image recording from the start irrelevant to the selected
resolution.
Referring to FIG. 4, a digital camera (surveillance camera) 10
of another embodiment is the same as the FIG. 1 embodiment except
that it has a still image recording function in place of a motion
image recording function, the resolution change-over switch 50,
the bank switch circuit 26, the video encoder 30 and the monitor
32 are omitted, the SDRAM 28 is formed with a photographed image
area 28c in place of the display image area 28a, and the ROM 52
is stored with a program corresponding to a flowchart shown in FIG.
5. Accordingly, a duplicate description will be omitted as little
as possible, giving the same reference number to the same circuit.
When a power witch 46 is turned on, a CPU 42 instructs a TG/SG
16 to read out all the pixels every time a vertical synchronization
signals is generated thirty times. The TG/SG 16 drives an image
sensor 14 in an all the pixel reading-out scheme. Thus, a high-resolution
RGB signals is outputted from the image sensor 14 at a ratio of
one to thirty frames. The outputted RGB signal is applied to a signal
processing circuit 22 through a CDS/AGC circuit 18 and an A/D converter
20 to be converted to a high-resolution YUV signal. The converted
YUV signal is written to the photographed image area 28c through
a memory control circuit 24. It is noted that the vertical synchronization
signal is generated every 1/7.5 second in this embodiment.
The CPU42 instructs, at timing the YUV signal is written to the
SDRAM28, a JPEG codec 34 to perform compression processing. The
JPEG codec 34 reads out the YUV signal stored in the photographed
image area 28c through the memory control circuit 24 and performs
the JPEG compression on the read YUV signal so as to determine a
compressed size. The determined compressed size value is applied
to the CPU 42, and the CPU 42 renews a compression ratio according
to the above-described equation 1. The YUV signal based on the next
all the pixels reading-out is compressed according to the renewed
compression ratio.
When a shutter button 48 is pressed, the CPU 42 instructs the TG/SG
16 to read out all the pixels. Thus, a high-resolution YUV signal
based on the all the pixels reading-out is stored in the photographed
image area 28c of the SDRAM 28 as in the above-described manner.
The CPU 42 instructs the JPEG codec 34 to compress the YUV signal
and to write the compressed YUV signal into the compressed image
area 28b. Accordingly, the JPEG codec 34 reads the YUV signal stored
in the photographed image area 28c through the memory control circuit
24, performs the JPEG compression on the read YUV signal and writes
the compressed YUV signal into the compressed image area 28b through
the memory control circuit 24. The determination of the compressed
size is also performed at this time, and the determined compressed
size is taken-in by the CPU 42.
When the taken compressed size value satisfies the size condition
shown in the equation 2, the CPU 42 sets a recording instruction
of the compressed YUV signal into an instruction list 42b, and renews
a writing address of the compressed image area 28c. The setting
of the recording instruction to the instruction list 42b is a process
for performing the BG process appropriately, and the renewal of
the writing address is a process for preventing the compressed YUV
signal from being overwritten.
On the other hand, when the taken compressed size vale does not
satisfy the size condition, the compression ratio is renewed according
to the equation 1, and instructs the JPEG codec 34 to compress the
same YUV signal and to write the compressed YUV signal to the compressed
image area 28b. Since the writing address of the compressed image
area 28c is not renewed at this time, the compressed YUV signal
based on the previous JPEG compression is overwritten by the compressed
YUV signal based on the current JPEG compression. When the compressed
size satisfies the size condition by repeating the JPEG compression,
the recording instruction of the compressed YUV signal satisfying
the size condition is set to the instruction list 42b, and the writing
address of the compressed image area 28b is renewed. The compressed
YUV signal accumulated in the compressed image area 28b is recorded
on the recording medium 40 by the BG process.
Referring to FIG. 5, processing of the CPU 42 according to a program
stored in the ROM 52 will be specifically described. It is noted
the description of the BG process will be omitted.
The compression ratio of the JPEG codec 34 and the writing address
of the compressed image area 28b are initialized in a step S41,
and a signal processing block is activated in a step S43. It is
determined whether or not the shutter button 48 is operated in a
step S45, and it is determined whether or not the vertical synchronization
signal is generated thirty times in step S47. When the vertical
synchronization signal is generated thirty times in a state the
shutter button 48 is not operated, "YES" is determined
in the step S47. Then, the TG/SG 16 is instructed to read out all
the pixels in a step S49, and the JPEG codec 34 is instructed to
perform compression processing in a step S51.
The TG/SG 16 performs all the pixels reading-out on the image sensor
14. The high-resolution YUV signal based on the all the pixels reading-out
is written to the photographed image area 28c of the SDRAM 28. The
JPEG codec 34 reads out the high-resolution YUV signal stored in
the photographed area 28c through the memory control circuit 24,
and actually performs the JPEG compression on the read high-resolution
YUV signal so as to determine the compressed size. The determined
compressed size value is taken from the JPEG codec 34 in a step
S53, and the compression ratio set to the JPEG codec 34 is renewed
according to the equation 1 in a following step S55. Thus, an optimal
compression ratio of the YUV signal obtained in response to an operation
of the shutter button 48 is predicted.
When the shutter button 48 is operated, "YES" is determined
in the step S45, and it is determined whether or not the vertical
synchronization signal is generated in step a S57. When the first
vertical synchronization signal after the operation of the shutter
button 48 is generated, the process proceeds from the step S57 to
a step S59 so as to set a counted value rty of a counter 42a to
"0". Successively, the TG/SG 16 is instructed to read
out all the pixels in a step S61, and the JPEG codec 34b is instructed
to perform compression processing on a high-resolution YUV signal
and to write the compressed YUV signal to the SDRAM 28 in steps
S63 and S65.
The high-resolution YUV signal based on the all the pixels reading-out
by the TG/SG 16 is stored in the photographed image area 28c of
the SDRAM 28, and the high-resolution YUV signal is subjected to
JPEG compression by the JPEG codec 34. The compressed YUV signal
thus generated is written to the compressed image area 28b of the
SDRAM 28. Furthermore, a size of the compressed YUV signal is determined
by the JPEG codec 34. The compressed size value is taken from the
JPEG code 34 in a step S67, and it is determined whether or not
the taken compressed size value satisfies the size condition of
the equation 2 in a following step S69.
When the size condition is satisfied, the recording instruction
of the compressed YUV signal is set to the instruction list 42b
in a step S71, and the writing address of the compressed image area
28c is renewed in a step S73. When the renewal of the writing address
is completed, the process returns to the step S45 so as to prepare
for the next operation of the shutter button 48. On the other hand,
when the size condition is not satisfied, the compression ratio
is renewed according to the equation 1 in a step S75, and the counted
value rty is identified in a step S77. Herein, where the counted
value rty reaches "10", the process proceeds to the step
S71. Where the counted value rty does not reach "10",
the counted value rty is incremented in a step S79, and the process
returns to the step S63. As a result, the high-resolution YUV signal
stored in the photographed image area 28c is compressed again at
the compression ratio after updating, and the compressed YUV signal
thus obtained is written to the compressed image area 28b.
Since the writing address is not renewed, the compressed YUV signal
obtained by the previous compression is overwritten by the compressed
YUV signal obtained by the current compression.
Also in this embodiment, since the optimal compression ratio is
predicted prior to the photographing instruction, it is possible
to shorten a time from an application of the photographing instruction
to a generation of the compressed YUV signal satisfying the size
condition. That is, it is possible to shorten a shutter interval.
It is noted that in a case the digital camera of the above-described
embodiment is utilized as a surveillance camera, a photographing
instruction is outputted from a timer for detecting a time or a
sensor for detecting unwanted people, and the monitor in the FIG.
1 embodiment is set in a control center.
Furthermore, either CCD type or CMOS type may be utilized as an
image sensor for photographing an object image.
Although the present invention has been described and illustrated
in detail, it is clearly understood that the same is by way of illustration
and example only and is not to be taken by way of limitation, the
spirit and scope of the present invention being limited only by
the terms of the appended claims. |