JPH02312486A - Picture data processing unit - Google Patents

Abstract

PURPOSE:To simplify circuit constitution by replacing linear product sum calculation in common to a digital filter processing including simultaneous and frame processing of a picture signal into a single linear product sum arithmetic circuit and setting an arithmetic coefficient one after another in response to the calculation applied to point sequential picture data in time series. CONSTITUTION:A digital filter circuit 120 whose arithmetic coefficient is set variably, applying linear product sum calculation to an input picture data with the set arithmetic coefficient and outputting a color component data subjected to picture signal processing, a coefficient generating circuit 124 setting the arithmetic coefficient in response to the input picture data to the digital filter circuit 120 synchronously with the input picture data and a FIFO memory 138 outputting an output picture data whose color component data timing is substantially made coincident are provided and common linear product sum calculation in common to the digital filter processing including simultaneous and frame processing of the picture signal is replaced into a single linear product sum arithmetic circuit 122. Thus, the circuit constitution is simplified.

Description

TECHNICAL FIELD The present invention relates to an image data processing apparatus, such as an image data processing apparatus for processing image data stored in a memory, for example, for file or playback purposes in a digital electronic still camera system.

BACKGROUND ART Digital electronic still cameras have been proposed that store images captured by an imaging device in the form of digital data in a semiconductor storage device such as a RAM memory card. See, for example, the applicant's pending patent application, Japanese Patent Application No. 132-270387.

Images taken with a digital electronic still camera can be reproduced by removing the memory card storing the image data from the camera, loading it into a reproduction device, and using a video camera or printer. Such a reproducing device is described, for example, in Japanese Patent Application Laid-Open No. 57-124364 and in the applicant's pending patent application, Japanese Patent Application No. 82-213058.

In these conventional digital electronic still camera systems, image data output from the camera to a memory card is processed by white balance adjustment, gradation (γ) correction, and synchronized video signals of the color separation filter array of the imaging device. It is a signal that has been subjected to various types of video signal processing, such as simultaneous processing, and is often in the form of color difference signals and luminance signals. However, in this conventional method, the video signal processing function was performed by the camera's analog processing circuit.

Therefore, this video signal processing does not always match ideal synchronization characteristics due to hardware constraints of analog circuits. Furthermore, the presence of such a processing circuit complicates the circuit configuration of the camera, which inevitably increases the size of the device.

By the way, in order to reproduce a still image video signal obtained from a video signal source such as a CCD as a clear visible image on an image reproducing device such as a TV monitor, it is necessary to remove noise and emphasize the outline of the image. Requires processing. The signal processing in FIG. 4 includes processing such as white balance adjustment 1601, gradation (γ) correction 1621, and edge enhancement 190, as illustrated in FIG.

For example, the input video signal has three primary colors, red (R),
If it is a dot-sequential field video signal of green (G) and blue (B), in order to synchronize it and convert it into a frame video signal, a luminance/color difference matrix] links 164, a low-pass filter (LPF) 178 etc., framing circuit 18
0 etc. and an RGBization matrix 184.

These functional units require dedicated digital circuits when implemented through pipeline processing.

Further, filter circuits such as the low-pass filter 176 are realized by a dedicated digital filter LSI. Furthermore, since the wide-area luminance signal YH is created at the timing of one pixel, there is a timing difference of two pixels with respect to the color difference signals R-YL and B-YL. In order to compensate for the difference in timing between the two, many FIFO circuits such as the FIFO circuit 166 must be used.
It is necessary to provide O memory.

Note that the low-pass filter technology for filtering point-sequential color signals in their time series is disclosed in Japanese Patent Laid-Open No. 57-145492.
known by. This conventional technique replaces a plurality of low-pass filters and high-pass filters with a single low-pass filter having a single function of low-pass filtering. This low-pass filter has no function other than low-pass filtering.

SUMMARY OF THE INVENTION It is an object of the present invention to overcome the drawbacks of the prior art and to provide an image data processing device with a simple device configuration.

DISCLOSURE OF THE INVENTION According to the present invention, an image data processing device that receives point-sequential color input image data and performs image signal processing including low-pass filtering and synchronization is provided, in which arithmetic coefficients can be variably set. A digital filter circuit performs a linear product-sum calculation using the set calculation coefficients and outputs color component data that has undergone image signal processing, and a digital filter circuit performs calculations according to the input image data in synchronization with the input image data. It has a coefficient generation circuit that sets coefficients, and a FIFO memory that outputs output image data by substantially matching the timing of color component data.

DESCRIPTION OF EMBODIMENTS Preferred embodiments of the present invention will now be described in detail with reference to the accompanying drawings. Referring to FIG. 2, an embodiment of an image data processing device according to the invention is applied to an image reproduction device advantageous, for example in an electronic still camera system, and has input capacitors 100 and 102. The image reproduction device is
A semiconductor memory device such as a RAM reads out image data stored in a memory card carried by a card-like base, performs image data processing, and stores the image data in a file storage device such as an optical disk, or on a video monitor or the like. This is an image processing device that reproduces images on an image output device such as an image printer. Such image data is input to the input port 100.

In a digital electronic still camera, a memory card that is removably attached to the camera has red (R) and green (
One field of video signal data is stored in the form of color component signals of G) and blue (B) by point-sequential rask scanning.

The memory card storing image data is removed from the camera and loaded into an image playback device. The image reproducing device carries the R data on the memory card connected to it.
The memory contents are read from the AM, and this image data is input into the input port 100 in bit parallel. Further, a synchronization signal synchronized with this readout is applied to the input port 102 and input to the synchronization signal circuit 104. The image signal data is repeatedly input in sequence for one field in this manner.

The image data processing device is a processing device that performs various image signal processing such as white balance adjustment, tone correction, pixel signal synchronization, edge enhancement, and framing. Therefore, the output +18 contains the RGB of the processed l frame.
Image data is output. The synchronized RGB data consists of one set of color component data R1G and B, which corresponds to one set of color component data R1G and B.
It constitutes a pixel. The synchronized RGB image data is stored in a frame buffer, for example, and is used for later image reproduction, for example, for visual display as a color image on a video monitor, or for image output in other forms such as a printer or analog communication device. Ru.

The image data processing device, whose processing functions are conceptually shown in FIG. 2, performs the following image signal processing. The white balance of the RGB image data input to the input port 100 is adjusted by a white balance adjustment section 180. The image data whose white balance has been adjusted is then sent to the gradation correction section 1.
The gradation (γ) is corrected in G2. If the imaging device of the digital electronic still camera used has γ=1, the gradation is corrected to a value of γ suitable for, for example, standard TV signal specifications.

The gradation-corrected RGB image data is then converted into a luminance signal YL and a color difference signal R- in a luminance color difference matrix 164.
YL and B-YL. The luminance signal YL is
It has the same frequency band as the color signal, for example, a low band of about 0.7 MHz, and in order to contribute to faithful color reproduction for a specific pixel, the color component signals of two adjacent pixels are adjusted to suit the visibility. This is a luminance signal mixed using a mixing coefficient.

In this embodiment, the color component signal obtained as the output of the gradation correction section 1B2 is a wide band luminance signal YH of about 4.2 MHz.
This wide area luminance signal YH is created at the timing of one pixel. The low frequency color difference signal YL obtained by the luminance color difference matrix 164 and the gradation correction section 162
The difference YL-YH from the wide-area luminance signal Y)I obtained from the subtraction section +74 is obtained. Note that the luminance signal YL has a frequency band comparable to that of the color signal, for example, a low band of about 0.7 MHz, and in order to contribute to faithful color reproduction for a specific pixel, the luminance signal YL is a color component signal of two adjacent pixels. are mixed with a mixing coefficient suitable for visibility. Therefore, timing for three pixels is required to create the luminance signal YL. In the functional configuration of FIG. 2, the matrix 164 generates the color difference signal R-YL by creating a low-band luminance signal YL and subtracting it from the color component signals R and B.
and create B-YL.

The color difference signals R-YL and B-YL are subjected to low-pass filters (LPF) 176 and 178, respectively, to remove high frequency components of, for example, 0.7 MHz or higher, and to be framed 180 and 182. Wide area luminance signal Y
H is band-limited to 4.2 MHz or less from the gradation correction section 162 through the low-pass filter 184, and then passed through the addition section 188.
is input. Further, the difference signal YL-YH is also input to the adder 188 through the low-pass filter 188. An adder 188 adds both signals to create a brightness signal Y, and then an outline emphasis section 190 emphasizes the outline of the picture represented by this brightness signal Y. The luminance signal Y whose outline has been emphasized in this manner is subjected to framing 182.

By the way, the image data input to the input port loo carries only image information for one field, as described above. Therefore, the color difference signals R-YL and B input to the framing units 180, 182 and I82, respectively.
-YL and the luminance signal Y, image information of other fields is created in a pseudo manner in the respective framing units, that is, they are framed. This framing is performed, for example, by additively averaging the color difference signals R-YL and B-YL and the luminance signal Y between two adjacent horizontal scanning lines in the I field pixel-wise. This is done by creating a pseudo-form.

These frames (color difference signals R-YL and B-Yt, framed in 11180, 182 and 192,
and the luminance signal Y is input to an RGB matrix 194, where it is interlaced scanned as a two-field I? C
It is converted into a B signal. This I? The Gfl signal is output from output port 116. Therefore, this output image data is synchronized and framed so that one pixel is composed of color component signals RGB of the three primary colors.

Of these functional parts of image data processing equipment.

Luminance color difference matrix 1θ4. Subtraction ffB174. Low pass filters 178, 178, 184 and 18B, adder 188. As well as I? In this embodiment, the various functions of the CB matrix 184 are realized by a single arithmetic circuit 122 including a digital filter circuit +20, as shown in FIG. This is because the individual functional units mentioned above have many linear product-sum operations that can be processed by the digital filter LSI, and the combination of multiple linear product-sum operations can be replaced with a single linear product-sum operation. This is because the circuit configuration has been simplified with special consideration.

The arithmetic circuit 122 includes a digital filter circuit +20,
Coefficient generation circuit 1 that gives filter coefficient 0n(i) to this
This is a linear product-sum operation circuit having 24 circuits. The digital filter circuit 120 is realized, for example, by a finite impulse response (FIR) filter, as shown in one specific example in FIG. This filter circuit 120 has N-
It has one stage (N is a natural number) delay circuit (T) 12B,
Pixel data P(K) is input from the gradation correction circuit 62 to the input 132 at the first stage. The delay circuit 12B at each stage is a circuit that applies a delay of one pixel data to pixel data in synchronization with a synchronization signal that is applied from the synchronization signal section 104 and corresponds to the pixel data. A multiplication circuit 128 is connected to the input/output of each stage 12B, and the filter coefficients On(i) are supplied from the coefficient generation circuit 124 to the N multiplication circuits 128 in synchronization with the pixel data. However, K is 3IIl+n
, m is 0 or a positive integer, n is 0.1 or 2. Also i
is an integer from O to N-1. The output of the multiplier circuit 12 is summed in the adder circuit +30, and the image data Q (
K) at the output 134 of the filter circuit 120.

Coefficient generation circuit 124 that generates filter coefficient 0n(i)
In this embodiment, is in the form of a ROM in which nxN filter coefficient values are stored in a semi-fixed manner.

This circuit outputs these values in synchronization with a synchronization signal corresponding to pixel data given from the synchronization signal section 104. In this embodiment, the filter circuit 120 and the coefficient generation circuit 12
4 is composed of one digital filter LSI.

For example, as shown in FIG. 5, when the input data 132 is input by repeating the pixel order of the color components RGB, the filter coefficient Cn(i) is λn for the R component, G Q
The minute is expressed as ILn, and the B component is expressed as νn. These coefficients are obtained by the following equations.

λn(i)=Ct(i)−p(i)・A
rgb(n+i)+K life τ c(i) δ (n
+1) gn(i) = Ct(i) −p (i)
・Argb(n+i)+Ke τc(i)δ(n+1
−1)v n(i)=Gt(i)−p(i)
・Argb(n+i) 10・τc(i)δ(n+i
+1) Here, if the resolution coefficients for the color components RGB of the low-frequency luminance signal YL are respectively at, ag, and ab, then YL=ar*R+ag*G+ab*B.

Therefore, if we set δ(K)-1 (when n-0) and δ(K)=O (when n=1 or 2), Argb(n+i) = ar * δ (1+i
)+6g red δ (n+1-1)+ab・δ(n+i+
1) It is expressed as In addition, 1 color difference signal system digital filter 1
76 and +78 coefficients as cc(i), luminance difference signal YL
−The coefficient of the YH system digital filter 186 is Cd(i)
.

Also, a digital filter 184 for the high-frequency luminance signal YH system.
If the coefficient of is Ch(i), then CL(i) = 0h(i)-Cd(i)ρ(i)=K
・τc(i)−τd(i)τc(i)=Gc(i−1
)+Cc(i)+Gc(n+1)τd(i)=Cd(
i-1)+Cd(i)+C:d(n+1) However, the constant is the color weight of the RGB conversion matrix 194, R=Y+(R-Y) B=Y+(R-Y) ) G = Y-(KIIar/ag)(RY)-(KΦ
ab/ag)(BY). Further, undefined constants are 0.

When the input image data P(K) is input to the input 132 of the digital filter circuit 120, the circuit +20 performs the following calculation using these fill coefficients input n(i), μn(i) and ′νn(i). is performed, and output data Q(K) is output.

Here, the coefficient 0n(i) is the filter coefficient input n(i).

.mu.n(i)8 and .nu.nc+) and is given from the coefficient generation circuit +24 according to the input image data P(K), as illustrated in FIG. Therefore, the calculation according to this formula is performed by the digital filter circuit 120, and the output data Q(K) is outputted to the output 134.
It is output in the form of envelope components RGB. In the example in the same figure, the coefficient is 0n
(i) is selectively set and used one after another from nine types. A coefficient Cr+(i) is selectively set in the digital filter circuit 120 in synchronization with the input image signal data P(K), and calculations are performed for each color component in a time-sharing manner.

As shown in the above equation, only 173 pieces of data of one color component data being processed can be obtained in one digital filter process. Therefore, in this example.

Field FIFO circuit 1 is connected to the output of filter circuit 120.
36, which rearranges each color component data so that complete data for one color component can be obtained at substantially the same timing.

The output 134 three-color component image data Q(K) is then subjected to an outline emphasizing section 138 where the outline of the picture it represents is emphasized, and a framing section 140 where image information of other fields is created in a pseudo manner. Framed.

Therefore, the image data output from the output JIB is in the form of 2-field RGB signal data that has been interlaced scanned, and is synchronized and framed so that one pixel is composed of 17 GB of color component signals of the three primary colors. It becomes something.

In this way, in this embodiment, the linear product-sum calculation, which is common in digital filter processing including image signal synchronization and framing, is performed using a single linear product-sum calculation circuit, that is, the calculation circuit 12.
2 to simplify the circuit configuration. As shown in FIG. 3, the arithmetic circuit 122 has the same circuit configuration as the low-pass filter with the longest number of taps, thereby eliminating the need for other low-pass filter circuits and addition and subtraction circuits. Furthermore, one set of framing circuits 140 is sufficient, and the cost of 2 cents can be omitted. Therefore, the luminance color difference matrix 164 shown in FIG. Subtraction unit 174. Low pass filter 1
78.178 and 188. Addition part】88. Framing units 180 and 182. In addition, the RGB conversion matrix 184 has a reduced amount of corresponding hardware.

Note that since the wide-area luminance signal YH is created at a timing corresponding to one pixel, there is a timing difference corresponding to two pixels with respect to the color difference signals R-YL and B-YL. For this reason, the second
If the functional blocks shown in the figure are realized by separate hardware, it is necessary to provide a FIFO circuit 168 at the output of the gradation correction section 162, as shown in FIG. This is F
It is used together with IFO circuits 168, 170 and +72 to compensate for the difference in timing between the two. However, in the embodiment shown in FIG.
fi8. +70 and +170 are not required, and the timing between each color component data is adjusted in the field FIFO circuit 136. Therefore, the circuit configuration required for signal processing is greatly simplified.

In this embodiment, all of the above-mentioned elements of the signal processing unit 108 are advantageously constructed as digital hardware circuits. Therefore, unlike video signal processing using analog circuits that was conventionally performed in electronic still cameras,
Since image data processing is realized by hardware using digital circuits, this image data processing has close to ideal characteristics and can be performed at high speed. In this embodiment, the image data output from this device is synchronized and framed RGB data that is commonly used in other systems, so it is compatible with other systems. .

The embodiments described here are for illustrating the present invention, and the present invention is not necessarily limited thereto, and variations and modifications that can be made by those skilled in the art without departing from the spirit of the present invention are possible. Within the scope of the present invention. For example, the input image f4' signal data is of RGB primary color components, but if it is l1ffi Europeanized image data, it may be complementary color system or luminance/color difference signal system image data. Can be applied effectively. Also in this example. It is also possible to obtain luminance color difference signal data at the output 116 using the same circuit configuration. In that case, the filter coefficient Cn(i) given to the digital filter circuit 120 may be set to a value suitable for the luminance/chrominance system. In order to adapt the present device to the requirements of these input/output signal data, it is sufficient to write corresponding storage data into the ROM of the coefficient generation circuit 122.

Effects As described above, according to the present invention, linear product-sum calculations common in digital filter processing including image signal synchronization and framing are replaced with a single linear product-sum calculation circuit, and point-sequential image data can be processed in time. The calculation coefficients are configured to be set one after another according to the calculations performed in series. Therefore, the circuit configuration is simplified. Since the image data processing is realized using a digital circuit/ware, it is possible to realize high-speed image data processing with close to ideal characteristics. The present invention is particularly advantageously applied to a playback device for a digital electronic still camera.

[Brief explanation of the drawing]

FIG. 1 is a circuit block diagram showing the overall configuration of an embodiment of the image data processing device according to the present invention, and FIG. 2 is a functional block diagram conceptually showing the function of image data processing in the image data processing device shown in FIG. Figure 3 is a block diagram showing a configuration example of a digital filter circuit applied to the same embodiment, and Figure 4 is a configuration example when the processing function of the same embodiment is realized by separate /\-ware. The functional block diagram shown in FIG. t55 is an explanatory diagram for explaining the operation of the embodiment shown in FIG. Explanation of symbols of main parts 120, . . . Digital filter circuit + 22. , Arithmetic circuit +24. ,,Coefficient generation circuit 136,,,Field FIFO circuit! 38. ．． ．． Contour enhancement part +40. ,,framing section 180,,,white balance adjustment section 1B2. ,, Tone correction unit patent applicant Fuji Photo Film Co., Ltd. Agent Number Number: Royusenzan Takaoso 7f Concave 7-1-in Kai Gokyu -11111------------------------□
-1FF, G, B Tool 2 recess □ Output 111a □ Earth ji map input value also meets, Wight Bown Kushibiki q /60r supplement, yo /6
2 //:, 4
/6 Otsu ■ degree color difference Mado 131,7 Ku
FrF;01″T4 *Kan F?-YL IL -Yl-I Y
l-1LPF LPF LPF LPFF
IFO 1/J3 7)'II 笥 1BF3 #5 Figure 170 /7. ? FIFOFIFO song /'? 0 -YL -YL Y frame
Framing Framing/q2・8-18ri

Claims (3)

[Claims] 1. An image data processing device that receives point-sequential color input image data and performs image signal processing including low-pass filtering and synchronization. a digital filter circuit that performs a linear product-sum operation using set calculation coefficients and outputs the color component data that has been subjected to the image signal processing; An image data processing device comprising: a coefficient generation circuit that sets the arithmetic coefficients; and a FIFO memory that outputs output image data by substantially matching the timing of the color component data. 2. 2. The image data processing apparatus according to claim 1, wherein the input image data is image data of a field, and the apparatus includes a framing means for framing the image data of the field. 3. 2. The image data processing apparatus according to claim 1, further comprising an outline emphasizing means for emphasizing the outline of a picture in said input image data.