JPH04170191A - Picture signal processing unit - Google Patents

Abstract

PURPOSE:To reduce production of a false color by applying color signal processing suitable for a picture pattern when the picture pattern from which a false color is easily generated is detected. CONSTITUTION:A matrix selection section 111 discriminates which RGB conversion section is to be used to minimize a false color in response to a pattern of an object and outputs the result to a switch 115. The switch 115 selects R,G,B of outputs of RGB conversion sections 112 114 depending on the result of the matrix selection section 111. The RGB conversion section 112 applies matrix calculation to minimize a false color with respect to a contrast pattern. The matrix calculation optimizing color reproduction under the condition of minimizing the false color is executed by the RGB conversion section 113, and the RGB conversion section 114 processes a color signal into an achromatic color when patterns are overlapped. Thus, occurrence of a false color is reduced.

Description

DETAILED DESCRIPTION OF THE INVENTION (Field of Industrial Application) The present invention relates to an adaptive image signal processing device that changes the method of color signal processing depending on screen conditions.

[Related technology]

FIG. 7, FIG. 9, and FIG. 11 are diagrams showing examples of arrangement configurations of color filters of conventionally known color solid-state imaging devices. In Figure 7, a green light transmission filter (hereinafter "Gr") is shown.
The red light transmitting filter (hereinafter referred to as "Rd filter") is vertically striped.
and a blue light transmission filter (hereinafter referred to as “B1 filter”)
) are arranged horizontally between the Gr filters in every second row and one column.

In FIG. 9, a magenta light transmission filter (hereinafter referred to as "Mg filter"), a green light transmission filter, a cyan light transmission filter (hereinafter referred to as "Cy filter"), and a yellow light transmission filter (hereinafter referred to as "Ye filter") are A total of eight color filters, two pixels in the horizontal direction and four pixels in the vertical direction, constitute one unit, and are arranged in the order shown in the figure.

Furthermore, FIG. 11 shows Rd, Or, B! Each filter has 3 pixels horizontally and 1 pixel vertically.

The pixels are arranged in an offset sampling structure of 141.5 pixels in the horizontal direction.

Figures 8, 10, and 12 refer to Figures 7 and 9, respectively.
The color light carriers in the color filter array configuration shown in FIG.

FIG. 3 is a characteristic diagram of the first quadrant when expressed by fν). In both cases, the pixel pitch in the horizontal direction is PH1, and the pixel pitch in the vertical direction is Pv, O≦fH≦1/P, 0≦fv≦1/2
It represents the range of PV. In both figures, arrows represent carriers of each color, the length of the arrow represents the size of the carrier, and the direction represents the phase relationship.

In FIG. 8, the colored light carriers are, in addition to (0,0),
(1/2PH, O), (1/P)l, O), (0,1/
4Pv), (1/2PH, 1/4Pv), (1/P+
−+, 1/4Pv). this house,(
o, o) and (1/PH, 0) are carriers generated in response to achromatic light and cause aliasing distortion; other carriers are completely canceled in response to achromatic light. However, it does not disappear for chromatic light, causing aliasing distortion.

Similarly, in FIG. 10, the colored light carriers are (0,0
), (1/2PH, 0), (1/PH, O), (1
/2P)l, 1/4Pv), (0,1/2Pv)
, (1/2PH, 1/2Pv).

(1/ Pl (, 1/2 P v ), of which (0,0) and (1/PH,0) are carriers generated for achromatic light; the others are achromatic. It is a carrier that disappears when exposed to light.

Furthermore, in FIG. 12, if the horizontal offset amount of the solid-state image sensor is PH/2, the color light carrier is (0
, 0), (2/3PH, O), (1/3PH, 1/
2Pv), (1/PH.

t/2PV), of which (0,0) and (1/P
o, 1/2Pv) are carriers generated in response to achromatic light, and the others are carriers that disappear in response to achromatic light.

Characteristics of such an offset sampling structure −
In a rectangular sampling structure as shown in FIGS. 7 and 9, (17P, . . . ) in the horizontal direction.

Since there is a carrier generated for achromatic light at the frequency of 0), f++ = 1/2 PII becomes the Nyquist frequency, and no frequency components beyond that can be obtained.
Although the horizontal resolution can only be obtained up to f,=172P,
In the offset sampling structure shown in FIG. 11, (1/PM) in the horizontal direction.

It is generally known that there is no carrier generated for achromatic light at the position 0), and that the Nyquist frequency can be set to f□=s/ps. Therefore, in the color filter arrangement shown in Fig. 11, although the sampling pitch is the same as that of the rectangular sampling structure,
The horizontal resolution can be doubled to f□=1/PH.

However, general subjects are not necessarily achromatic, but generally have colors, so FIG.

Folding distortion occurs from the colored light carriers shown at all positions in FIGS. 10 and 12, and depending on the scene, it becomes very unsightly. For this reason, it is necessary to use an optical low-pass filter or the like to cut out harmful colored light carriers, which leads to a decrease in resolution.

For example, in the color solid-state imaging device with the offset sampling structure shown in FIGS. 11 and 12, (2/3P, ,
Since colored light carriers are generated at the position O), an optical low-pass filter that cuts frequency components of fH = 2/3 PH or more in the horizontal direction is required. Where horizontal resolution can be obtained up to pH, t o = 2 / 3 P of 273
Horizontal resolution can only be obtained up to s.

In order to resolve such problems, the applicant has separately proposed the third
We have proposed a color filter array as shown in the figure.

A solid-state image sensor using the color filter array shown in FIG.
Horizontal pixel pitch PH9 Vertical pixel pitch PV
, horizontal pixel offset amount P, has an offset sampling structure of /2, and has four types of color light transmission filters (color filters) Mg: magenta, green, cyan, and yellow.
, Gr, Cy, and Ye are arranged at positions corresponding to each pixel. A combination of these color filters is called a color filter array. As shown in Figure 3, this color filter array has an offset sampling structure in which each type of color filter has a pitch of 2P in the horizontal direction, a pitch of 2Pv in the vertical direction, and an offset amount of PH in the horizontal direction. There is.

FIG. 4 is a characteristic diagram showing the color light carriers in such a color filter array in terms of two-dimensional frequency ten planes (f, , fv).
2 Indicates the range of Pv.

In the same figure, the colored light carriers are (1
/P□, 0), (1/2P,, 1/4Pv), (0,
1/2Pv ), (1/Po , 1/2Pv), among which (0,0) and (1/PH, 1/2PV
) are carriers generated in response to achromatic light, and the others are carriers that disappear in response to achromatic light.

As is clear from the figure, in the horizontal direction (1/p, , o
) No color carriers are generated up to the position, and therefore frequency components up to f,=1/P can be obtained. That is, the color solid-state imaging device in this example is
Figure 11.

A horizontal resolution up to fH=1/PH, which is 1.5 times the horizontal resolution of the color solid-state imaging device shown in the sampling structure of FIG. 12, can be obtained.

Furthermore, the color light carrier closest to the origin is (
1/2 P H, 1/4 P v ), but this is a colored light carrier that has a sufficient distance from the origin and disappears against achromatic light, so it is
No significant aliasing distortion occurs for vertical frequency components. Especially when the color filters are complementary colors, the aliasing distortion is small.

As described above, the color solid-state imaging device using the color filter array shown in FIG. 3 has high resolution and less occurrence of moiré.

(Problems to be Solved by the Invention) An example of signal processing in such a color solid-state imaging device is shown in FIG.

In this case, the color signals are interpolated four complementary color signals Mg,
From Cy, Ye, Gr, R,
After converting into G and B primary color signals, they are processed. This color processing method is called a matrix method or a primary color separation method, and is described in the specification of Japanese Patent Application No. 63-281456 filed by the present applicant or "Simultaneous R in Image Mixing CCD Camera" by Nishimura et al.
GB Processing ``Television Society Techniques TEBS 89-9
ED 89-13 Feb 1989.

In the RGB conversion unit 510 in FIG. 5, the signal is converted into primary color signals RGB by matrix calculations such as the following.

At this time, consider a diagonal light and dark pattern as shown in FIG. It is assumed that thick lines represent bright areas and thin lines represent dark areas. This pattern is for the wave at point P shown in FIG.

As is clear from FIG. 13, for such a pattern of waves, the responses of Mg and cy are large, and the responses of Gr and Ye are small. Therefore, according to equation (1), the signal R,
When calculating G and B, there is a problem in that false colors occur even if the subject is achromatic.

Also, for the pattern shown in Figure 14, Mg and Ye
There is a similar problem in that the response of Gr and Cy is large, and the responses of Gr and Cy are small.

The present invention has been made to solve such problems, and an object of the present invention is to provide an image signal processing device that can realize high image quality with less occurrence of false colors.

(Means for Solving the Problems) In order to achieve the above object, the present invention configures an image signal processing device as shown in (1) and (2) below.

(1) an image pattern detection means for detecting an image pattern that is likely to cause false colors; a color signal processing means for performing color signal processing adapted to the image pattern; An image signal processing device comprising a selection means for selecting a signal processing means and causing color signal processing to be performed.

(2) The image signal processing device according to (1) above, wherein the color signal processing means performs achromatic processing.

[Effect]

According to the configurations (1) and (2) above, when an image pattern in which false colors are likely to occur is detected, color signal processing appropriate to the pattern is performed. According to the configuration (2) above, achromatic processing is performed as the color signal processing adapted to the above-mentioned pattern.

〔Example〕

The present invention will be explained in detail below with reference to Examples.

FIG. 1 is a block diagram of a "video camera signal processing device" which is an embodiment of the present invention.

A color filter array shown in FIG. 3 is attached to the sensor 101, and as shown in FIG. 3, two rows are read out in a zigzag manner in one horizontal scanning period (IH).

Therefore, all pixels are read out once per field. Now set the frequency of this read clock to f8
Then, the output of the sensor 101 becomes Mg→Cy−+Gr−+Ye for each clock. This signal is subjected to processing such as AGC (automatic gain control) in an analog processing section 102, and then sent to an A/D (analog-digital) converter 1.
A-D conversion is performed at 03. It is desirable that the A/D converter used has an accuracy of 10 bits or more.

This signal is input to an interpolation filter 104, then subjected to standard video signal processing such as gamma conversion and blanking in a brightness processing section 105, and D-A converted in a D/A (digital-to-analog) converter 106. Ru. Furthermore, a synchronization signal of a standard television signal is added to this by a periodic signal adding section (not shown).

On the other hand, the outputs of the A/D converter 103 are sent to color interpolation filter sections 107 to 110, respectively, for Mg and Gr.

After being separated into Cy and Ye, they are interpolated two-dimensionally.

This process is carried out, for example, in the patent application filed in 1983 by the present applicant.
-281456, and the explanation here will be omitted. RGB conversion units 112, 113, and 114 convert Mg and Gr using different matrix coefficients according to equation (1).
, Cr.

The Ye signal is converted into three primary color signals of R, G, and B.

As will be described later, the matrix selection section 111 determines whether false colors can be minimized by using the RGB conversion section according to the pattern of the subject, and outputs the result to the switch 115. The switch 115 connects the RGB converter 112,1
13, 114 output R, G, and B are matrix selector 1
Select based on the results of step 11.

Next, the difference between the RGB converters 112, 113, and 114 will be explained.

The RGB conversion unit 112 performs matrix calculation to minimize false colors on the bright and dark pattern shown in FIG. For the pattern shown in FIG. 13, assuming that Mg and cy are the bright parts and Gr and Ye are the dark parts, R is as follows according to the above equation (1). As mentioned in the above-mentioned reference, the red false color Rm for the pattern in FIG.
It is sufficient to multiply the term by +1 and the terms of Gr and Ye by -1, so the following is obtained. When Rm is O, the occurrence of false colors is minimized.

That is, a++Mg+a, Cy=a, 2Gr+a, 4Ye=-
(4) The same holds true for other colors, so A method for determining matrix coefficients that also optimizes color reproduction under the constraint condition (5) was proposed in the aforementioned patent application No. 63.
This is as shown in No.-281456. In this way, the determined matrix calculation power is executed by the RGB conversion unit 112, so that the false colors are minimized compared to the pattern of FIG. 13 as a result.

Similarly, for the pattern shown in FIG. 14, the conditions under which false colors are minimized are as follows. Since the matrix calculation that optimizes color reproduction under the constraint condition (6) is executed by the RGB converter 113, the output has minimal false colors compared to the pattern shown in FIG.

The RGB conversion unit 114 performs an operation to make the color signal achromatic when both the patterns shown in FIG. 13 and FIG. 14 overlap in the color matrix. As a means to make the color achromatic, for example, a++=a2t=as+ (1=1.2.3.4)・
...-(7) can be used. When both the patterns in FIGS. 13 and 14 overlap, this is the case at the edges of the object, and in such areas, there is little deterioration in image quality even if the color signal is made achromatic.

FIG. 2 shows the configuration of the matrix selection section 111. The input signal is a signal read out in a zigzag manner from the sensor 101, which is manually input by repeating Mg-+Cy-+Gr→Ye. Therefore, the human signal and the delay 201
The result of adding the delayed data by the adder 202 is Mg+Cy/Cy+Gr/Gr
+ Y e / Y e + M g is repeated. When the subtracter 205 takes the difference between this and the delay 203° 204 that is delayed by 2 data, the result is (Mg+Cy)-(Gr+Ye)/(Cy+
Gr)-(Ye+Mg)/(Gr+Y
e) −(Mg+C'I)/(Ye+Mg
)-(Cy+Gr) is repeated. If the component of the pattern in Figure 14 of the subject is large (Cy+Gr)
-(Ye+Mg) becomes larger, or the 13th
If the component of the pattern in the figure is large, (Mg+Cy)-(G
The absolute value of r+Ye) increases. Therefore, switch 2
If you switch the data one by one with 06, the upper side will have ±[(Mg+Cy) (Gr+Ye)] and the lower side will have ±
Since the outputs of [(Cy+Gr)-(Ye+Mg)] are separated, their absolute values are taken by absolute value detection units 207 and 208, and further compared with predetermined thresholds Thl and Th2 by comparators 209 and 210. Therefore, P is >T
1 in hl. <Thl is 0, Q is>Th2 and 1, <Th2
The logical outputs P and Q of 0 are obtained.

The judgment logic unit 211 determines the matrix to be used based on the logic shown in FIG.

Note that the RGB converter 114 does not convert the color signals into achromatic colors, but the color signal processor 116 converts the color difference signals R-Y, B-
It may be realized using so-called chroma killer processing to set Y to 0.

Further, when the horizontal address X and the vertical address Y of the pixel are manually input to the switch 115, and the outputs of P and Q are 0, M1 and M2 are alternately switched as shown in FIG. You may.

Further, the color filter of the sensor 101 is made of Mg.

Not limited to Cy, Gr, Ye, but also R, G,, G2. It may be a pure color like B, or any color as long as four colors are arranged in an offset structure.

In addition, in the example, there are two image patterns that are likely to cause false colors,
Color signal processing means adapted to the image pattern 112~
114, but the present invention is not limited thereto, and includes one or more image patterns in which false colors are likely to occur, and one or more color signal processing means adapted to the image patterns. It can be implemented in the form of

(Effects of the Invention) As described above, according to the present invention, high resolution and high image quality with less false color generation can be achieved even for patterns that are likely to cause false color generation.

[Brief explanation of the drawing]

Fig. 1 is a block diagram of an embodiment of the present invention, Fig. 2 is a block diagram of a matrix selection section of the embodiment, Fig. 3 is a diagram showing a color filter arrangement used in the embodiment, and Fig. 4 is a diagram showing a color filter arrangement used in the embodiment. Figure 3 is a characteristic diagram of colored light carriers in a color filter array, Figure 5 is a block diagram for explaining related technology, Figure 6 is an explanatory diagram of the operation of the navigation example, and Figure 7 is a color filter array. A diagram showing an example, FIG. 8 is a characteristic diagram of colored light carriers in the color filter arrangement of FIG. 7, FIG. 9 is a diagram showing an example of the color filter arrangement, and FIG. 10 is a color light carrier in the color filter arrangement of FIG. 9. 11 is a diagram showing an example of the arrangement of color filters, FIG. 12 is a characteristic diagram of colored light carriers in the color filter arrangement of FIG. 11, and FIGS. 13 and 14 are images in which false colors are likely to occur. FIG. 15, which is a diagram showing the pattern, is an explanatory diagram of the operation of the embodiment. 111-...Matrix selection section 112-114-'-RGB conversion section 115...Switch

Claims (2)

[Claims] (1) An image pattern detection means for detecting an image pattern in which false colors are likely to occur; a color signal processing means for performing color signal processing adapted to the image pattern; An image signal processing device comprising a selection means for selecting a processing means and causing color signal processing to be performed. (2) The image signal processing device according to claim 1, wherein the color signal processing means performs achromatic processing.