JPS6170865A - Image pickup device - Google Patents

Abstract

PURPOSE:To improve the sensitivity by providing a delay circuit delaying the 1st input signal for a prescribed time, a change detection circuit, and a noise separation circuit using a signal representing the image pickup state to separate noise thereby improving the deteriorated S/N. CONSTITUTION:A signal subject to photoelectric conversion from an image sensor is amplified to a prescribed level by an AGC15 and converted into a digital signal by an A/D converting section 11. The signal is converted into three primary color signals R, G, B at a color separation circuit 2 and converted into a luminance signal Y and two color difference signals R-Y and B-Y at a matrix circuit 3. The color difference signal is inputted to S/N improving circuits 13, 14, a difference signal between the said value and an output signal before one horizontal scanning period is separated into difference and noise components by a noise separation circuit using a signal representing the image pickup state, the noise component is outputted while being subtracted from the input signal ad outputted via an encoder 6 and a D/A converter 12.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of Industrial Application The present invention relates to an imaging device such as a television camera.

2. Description of the Related Art Structures of Conventional Examples and Their Problems In recent years, imaging devices such as television cameras are becoming more sensitive due to improvements in imaging devices. In particular, there is a strong demand for higher sensitivity, that is, lower noise, for ENG A and those used in places where sufficient illumination is not available, such as for home use.

A conventional imaging device will be described below. The first problem is a diagram showing the configuration of a conventional single-chip imaging device. 1 is an image sensor that converts an optical image into an electrical signal. The signal from the image sensor converted into an electric signal is converted into the three primary color signals of IIGB by the color separation circuit 2, and the luminance signal Y and the two color difference signals RY and B-Y are converted by the matrix circuit 3.
Convert to The two color difference signals have an S/N of 4 and 6, respectively.
The signal-to-noise ratio is increased by the Bunzen circuit, and the signal is converted into an NTS signal by the encoder 6 together with the luminance signal, and is output.

In single-chip color cameras, in order to achieve high sensitivity and high resolution, complementary color pigments as shown in FIG. 3 are generally used in mosaic color filters. W is golden light transmission, ye is red light (hereinafter R) and green light (hereinafter R)ay
is a color filter that transmits blue light (hereinafter referred to as B) and G,
Separation of the respective primary color lights R and H is performed as shown in the following equation.

R=W −ay + YlB −G
--------(1 B =
W -Ye + Cy -G ・Mountain・
・@)y: w + Ye + ay + G
・−・−(3) In this type of color separation method, compared to those that use primary color filters using RGB pigments, the brightness signal has a higher S/N ratio because the utilization rate of light increases. I get a signal, but R,! :Conversely, for the B primary color signal, the S/N becomes worse when the signal amount is lower. For example, if a primary color filter is used, a primary color signal equal to the saturation level of the image sensor can be obtained, but W.

Ye, c7. In the case of G, under the condition of equation (4), the R signal obtained according to equation (1) is assumed to be W=1 (saturation level) - lowercase letters indicate the level when separated into each of the three primary colors.

””+((r×g+b)−(g+b)+(r+g)−
g ) =4r (5
), the level of the primary color signal is lower than that of the primary color system where R=r, and an improvement in SIN is desired.

Therefore, even in the conventional single-chip color imaging device, the SI
Improvements are being made to N. Figure 2 shows the S/
A block diagram of the N improvement circuit is shown. 7 is the subtraction circuit, 8 is I
H (horizontal scanning period) delay circuit, 9 a constant circuit, and 10 a clip circuit. Take the difference between the input signal and the signal 1H ago. If this signal is Δ, Δ is the noise component and the signal difference between lines. A Δ signal with a level above a certain value is regarded as a signal difference between lines, a low level is regarded as noise, and the part with a high level of the Δ signal is clipped to separate the part that is considered to be a noise component. . The separated noise component is multiplied by a constant α 1 to obtain N', and this noise component N' is subtracted from the input signal to output a noise-improved signal.

When the same signal comes in the vertical direction and the noise is small,
The SIN improvement is 8 dB at α=0.6 and 1 at a=0.9.
This results in an improvement of 2 dB.

However, with this method, if the level of the Δ signal, which is the difference from the signal 1H before, exceeds the clip level, the S/N
Improvements are not made. Furthermore, if the level of the signal passing through the clip circuit is increased with emphasis on S/N improvement, the resolution of the color difference signal in the vertical direction decreases, causing color mixing in the vertical direction, which poses a problem.

As described above, the conventional configuration has the problem that there is no SIN improvement effect in areas with poor SIN where the effect of S/N improvement is expected to be high. This included the problem of a decrease in directional resolution.

Purpose of the Invention The present invention solves the above-mentioned conventional problems.It is possible to expect an improvement in the SIN even in areas where the SIN of the color difference signal is poor, and the imaging condition is good (even when the SIN is high, the resolution of the color difference signal is improved). It is an object of the present invention to provide an imaging device having an S/N improvement circuit with very little deterioration.

Structure of the Invention The present invention includes a delay circuit that delays an input signal for a certain period of time;
An imaging apparatus is provided with a change detection circuit that detects a change component, a circuit that separates a noise component from the change component using another signal corresponding to the input signal, and subtracts the separated noise component. , SIN of the input signal from the image sensor
The configuration is such that the SIN can be improved even when the SIN is poor.

DESCRIPTION OF EMBODIMENTS FIG. 4 shows a block diagram of an imaging apparatus in a first embodiment of the present invention. In Figure 4, 11 is Mu/D
12 is a D/MU converter, 13 and 14 are S/N improvement circuits, 15 is an automatic gain control circuit (hereinafter referred to as "JINGO"), and 20 is a level detection circuit.

The operation of the imaging apparatus of this embodiment configured as described above will be described below. A photoelectrically converted signal from the image sensor 1 is amplified to a predetermined level by the GC 15, and converted into a digital signal by the A/D converter 11.

The color filter of the image sensor is similar to that of the conventional example and is shown in FIG. Therefore, color separation is also possible (1) ~
This is carried out using the method of equation (3). The matrix circuit 3 converts the luminance signal into a luminance signal and two color difference signals, and converts the two color difference signals into S using a signal indicating the level of the luminance signal or a signal indicating the gain of the luminance signal circuit or the GO circuit. /N Make improvements. A detailed block diagram of the S/N improvement circuit is shown in FIG.

7 is a subtraction circuit, 8 is a 1H (1 horizontal scanning time) delay circuit,
9 is a constant circuit, 16 is a clip control circuit, 17 is a signal conversion circuit, 18 is a clip circuit, and 19 is a signal inverse conversion circuit.

From the input signal Sin (color difference signal in this example), subtract the noise Nn separated by a noise separation circuit described later,
Let the output signal be Son. The subscript n indicates the order of horizontal scanning. The change detection circuit takes the difference between the output signal 5on- from 1H before and the input signal Sil, and calculates the change component as Δ. shall be. Δ.

is the difference component of the signal between lines and the amount of noise contained in the input signal. The control input (CONT) is used to separate the difference component and noise component of the Δ0 signal.

IN). As a control input for the noise separation circuit, a signal indicating the level of a luminance signal or a signal indicating the gain of the human GO circuit is used. In other words, the control input signal used here is obtained by averaging the output of the imaging probe, by detecting the peak value and averaging it, or by mixing these and performing calculation processing such as a value between the average value and the peak value. Use signals.

The arithmetic processing is performed by the level detection circuit shown in FIG. 420. The signal obtained by processing the output of this image sensor is the input signal S.

The value can be approximately proportional to the S/N ratio of , and becomes a signal indicating the imaging state. Δ. The signal is converted by frequency or pattern, and the clip level is set to the S/N of the input signal.
Control according to Δ. This makes it possible to more reliably separate the noise contained in the signal from the signal component, thereby further improving the S/N ratio of the input signal. Assuming that the characteristics of the 17 signal converters are shown in equation (6), each signal component becomes as shown in FIG. 6 by each conversion.

Do-D3:Δ. Each value Ko-, to: Converted signals Do' to D3': Inversely transformed Δn Ko/ to 3': K after clipping. ~5 Figure 6a is,
Difference signal Δ between lines. This is the waveform of b is K. The waveforms c, d, and e shown by the components are respectively 4. K2. K,
This is the waveform shown by the component. The signal decomposed into each component is a signal corresponding to each frequency or turn component,
The signal is caused by noise dispersed in each component, and it is easy to distinguish between noise and signal no(ter/).However, even in this case, even though t=4Q, the signal at ~3 has a small level and is at the threshold as shown by the thin line. is regarded as fixed noise.

Similarly, in the case of t=62, the signal of ~5 has a low level and is considered to be noise. This occurs because the threshold value that separates the signal and noise is constant, and separation becomes possible by controlling the threshold value using a control input signal that corresponds to the S/N ratio of the present invention. In the case of the example shown in Figure 6, 3 is not clipped but inversely transformed into K, ~, so
It is regarded as noise and is subtracted from the input signal Sin. As a result, the rising and falling portions of the signal become dull.

In particular, the changing part between t=40 and t=62 is K. It will be indicated only by The clip level of 3 in Ko~ is determined by the amplitude of the noise of the signal for which S/N improvement is to be performed. In the present invention, this clip level is increased or decreased according to the S/H of the input signal, and in imaging conditions with a good S/N, the clip level is lowered and the edge components of the signal are separated. The signal level is increased by 11 to prevent edge components from being subtracted from the input signal, and the level is increased in imaging conditions with poor S/N to improve the S/N for large amplitude noise. For example, when considering an imaging state with good S/N, it becomes possible to detect edge portions at t=40 and t=62 by lowering the level to the level 11, as shown by the broken line in FIG.

As described above, the signal (S
/N or a signal indicating the gain of 5GC) to separate the noise component and change component contained in the signal, thereby performing noise separation suitable for the state of the imaged signal, It is possible to improve the S/N.

The noise separation described above is performed using 18 clip circuits and 16 clip control circuits, and the separated noise components are separated into 19 clip circuits and 16 clip control circuits.
The inverse conversion circuit converts the signal into a signal indicating the amplitude of the time series,
The input signal Si is multiplied by a certain number α and used as the noise component Nn.
Subtract from n and output.

As shown above, by improving the S/N, the encoder 6
The pNTsG is modulated by the pNTsG, converted into an analog signal by 12 D/mu converters, and output.

As described above, by separating the noise component and the change component using the signal indicating the imaging state, the threshold value that forms the boundary between the signal and the noise is effectively changed. In other words, it is possible to realize an imaging apparatus that lowers the threshold value when the imaging condition is good to reduce blur in the changing part, and increases the threshold value when the imaging condition is poor and has a sufficient effect of improving the S/N.

.

In this embodiment, an example was shown in which the signal of a 1H line difference is separated from the noise to improve the S/N, but it is not necessary to limit the signal to a 1H line difference signal; Of course, it is also possible to use field differences. It goes without saying that the color filter of the imaging device can also be used in addition to the filter shown in FIG.

Also Δ. The conversion matrix of the signal conversion unit and inverse conversion unit that converts a signal into a frequency or pattern is expressed by equation (6).

There is no need to limit the method to the following: it is sufficient that the signal can be converted into and inversely converted into the required components. Also, the imaging device is RG
There is no need to limit it to B processing or digital processing. Also, the output does not need to be limited to NTSC. Furthermore, the second signal (control input) used for noise component separation need not be limited to a signal indicating the luminance signal level or a signal indicating the gain of the luminance signal circuit, but rather a signal that can determine the imaging state or S/N of the nine signals. It's obvious that it's fine if that's the case.

Effects of the Invention The imaging device of the present invention has a delay circuit that delays the first input signal for a certain period of time, a change detection circuit, and a noise separation circuit that separates noise using a signal indicating the imaging state. It is possible to realize an imaging device including an S/N improvement circuit that has a large S/N ratio improvement effect even in areas with a poor S/N ratio and has a good imaging condition (the resolution does not deteriorate even in a good S/N ratio). That is, it is possible to realize a highly sensitive imaging device, which can be used in home television cameras and the like used in places where there is insufficient illumination light, and its practical effects are great.

[Brief explanation of drawings]

Fig. 1 is a block diagram of a conventional imaging device, Fig. 2 is a block diagram of an S/N improvement circuit in a conventional imaging device, Fig. 3 is a layout diagram of color filters in a conventional imaging device, and Fig. 4 is a block diagram of a conventional imaging device. Block diagram of the imaging device in the embodiment of the invention, No. 6
The figure is a block diagram of the S/N improvement circuit in the present invention δ (example), and FIG. 6 is a diagram showing waveforms of each part of the S/N improvement circuit. 1... Image sensor, 2... ...Color separation circuit, 3...Ma) 11nox 1M path, 6...
... Encoder, 11 ... A/D conversion section,
12...D/A converter, 13°14...
・S/N improvement circuit, 15...Music GC circuit, 16
... Clip control circuit, 17 ... Signal converter, 18 ... Clip circuit, 19 ...
...Signal inverse conversion circuit, 20...Level detection circuit. Name of agent: Patent attorney Toshio Nakao and 1 other person No. 1
Figure 2 Figure 3 Figure 5

Claims (2)

[Claims] (1) a delay circuit that delays a first input signal for a certain period of time; and a change detection circuit that separates a change component of the input signal;
a signal conversion circuit that decomposes the changing components into frequencies or patterns, and processing that combines the converted frequency-based or pattern-based signals with a second signal indicating the imaging state of the first signal. A noise separation circuit that separates a noise component included in the change component by a circuit and an inverse conversion circuit that converts a processed signal into a signal indicating a waveform, and a subtraction circuit that subtracts the separated noise. An imaging device is a feature. (2) The imaging device according to claim 1, wherein the first input signal is a color difference signal, and the second signal is a signal obtained by calculating the output of an image sensor or a signal indicating a gain of a signal processing circuit.