JPH04310073A - Flair correcting circuit for video signal - Google Patents

Abstract

PURPOSE:To correct flair properly even when a highlight portion reaching a pre-knee point is in existence in an object. CONSTITUTION:A digital video signal S5 from an A/D converter 5 is fed to a subtractor circuit 61 as a main signal and fed to a low pass filter 62, which extracts a signal component corresponding to a high frequency component in the original analog video signal S4 from the main signal and from which a resulting digital video signal S62 is extracted. The signal S62 from the filter 62 is fed to an inverse conversion circuit 63. The inverse conversion circuit 63 has an input output characteristic complementary to a pre-knee characteristic of a pre-knee circuit 4. Thus, a video signal S63 of the same linearity as that of the video signal S2 is outputted from the inverse conversion circuit 63. The signal S63 is fed to an adder circuit 64, in which the signal S63 is added or integrated by one picture element each for each field period and its sum output S64 is fed to a microcomputer 65 for one field period each.

Description

[Detailed description of the invention]

0001

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a flare correction circuit for video signals.

0002

[Prior Art] When taking pictures with a video camera, part of the incident light from the subject is diffusely reflected inside the camera, etc.
The image of the subject may float toward white. This phenomenon is called flare, and this flare becomes more pronounced as the subject has higher brightness areas, and degrades the quality of the reproduced image.

[0003] Therefore, video cameras are generally provided with a flare correction circuit to correct the video signal.

FIG. 6 shows an example of a video camera from an image sensor to a flare correction circuit.
An output signal S1 is supplied to a sampling/holding circuit 2 to extract an effective video signal S2, and this signal S2 is supplied to a Pliny circuit 4 through a gain control circuit 3.

This Pliny circuit 4 has input/output characteristics as shown in FIG. 7, for example, and is configured such that when the level of the input video signal S2 exceeds the Pliny point, the gain for the output video signal S4 becomes small. . Therefore, the level of the high brightness portion of the video signal S4 is more compressed than that of the video signal S2.

[0006] Then, this video signal S4 is
The signal S4 is supplied to the converter 5, and one sample of the signal S4 is converted into a digital video signal S5 of, for example, 10 bits.
/D conversion, and this signal S5 is supplied to the flare correction circuit 6.

The correction circuit 6 is composed of a subtraction circuit 61 and an integration circuit 69 having a time constant of several field periods, and the video signal S5 from the converter 5 is supplied to the subtraction circuit 61 as a main signal. Integrating circuit 6
9, a DC component S69 (a digital signal corresponding to the DC component of the signal S4) is extracted, and this DC component S69 is supplied to the subtraction circuit 61.

In this case, this DC component S69 corresponds to the amount of floating of the subject in the white direction due to flare. Then, in the subtraction circuit 61, this DC component S69 is subtracted from the video signal S5, so the subtraction result S6
is the original video signal S5 from which the floating in the white direction due to flare has been removed. That is, the signal S6 is a video signal subjected to flare correction. This video signal S6 is then taken out to the terminal 7.

[0009]

[Problems to be Solved by the Invention] However, in the case of the flare correction circuit 6 as described above, if there is a high-brightness part in the subject, the amount of flare correction may be insufficient due to the following reasons I and II. There is.

[0010] That is, since the size of I flare corresponds to the brightness of the subject,
When the video signal S2 includes a DC component due to flare, the magnitude of this DC component corresponds to the level of the video signal S2. However, since the level of the high-brightness portion of the video signal S4 is compressed by the Pliny circuit 4 compared to the level of the original video signal S2, extracting the DC component from such a video signal S4 would not be possible. The level of the DC component is the same as the original video signal S2.
The level will be lower than the level of the DC component extracted from the source.

Therefore, even if the correction signal S69 is formed from the video signal S5 obtained by A/D converting the video signal S4 to perform flare correction, the amount of correction will be insufficient.

II If the object is a high-intensity white circle on a black background, for example, as shown in FIG. 8A, the video signal S4 has a waveform with varying levels as shown in FIG. 8B. Then, for such a signal S4,
If the allowable input level (dynamic range) of the A/D converter 5 is level V5 in FIG. 8B, the signal S4
is clipped at level V5, and only the shaded portion in FIG. 8B is A/D converted to digital video signal S5.

Therefore, even if the flare correction signal S69 is formed from such a signal S5, the portion of the signal S4 at level V5 or higher (the portion without diagonal lines) is not used to form the flare correction signal S69. Therefore, when such a flare correction signal S69 is used, the correction will be insufficient.

In this case, the Pliny circuit 4 is provided before the A/D converter 4, but its Pliny characteristic does not become flat even in high-brightness portions, so The level may exceed the allowable input level V5 of the converter 5.

[0015] Furthermore, the gain of the gain control circuit 3 is adjusted and the signal S4 is supplied to the A/D converter 5.
If the peak value of is below the allowable input level V5, then I
The problem in Section I does not arise.

However, if this is done, the resolution during A/D conversion in the A/D converter 5, that is, the range of analog levels per bit becomes wide, and the quantization error becomes large. In other words, the quantization of the visually noticeable gray level becomes rough due to the high brightness portion.

The present invention is intended to solve the problem of insufficient flare correction amount due to the I term among the reasons for the I term and II term.

[0018]

[Means for Solving the Problems] Therefore, in the present invention, the reference numerals of each part correspond to the embodiments described later. The video signal S4 from the circuit 4 is supplied to an inverse conversion circuit 63 whose input/output characteristics are complementary to that of the Pliny circuit 4, and an addition circuit 64 which adds the output signal S63 of this inverse conversion circuit 63 over a predetermined field period. , a circuit 65 which averages the addition output S64 of the adder circuit 64 over a predetermined field period and outputs a signal indicating the average luminance level, and a video signal S4 from the Pliny circuit 4, which calculates a signal indicating the average luminance level. and an arithmetic circuit 61 that outputs a video signal S6 subjected to flare correction.

[0019]

[Operation] Flare correction is performed on a video signal in which the level of high-brightness portions has been level-compressed by the Pliny circuit, but the flare correction signal for this purpose is a video signal whose level has been corrected using a characteristic complementary to the Pliny characteristic. Therefore, even if there is a high-brightness part in the subject, appropriate flare correction is performed.

[0020]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The example shown in FIG. 1 is a case in which the lack of flare correction amount due to the reasons in terms I and II is solved.

That is, in FIG. 1, the digital video signal S5 from the A/D converter 5 is supplied to the subtraction circuit 61 as a main signal, and the low-pass filter 6
A digital video signal S62 from which signal components corresponding to high frequency components in the original analog video signal S4 have been removed is extracted. This high frequency removal is for reducing the number of bits of the addition result in signal addition, which will be described later. Note that this high frequency removal results in ignoring a part of the subject, but since it is the details that are ignored, the effect is small and there is no problem.

The signal S6 from this filter 62
2 is supplied to the inverse conversion circuit 63. This inverse conversion circuit 6
3, as shown by the solid line in FIG. 2, has an input/output characteristic complementary to the Pliny characteristic (shown by the broken line) of the Pliny circuit 4. Therefore, from this inverse conversion circuit 63, a video signal S63 having the same linearity as the video signal S2 is output.
is output.

However, even in this video signal S63, the portion exceeding the allowable input level V5 of the converter 5 is clipped. That is, if the subject is, for example, FIG. 3A
As shown in Fig. 8A (this is the same as Fig. 8A), if the video signal S2 is a high-intensity white circle on a black background, then the video signal S2 is as shown in Fig. 3.
As shown in B, the waveform changes in level, and the shaded portion in FIG. 3B becomes the video signal S63.

Then, this signal S63 is supplied to an adder circuit 64, and for each field period, the signal S63 of that field period is sequentially added or integrated by one pixel (one sample), and the addition output S64 is 1 The data is supplied to the microcomputer 65 for each field period. Note that the addition output S64 indicates the area of the shaded portion in FIG. 3B.

At this time, the microcomputer 65 is supplied with the number of times N64 of pixel addition (the number of samples of the signal S63) in the field period.

[0026] Then, in the microcomputer 65, X64=
A division of S64/N64 is performed. In this case, the value S64 is the area of the shaded area in FIG. 3B, and the value N64 corresponds to its horizontal length (one field period), so the quotient X64 is the area of the shaded area in FIG. 3B over one field period. value, that is, the average brightness level.

Further, the signal S62 from the filter 62 is supplied to the comparator circuit 66, and a signal indicating the allowable input level V5 of the A/D converter 5 is supplied to the comparator circuit 66.
is supplied as a reference level to S6 from the comparator circuit 66.
When 2≧V5, a comparison output S66 which becomes “1” is taken out. Therefore, the level of analog video signal S4 is A
When the permissible input level V5 of the /D converter 5 is exceeded, the comparison output S66 becomes "1".

This comparison output S66 is supplied to the pixel counter 67 as a count enable signal, and the signal S62 from the filter 62 is supplied to the counter 67.
The number of pixels of the signal S62 during the period when S66="1" is counted for each field period, and this count output S67 is supplied to the microcomputer 65 for each field period.

Then, in the microcomputer 65, Y=Y1
+Y2 ‥‥‥ ■Y1 =
a・X64+b ‥‥‥ ■Y2 =c・S67+
d ‥‥‥ ■a to d: Operations indicated by constants are performed.

Although the details will be described later, the equation (2) is illustrated as shown in FIG. 4, where the vertical axis Y1 indicates the first correction amount of flare. Further, when formula (2) is illustrated, it becomes as shown in FIG. 5, where the vertical axis Y2 indicates the second correction amount of flare.

Furthermore, in the microcomputer 65, the value Y is
is determined for each field period, and an average value S65 is calculated for the value Y for the latest 8 field periods. This average value S65 is supplied as a flare correction signal to the subtraction circuit 61, and the video signal The average value S65 is subtracted from S5, and the subtraction result S
6 is taken out to terminal 7.

In this case, the component of equation (2) in the average value S65 is the average luminance level X64 for the latest 8 field periods, so this component is as shown in FIG.
This corresponds to a flare correction signal S69 obtained by integrating the video signal with a time constant of several field periods in the flare correction circuit 6 of FIG.

Although the video signal S4 is level-compressed in the Pliny circuit 4, the linearity of the video signal S63 supplied to the adder circuit 64 is returned to its original level by the inverse conversion circuit 63, so that the average value S65 is The component of equation (2) is equivalent to the flare correction signal formed from the pre-Pliny video signal S2.

Therefore, if the video signal S4 is below the allowable input level V5 of the A/D converter 5, appropriate flare correction is performed using this component. That is, I
The term is now resolved.

Furthermore, the component of equation (2) in the average value S65 is determined by the level of the A/D converter 5 in the video signal S4.
This corresponds to the period in which the allowable input level V5 is exceeded.
This component has a value corresponding to the portion of the signal S4 that is at level V5 or higher.

Therefore, ■ in this average value S65
The components of the equation mean that term II has been solved.

Therefore, even if the original video signal S4 has passed through the Pliny circuit 4, the video signal S6 taken out from the subtraction circuit 61 is
Even if there is a high-luminance portion exceeding the allowable input level V5 of the A/D converter 5 in the video signal 4, the video signal will be appropriately flare-corrected.

In addition, in a color video camera, the above-mentioned signal lines are provided for three channels for a red signal, a blue signal, and a green signal, and the output signal is, for example, NTSC.
Just encode it into a signal. Further, instead of counting the number of pixels of the signal S62, the counter 67 may count the number of clocks used to obtain the signal S62 and use the count as a measurement value of the clip period of the A/D converter 5.

[0039]

According to the present invention, when forming the flare correction signal S65 corresponding to the average luminance level from the video signal S4, the Pliny circuit 4
Even if the level of the high-brightness part of the subject is compressed, the linearity of the video signal S63 used to form the flare correction signal S65 is corrected by the inverse correction circuit 63, so even if there is a high-brightness part of the subject, Appropriate flare correction can be performed without the amount of flare correction being insufficient.

[0040]Furthermore, the processing of formulas
The microcomputer 65 carries out the process of averaging the correction signal S65.
Since this is executed by software, the amount of flare correction can be easily changed, for example, by adjusting the amount of flare correction while looking through the viewfinder.

[Brief explanation of drawings]

FIG. 1 is a system diagram of an example of the present invention.

FIG. 2 is a characteristic diagram for explaining the present invention.

FIG. 3 is a diagram for explaining the present invention.

FIG. 4 is a characteristic diagram for explaining the present invention.

FIG. 5 is a characteristic diagram for explaining the present invention.

FIG. 6 is a system diagram of a conventional example.

FIG. 7 is a characteristic diagram for explaining a conventional example.

FIG. 8 is a diagram for explaining a conventional example.

[Explanation of symbols]

1 CCD image sensor 2 Sampling/hold circuit 3 Gain control circuit 4 Prinny circuit 5 A/D converter 6 Flare correction circuit 61 Subtraction circuit 62 Low pass filter 63 Inverse conversion circuit 64 Adder circuit 65 Microcomputer 66 Comparison circuit 67 Pixel counter

Claims (1)

[Claims] 1. A Pliny circuit that compresses the level of a high-luminance portion of a video signal, an inverse conversion circuit to which a video signal from the Pliny circuit is supplied and whose input/output characteristics are complementary to that of the Pliny circuit, and an inverse conversion circuit for the inverse conversion. an adder circuit that adds output signals of the circuit over a predetermined field period; a circuit that averages the addition output of the adder circuit over a predetermined field period to output a signal indicating an average luminance level;
A video signal flare correction circuit comprising: an arithmetic circuit that calculates a signal indicating the average brightness level on the video signal from the Pliny circuit and outputs a flare-corrected video signal.