JPS60156176A - Correction system in data processing - Google Patents

Abstract

PURPOSE:To decrease luminance unevenness by applying D/A conversion to a data corresponding to a uniform pattern and changing a reference of an A/D converter so as to bring the data to a maximum data of the initial data when the data is applied again with A/D conversion. CONSTITUTION:A video signal of a reference object without any uneven luminance is formed into a data corresponding to a uniform pattern by a reference voltage source 18, A/D-converted by an A/D converter 6, a prescribed address is designated to a video RAM9 and stored. Then the maximum value of the data in the video RAM9 is obtained by a CPU10. Each data read from the video RAM9 is converted into an analog data by the 2nd D/A converter 8 and it is converted into a digital data again by the A/D converter 6, then the value in a correction memory 11 is changed so that the value is made identical to the said maximum value.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of the Invention The present invention relates to a correction method for correcting brightness unevenness in an image processing apparatus that obtains data by A/D converting a video signal.

(b) Prior art When a C0D (charge storage device) is used as a sensor in a general Nihiteo camera, endoscope, etc., there is non-uniformity in sensitivity between each element of the CCD due to illumination with something close to a point light source. Although it is a pure white image, as shown in FIG. 2, the relative sensitivity near the center P is small depending on the position, and the relative sensitivity decreases as the distance from P increases. Up to this point, there are inconveniences in data processing after A/D conversion, so the relative sensitivity (signal size')'fr
: Must be corrected.

Therefore, one method for compensating for screen uniformity was proposed at the ``Third Digital Image Processing Application Technology Forum'' (December 16, 1982) sponsored by the Japan Management Association.
Proceedings of ``Camera for two-dimensional image processing combining 1-order ycccD and mechanical scanning''.

12 to Step 4.

However, in the example using COD, white nonuniformity (WU) and nonuniformity (DU
), it is shown that the variation in WU and DU is measured and corrected, but it does not specifically show the means for achieving this purpose.

(C) Object of the Invention The present invention aims to eliminate the above-mentioned drawbacks, and particularly to eliminate uneven brightness during imaging with a camera such as an endoscope.

B) Structure of the Invention The present invention comprises an address counter to which a horizontal synchronization signal and a vertical synchronization signal are applied, a first switch to which a video signal is applied, and an A/D connected sequentially to the first switch.
a converter, a second switch, a first D/A converter, a correction memory, a video RAM connected to the output side of the A/D converter, and a second D/A converter connected to the first switch. the third connected to the output side of the converter.
a switch, the video RAM, an address counter,
a fourth switch connected to the correction memory and the CPU;
a fourth switch connected to the CPU and the video RAM; a fifth switch coupled between the A/D converter and the third switch; and an output side of the first D/A converter. , the fifth switch and a sixth switch connected to the CPU, and the CPU applies a control signal to the control terminals of the first to sixth switches to display data corresponding to a homogeneous screen. When the data is written to the video RAM, it is D/A converted, and when A/D conversion is performed again by the A/D comparator 2, the value obtained by the A/D conversion becomes the initial data. The reference voltage of the A/D converter is changed so as to be the same as the maximum value among them, the digital input value of the first D/A converter is stored in the corresponding address of the correction memory, and the value according to the digital input value is This configuration reduces uneven brightness by changing the reference voltage of the A/D converter.

(e) Embodiment The correction method in data processing of the present invention will be explained with reference to the drawings. FIG. 1 shows a system block diagram for explaining the method, and FIG. 2 shows a characteristic diagram for explaining the method.

In Figure 1, (1) is a horizontal synchronizing signal terminal, (2) is a vertical synchronizing signal terminal, (3) is a clock signal (CLK) terminal, (4) is a video signal terminal, (the side is an address counter, ) is an A/D converter, (71 (81 is each first
and 2nd] D/A: 177 Hal, (9J Javide, tRAM, (10) microprocessor (CP
(referred to as U), Suha correction memo IJ, (13 (+3)
(I plant ■αQαD are each electronically controlled bus (C
B) first, second, third, fourth, fifth, and sixth switches whose switching operations are performed by control signals from the CPU through the CPU, a reference voltage source, AB an address bus, and DB a data bus. show.

First, a case will be described in which correction data is written in the correction memory and its contents are obtained.

The first, second, third, and fourth switches (I blunt 3) α4) and α are set to the x, y, and x1x sides, respectively, and the fifth switch is turned on.

A horizontal synchronization signal and a vertical synchronization signal terminal are added to the address counter (marked), and a video signal is added to the first switch (12+), so that the video signal of the reference object with no shading, that is, with no uneven brightness, is supplied to the reference voltage source α. A/D converter (6) converts the data into A/D converter (6) as data corresponding to a homogeneous screen.
The data is converted into D and stored in the video RAM (9) with a predetermined address designation.

Next, the third, fourth and sixth switches α4) (c) and (
17) to X, Y, .
AM (Calculate the maximum value of the occupancy data.

Next, 1st, 2nd, 3rd, 4th and 6th switch: y
IZIU (14) (15) and H (17) Y respectively
Y, X, Y, Y, X (or Y) and the fifth switch (16) is set to the on state, the video
When each data read by M(iVc) is converted into analog data by the second D/A converter (8) and then converted into digital data by the A/D converter (6), the value is Change the value in the correction memory so that it is the same as the maximum value.
) and the address in the correction memo IJ (11) are made to match, and the data contents in the correction memory I are completed by performing the above processing for one screen.

On the other hand, a case where the above-mentioned corrected data contents are stored in the correction memory I will be explained. In this case, the first, second, third, fourth and sixth switches (121(13)(
14) (151 and h (17) t4, X, X, Y, x,
xic switch. The correction memory U is addressed for each predetermined location by an address counter, and the value on the data bus is converted into analog data by the first D/A converter (8), which is used as the reference voltage for A/D conversion of the video signal. Therefore, the data obtained by A/D conversion is corrected, and the corrected data is stored in the video RAM (9).

Next, regarding image processing, the third, fourth, and sixth switches αaαωαη are set to the Y, Y, and X sides, respectively, and the fifth switch (16) is in the off state and the CPU (IIQI
and video RAM (J9) are connected, and after the data is processed by the CPU, it is output from terminal 0.

As a result of the above operation, data corresponding to the VC is written into the video RAM (91) with a homogeneous screen (obtained as a relative sensitivity of 10 levels as shown by the broken line in FIG.
The reference voltage of the A/D converter (6) is changed so that the A/D converted value is always the same as the maximum value in the initial data when converting from A to D/converting again. Since the configuration is such that the correction memory uVc is stored in the correction memory uVc, it is possible to remove brightness unevenness obtained as a video signal due to an error in the photographing system, for example.

(F) Effects of the Invention According to the correction method in data processing of the present invention, uneven brightness in a video signal obtained by a camera such as an endoscope can be uniformly reduced in the corrected signal when generated as a real screen. The signal obtained from the screen is derived.

[Brief explanation of drawings]

FIG. 1 shows a correction method in data processing of the present invention, and FIG. 2 shows an explanatory characteristic diagram of the same method. Explanation of main drawing numbers (1)...Horizontal periodic signal terminal, (2)...Vertical synchronization signal terminal, (3)...Clock signal terminal, (4
)...video signal terminal, (phrase...address counter, (6)...to/D converter, (7)...first
D/8 converter, (8)...Second D/A converter, (9J...Video RAM, Wing...c p
U, c++...correction memory, tLa...first switch, u3)...second switch, 04)...
Third switch, Q51...Fourth switch, α
6)...Fifth switch, (I7)...Sixth switch.

Claims (1)

[Claims] (1) An address counter to which a horizontal synchronization signal, a vertical synchronization signal, and a clock signal are applied, a first switch to which a video signal is applied, an A/D converter sequentially connected to the first switch, and a second a switch, a first D/A converter and a correction memory, a video RAM connected to the output side of the A/D converter, and an output side of a second D/A converter connected to the first switch. a third switch connected to the video RAM, the address counter, the correction memory and the video RAM; a fourth switch coupled between the A/D converter and the third switch; a fifth switch and the first D/A
the output side of the converter, the fifth switch, and a sixth switch connected to the CPU, and the control terminals of the first, second, third, fourth, fifth, and sixth switches are controlled by the CPU. When a control signal is applied to the data corresponding to a homogeneous screen, data corresponding to a homogeneous screen is written to the video RAMK, D/A conversion is performed, and A/D conversion is performed again by the A/D converter.
The reference voltage of the A/D converter is changed so that the value obtained by the A/D conversion is the same as the maximum value in the initial data, and the digital input value of the second D/A converter is changed. Store it in the corresponding address of the correction memory,
A correction method in data processing, characterized in that brightness unevenness is reduced by changing a reference voltage of the A/D converter according to the digital input value.