JPS5922476A - Ripple correcting method - Google Patents

Abstract

PURPOSE:To decrese the quantizing error due to the change in luminous amount, by setting a reference voltage of a quantizer based on the ripple amount of an exposure light source, in an original reader reading out repetitively an output signal of a photoelectric converting element. CONSTITUTION:A sensor 7 detecting the luminous amount of a lamp 3 is provided in a reader of an original where a reflecting light from a light source 3 is detected at photoelectric converting element 6 and taken as a picture signal. The light detected at the sensor 7 is integrated at an integrating circuit 8 and the ripple amount is detected at a ripple extracting circuit 9. The detected ripple amount is held at a sample-and-hold circuit 11, converted into a suitable level at a level converting circuit 12 and inputted to a reference voltage setting circuit 14. The reference voltage setting circuit 14 sets a reference voltage in response to the ripple amount and applied to an A/D converter 15 for picture signal quantization.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a ripple correction method for correcting image signal ripple caused by light intensity riffle of a light source for exposing an original in an original reading device using a photoelectric conversion element in a copying machine, facsimile, etc. The original is transferred to the original by a lamp, and the reflected light is converted into an electric number by an optical conversion element such as a CCD or a photodiode array through an optical system including mirrors and lenses. A recording device that forms a static latent image on a photoconductor or dielectric material using a fiber tube or a needle, and then develops this static latent image to create a recorded image is already known and has been put to practical use. . Furthermore, as a method for varying the output of a light-converting element such as a CCD, a method is known in which the charge accumulation time on the light-receiving surface of the light-converting element is changed according to the illuminance of the light-receiving surface. A light source such as a CCD used in an Ijt document reading device in such a recording device
The output of the conversion element takes a certain amount of storage time when using a fluorescent light source that is powered by an AC power source as a light source for h-light. It will fluctuate. 1, that is, in FIG. 1, the waveform a shown by the broken line is the ACC power supply of the commercial frequency (50- or 6o) 1□) that illuminates the fluorescent lamp. C is the transfer control mountain number of the CCD for document reading, and the transfer fljJ? The interval between the 1jll signals indicates the charge accumulation time. In the figure, since the area of the shaded portion represents the exposure amount, it is easy to understand that the level of the image output signal varies each time. Therefore, one possible method is to use a high frequency power source of about 20 KHz as an AC power source to turn on a fluorescent lamp as a light source for the chest, so that the amount of exposure within each charge accumulation time is made almost constant.

For example, if a high frequency IT source of 20 KH is used, the period of light change of the fluorescent lamp is as follows. In addition, when the number of pixels of the CCD is 2048, the video frequency is IMB, and the minimum charge accumulation time is 6, the minimum charge accumulation time is 2048 (ms), and 82 waves enter each time and the image output signal becomes a trap. The fluctuation in the pressure level will almost disappear. However, in reality, since the commercial frequency (50H or 60H) is applied to the high frequency power source of 20 volumes, there is a problem that the level of the III image output signal fluctuates.

The present invention has been made in view of the above points, and includes a device for repeatedly reading the output signal of a photoelectric conversion element.
, detects the light intensity of the light source for document fog light, extracts only the riffle amount from that value, holds this riffle lid for the entire scanning period, and sets the standard of the A//D converter based on this ripple amount. This paper proposes a riffle correction method that performs quantization without any riffle carvings when setting the m-river.

This is explained based on the drawings.

FIG. 2 schematically shows an example of the configuration of a document reading section of a photocopying machine having a movable document table 1ii' and a reading device. is exposed to light by a lamp 3 such as a fluorescent lamp, and the reflected light from the original 2 is incident on a photoelectric conversion element 6 via a mirror 4 and a lens 5, and is converted into a blur image signal.In this example, the lamp 3 Non-image area (area outside the image sounding area at one longitudinal end of the lamp)
A sensor 7 such as a fumetode is provided to detect the K1 light.

Here, the principle of riffle correction will be explained with reference to FIG. 2. The sensor 7 detects the amount of light from the lamp 3, and the output change 11JK of the photoelectric conversion element 6 is integrated for frequencies with no shadow change.

For example, if the number of COD pixels is 2048 and the video frequency is IMH, the minimum charge accumulation time will be 2048 ms as explained above, and if the period of the light intensity ripple is less than 2048 ms, output fluctuations will be a problem. It doesn't matter. In this case the frequency is approximately 5001. So 500H
, and the above is integrated as shown in Figure 3 (a).

If only the ripple is extracted from this signal, Figure 3 (b)
as shown by the solid line. CO as the photoelectron conversion element 6
When D is used, the charge accumulated in a certain charge accumulation time is transferred to the analog shift register by the COD transfer control symbol in Figure 3 (c), and is output as an image signal within the next charge accumulation time. .

Therefore, it is necessary to maintain the ripple waveform information in FIG. 3(b) at the level shown by the chain line in the same figure.In the present invention, the ripple waveform is held, and after level conversion, a preset constant voltage c is added to determine the reference m pressure (R) of the A/D converter as shown in Fig. 3).

Fig. 4 shows an example of implementing the riffle correction method according to the present invention. 9
is a complete extraction of only the ripple portion of the borrowing from the integration circuit 8.
6, 7 ゛ extraction circuit for 6, ]() is a full bold timing circuit that outputs a hold signal for one scanning time based on CCD transfer control I and others, 11 is output from a simple hold timing circuit 10. hold signal r
Sunfull hold circuit that holds the riffle value from C, 12 is the held Ritz/” * rj:?jr
13 is a level converter circuit for a suitable level r (not described later); 14 is a level conversion circuit; 14 is a level conversion circuit; 12 riffle wall and reference m pressure 13 are added to set the reference m pressure vR of the A/D converter, which will be described later, according to the riffle, and the standard for output 1 [i setting circuit, 1
5 is an A/D converter for quantizing the image signal from the photoelectric conversion element 6.

As shown in FIG. 3, the reference voltage of the A/D converter 15 is set to a large value where the level of the trap or ripple waveform is high. Further, settings are made in advance so that quantization errors do not occur due to the resolution of the A/D converter 15 and the level fluctuation of the image output signal from the photoelectric conversion element 6.

In the above embodiment, the ripple waveform is extracted from a signal obtained by integrating the signal from the sensor 7 over a frequency that is not affected by fluctuations in the image output signal, or the ripple waveform is extracted by integrating the signal from the sensor 7 every seven scans. It is also possible to extract ripple waveforms from signals.

In the above embodiment, the sensor 7 that detects the brightness of the lamp 3 is attached directly to the lamp 3, but the sensor detects the light amount of the lamp via an optical fiber to prevent shadows such as heat caused by the lamp. You can also do this.

As explained above, in the present invention, the change in the light intensity of the light source for illuminating the original is detected, the amount of riffle is extracted from this extraction signal, this amount of riffle is held for a lifetime of scanning5, and the amount of riffle is based on this amount of light. Since the reference vehicle pressure is set without quantizing the original reading image output signal, the image signal can be quantized regardless of whether the sandals are open or not. According to the present invention, a+
Since the i-image signal is not directly corrected, there is no signal deterioration due to correction.

In addition to the six types in which the reflected light from the original is not directly incident on the photoelectric conversion element, the present invention may be of a type in which the optical image of the original is incident through a fiber. Furthermore, the ripple that is the object of the present invention is caused by fluctuations in the commercial frequency.
In addition to IJ tunnels, ■ caused by fluctuations in source pressure, etc. 6
It also includes relatively short-term changes in light intensity.

[Brief explanation of the drawing]

FIG. 1 is a diagram illustrating the important aspects of the image output signal level of the optical It conversion element due to riffling. FIG. FIG. 4 is a block diagram of an embodiment of a device implementing the ripple correction method according to the present invention. 1...Original table, 2...Original, 3...-Ranff°, 4
...Mirror, 5...Lens, 6...Photoelectric conversion element, 7...Sensor, 8...Integrator circuit, 9...Ripple extraction circuit, 10...Sandal hold timing circuit, 11 ...Sample hold circuit, 12...Level conversion circuit, 13...Constant voltage generator, 14...Reference voltage setting circuit, 15...j%/D converter patent applicant Konishiroku Photo Industry Co., Ltd. Company - Agent Patent Attorney Hiroshi Suzuki Blade Edge 1 Figure 2

Claims (2)

[Claims] (1) Integrate the output of a light sensor that electrically detects the amount of light irradiated by the light source for document exposure, calculate the ripple amount based on the integrated output, hold the ripple amount for one main scan, and A ripple correction method characterized in that ripple correction is performed by changing the reference voltage of a j%/D converter without quantizing the original read image signal based on the amount. (2) The ripple correction method according to claim 1, wherein the output of the light amount sensor is integrated for each main scan of the original image.