JPS62180662A - Picture reader - Google Patents

Abstract

PURPOSE:To eliminate a dark current component without lowering the dynamic range of an A/D converter with a simplified circuit constitution, by providing a means which makes a dark time output from the effective picture area of a solid-state image pickup element coincide virtually with the output value of a non-picture area. CONSTITUTION:When a picture data A obtained from a light receiving part at a CCD10 which is placed under a light shielding state when a bit of picture information is scanned, is larger than a set value B, an input on one side of an AND gate 14 goes to H because A>B, and at such a time when a count pulse arrives, the count pulse is inputted to the UP terminal of a counter 17, then being counted up. Therefore, by an output from an D/A converter 18, a subtracted quantity for the picture signal from the CCD10 becomes large at a subtractor 11, and the picture data A is reduced, then being coincided with a set value B. On the other hand, as for the picture data A obtained from the light receiving part of the CCD10, when it is A<B, the counter 17 is down- counted, and the picture data A is increased, then being coincided with the set value B. In this way, the dark current and an offset voltage can be eliminated.

Description

DETAILED DESCRIPTION OF THE INVENTION (Field of Industrial Application) The present invention relates to an image reading device that reads images using a CCD.

(Background of the Invention) Conventionally, image information such as a document is irradiated with a lamp, and the reflected light is guided through an optical system to a solid-state image sensor such as a CCD (charge-coupled device) or a photoelectric conversion element such as a photodiode array. 2. Description of the Related Art An image reading device is known that converts image information into an electrical image signal and extracts it. FIG. 6 is a block diagram showing an example of this type of original i reading device. The document 2 placed on the document table 1 is illuminated by the illumination lamp 3a of the optical scanning unit A4 while moving in the direction of the arrow.
The reflected light is reflected by the mirror 3b', further reflected by the mirrors 4a and 4b of the mirror unit 4 which moves in the direction of the arrow at a speed of % of the optical scanning unit 3, and is received by the CCD 6' via the lens 5. Accordingly, an image signal representing document information is obtained from the CCD 6.

However, in a CCD used in an image reading device, charges due to thermal excitation 1 from a silicon substrate flow into a potential well, charges thermally excited in a boning well region are accumulated, or a surface level It is known that dark current is generated due to various reasons such as the presence of current generated from the background, and that this is one of the causes of lowering the dynamic range of gradation.

This point will be explained in more detail. normal CCD
A constant offset voltage is superimposed on the image output voltage. Therefore, if the CCD is irradiated uniformly, a constant voltage corresponding to the amount of light received or a voltage subtracted from the offset voltage will be obtained from the light receiving section, or a constant offset voltage higher than that will be obtained from the light shielding sections at both ends. is obtained, and this offset voltage becomes the reference voltage in the dark state. However, CC
When the irradiation to D is stopped, the output voltage of the light receiving part does not become the offset voltage of the light shielding part, but is slightly lower than that (for example, several tens of nV), as if a small amount of light was being irradiated.
low) voltage. This is due to the dark current of the CCD, which increases in proportion to the scanning period T of the CCD and the temperature (which increases approximately twice as much as the temperature rises by 8° C.).

For this reason, there is a problem in that the dynamic range of gradation becomes low when image information is read by a CCD.

Therefore, conventionally, in order to reduce the dark current of CCD,
Is there a known method for cooling D using a cooling means such as a Beltier element?In addition to the Beltier element, its drive circuit,
A power supply, heat sink, etc. are required, which increases costs and risks condensation. Alternatively, memory (R
AM), the A/D conversion value of the dark current for each pixel at dark time is stored in its memory, and the dark current component of the corresponding pixel is removed by subtracting it from the read data when reading the image. However, the problem remains that the circuit configuration for controlling writing and reading into the memory and performing subtraction processing for each pixel is complicated, and the dynamic range of the A/D converter is narrowed by the same amount as the darkroom.

(Objects and Structure of the Invention) The present invention has been made in view of the points mentioned in -1- above, and aims to remove dark current components without reducing the dynamic range of the A/D converter using a simple circuit structure. In order to achieve this purpose, when the dark output value from the effective image area of the solid-state image sensor is larger than a predetermined value, the value to be subtracted from the output value of the solid-state image sensor is increased, and the dark output value is increased. When the value is smaller than a predetermined value, the value to be subtracted from the output value of the solid-state image sensor is reduced to make the dark output value from the effective image area almost match the output value of the image area. .

(Example) Hereinafter, the present invention will be explained based on the drawings.

FIG. 1 shows a block diagram of an embodiment of an image reading device according to the present invention, in which 10 is a CCD that photoelectrically converts optical image information and outputs it as an electric image signal, and 11 is a CCD.
10 is a subtracter having the function of amplifying the output image signal; 12 is an A/D converter that converts the output of the subtracter 11 into a digital signal (for example, 8 bits); and 13 is the output from the A/D converter 12. Digital image data A is compared with a preset value B (for example, 8 pips 1-), and if A>B, it is sent to terminal Q, if A=B, it is sent to terminal Q2, and if A<B, it is sent to terminal Q3. Comparators that each output "H" and otherwise output "L"; 14, 15, and 16 are ant gates; 17 is an uptown counter; 18 is a D/A converter for the count value of the uptown counter 17. D
/A converter. It should be noted that the ant gate 15 can open the gate and accept count pulses at times other than A-B.

Next, the circuit operation will be explained.

The circuit operation is based on image data A output from the A/D converter 12.
The output of the D/A converter 18 is feedback-controlled so that the set value B matches the set value B, and the set value B is now set to the black level rl 1111111J and the own level roooo000.
0J, for example rllllllllOJ, and the count pulse of the up/down counter 17 is set to
If the pixel corresponding to the light receiving section D is outputted, the dark current level can be fixed at rllllllllOJ.

Now, to explain the setting value B, in order to use the full dynamic range of the A/D converter, B is the minimum input of the A/D converter (for 8 bits, black, OO1, white, F
F) Or, it is possible to set it to the maximum input (black or FF, white or OO when there are 8 hits). However, the feedback system does not work properly with this $. This is because, assuming the setting value B is the maximum input, B=FF or 8
In the case of bits, FF is the maximum value and there is no larger value. Therefore, since there is only a value smaller than the set value B=FF, no feedback can be applied. (Feedback here means to increase G when it is smaller than the set value, and to decrease it when it is larger.) Therefore, set the set value to (minimum value + 1) or (maximum value - 1). It can be seen that the dynamic range of the A/D converter can be expanded the most. For example, in the example above, FF-1=
FE or its value. If the set value B is FE, then A is less than or equal to FC, so A<B, and A is FF, so A>B. A
Of course, A=B in FE.

That is, the CC is in the t-light state when scanning image information.
Image data A or setting value B obtained from the 1st pattern part of D10
If it is larger than A>B, the ant gate 14
One of the inputs becomes H, and when a count pulse arrives at this time, the count pulse is input to the UP terminal of the counter 17 and is counted up. Therefore, by the output from the D/A converter 18, in the subtracter 11, CC
The amount of subtraction for the image signal from D10 increases, and image data A decreases to match set value B.

On the other hand, if the image data A obtained from the light receiving section of the CCD 10 is A<B, the counter 17 is counted down and the image data A increases to match the set value B. This feedback operation is repeated every time a count pulse arrives, that is, every time image data A arrives from the shielded light receiving section, and continues until A=B.

In this case, if a count pulse was input every time the light receiving section of the CCD 10 scans, it would take time for A=B to be satisfied. Therefore, if the scanning pulses of the plurality of light receiving sections of the CCD 1O which are shielded from light are applied as count pulses to the ant gate 15, it is possible to set A=B by scanning one line of the light receiving section. Alternatively, if the count pulse is set at a high speed so that a plurality of count pulses are inputted to a light receiving section that is shielded from light, the time required to establish A=B can be further shortened.

When A=B, the shielded light receiving portion, that is, the black level portion (dark time output) is obtained as a constant value (setting value B). First, as long as the subtraction voltage (this voltage becomes the offset removal voltage) is applied to the subtracter 11 under the condition of , the black level is fixed. In other words, dark current and offset voltage are removed. During document reading, count pulses from 1 to 1 are not input as a matter of course, and the output of the D/A converter 18 is a constant subtraction type □
It is fixed by pressure, making it possible to fix the black level completely.

What is important here is that the count pulse is applied during non-reading mode, but it is made to correspond to the position of the CCD light-receiving pixel. It is also possible to remove offset voltage drift due to changes or changes over time. The color pulse may be applied to each specific light-receiving part pixel once per horizontal scanning period, or may be applied to a plurality of pixels.

Note that in order to completely remove dark current, even the slightest amount of light must not enter the light receiving section of the CCD in the non-reading mode. Normally, the optical scanning unit 3 (see Fig. 6) of the image reading device is in a reference position (called the home position) retracted inward from the document table l when not reading, so the CCD does not read. Although it is designed to prevent light from entering, in order to be even more reliable, a shutter may be raised in front of the lens 5 and the shutter may be closed when not reading.

: FIG. 2 shows a block diagram of another embodiment of the image reading device according to the present invention. □ ・Generally, we focused on the fact that in CCDs, a current proportional to the light obtained at the light receiving section is taken out through separate paths for even and odd pixels, and in this embodiment, black is drawn separately for even and odd pixels. It is designed to maintain the level and synthesize at the end.

The circuit configuration is such that the image signals of even-numbered pixels and the image signals of odd-numbered pixels outputted from the CGD 10 are separately feedback-controlled by feedback circuits 20 and 30 (both are shown enclosed by broken lines). The feedback-controlled image data is transferred to the timing control circuit 40.
The data is synthesized by a data selector 50 based on a select signal from the data selector 50. The circuit configuration and circuit operation of the feedback circuits 20 and 30 are exactly the same as the embodiment shown in FIG. The feedback circuit 30 will be omitted and the explanation of the circuit operation will be omitted.

FIG. 3 shows the signals output from the timing control circuit 40, and the even pixel count pulse and the odd pixel count pulse applied to the AND gates of the feedback circuits 20 and 30 (denoted as 25 for the feedback circuit 20), respectively. The count pulse is shifted by one shift clock pulse of the CCD I O, and the select signal alternately selects even-numbered pixels and odd-numbered pixels.

With such a circuit configuration, the black level is maintained separately for even-numbered pixels and odd-numbered pixels, thereby stabilizing the removal of dark current components and also eliminating variations due to differences in temperature characteristics.

FIG. 4 shows a block diagram of still another embodiment of the image reading device according to the present invention.

In this embodiment, in order to make the output signal of the subtracter 1 constant when scanning the light-shielded light receiving section, and instead of using a digital comparator, the reference voltage V RF:F power supply 61, analog comparator 62 and Reference voltage vRI! Using a power supply 63 of F + △V and an analog comparator 64, two analogues 65 and 66, an uptown counter 67, and a D/A converter 68 similar to the embodiment shown in FIG. There is. The count pulse is given at a timing corresponding to the light receiving section in a state where no light is incident on the CCD.

The reference power supply vRF:F or the black level will be given. This V RF:F may be any value that depends on the input level at the subsequent stage (after the density information in FIG. 4).

Now, let the output of the subtractor 11 be v7, and the outputs of the comparators 62 and 64 be V, , V, respectively, then V Rpr <
When V z < V REF +△V, is vx, vY °
It becomes "L" and the feedback loop is stabilized. That is, the system is stabilized when the output signal (7) of the subtracter 11 when scanning the light-shielding portion satisfies the condition of the above equation. The black level at this time is fixed by the memory function of the up/down counter 67, and the influence of dark current and offset current can be removed.

Note that this embodiment can also be applied to a method in which even-numbered pixels and odd-numbered pixels are extracted separately, similar to the embodiments described above.

However, in the embodiments shown in FIGS. 1 and 2, it is conceivable that the dark current component fluctuates due to CCD noise (the amount of fluctuation is several LSB or less). Therefore, in order to correct this fluctuation, it may be possible to average the image data including the A/D-converted dark current component.By such averaging, the random noise can be reduced to 1/'P-f,](l
It is possible to keep it down to Z 100 a. FIG. 5 shows an example of a circuit for correcting such fluctuations in the dark current component, which consists of an adder 70 and a latch circuit 71 that holds the addition result in accordance with the timing of the count pulse, and the circuit that is output from the A/D converter 12. The image data A (dark current + offset voltage) is sent to the adder 70.
The averaged value is added to the value held in the latch circuit 71. This addition result is 1 from the image data.
Either increase the number of bits, ignore the least significant bit (LSB) and give the operation result of the upper bit and carry, or the addition value x%.

The latch circuit 71 rewrites the previous value with a new calculation result at the timing of the count pulse, and provides the held value to the digital comparator 13.

This correction circuit is connected to the digital comparator 13 (Figs. 1 and 2).
(see figure), it is possible to correct fluctuations in the dark current component due to noise.

(Effects of the Invention) According to the present invention, the dark current component of the CCD becomes zero through feedback control, so that the dynamic range of gradations when image information is read by the CCD can be expanded.

[Brief explanation of drawings]

FIG. 1 is a block diagram of one embodiment of the image reading device according to the present invention, FIG. 2 is a block diagram of another embodiment of the image reading device according to the present invention, and FIG. 3 is a block diagram of an embodiment of the image reading device according to the present invention. FIG. 4 is a block diagram of still another embodiment of the image reading device according to the present invention;
FIG. 5 is an example of a correction circuit that corrects fluctuations in dark current components due to CCD noise, and FIG. 6 is a transport route diagram of an image reading device using a CCD.

Claims (3)

[Claims] (1) An image reading device that reads images using a solid-state image sensor, characterized in that a matching means is provided for matching a dark output value from an effective image area of the solid-state image sensor to a preset value. image reading device. (2) The image reading device according to claim 1, wherein the matching means is a feedback circuit. (3) The set value is less than or equal to (maximum value - 1) of the A/D converter that digitizes image data, or (
The image reading device according to claim 1, wherein the image reading device has a minimum value +1) or more.