JPS59158173A - Generating method of multiple gradation picture data - Google Patents

Abstract

PURPOSE:To read gradations of intermediate density without increasing the number of gradations by providing two black-and-white threshold values with which black information having some density or higher is evaluated as black, and dividing density data between those threshold values into a specific set number of gradations. CONSTITUTION:There is no picture normally at edges of an original when a picture signal is read out of the original, so the 1st one line or several lines are sampled to regard the lowest level of a ground part as a comparison voltage B. When a gain setting signal goes up from low to high, and once a switch 3 outputs a signal (comparison voltage B) from a terminal (d), a voltage generating circuit 10 varies the voltage at the gain control terminal of a variable gain amplifier 8 until a comparator 9 outputs a coincidence signal. Then, the output level of the variable gain amplifier 8 vaires and a ready signal K is outputted. When the ready signal K is outputted, the gain setting signal I goes down to the low level and a discrimination start signal J goes up to the high level from the next line, so that a gradation discriminating circuit 4 starts discrimination.

Description

[Detailed Description of the Invention] (Field) The present invention relates to a multi-gradation image reading device, and in particular, to a reading level of the fabric of a read document, that is, a signal level to be determined as white, which depends on minute changes in the reflectance of the fabric. There is a white threshold (a degree) that determines white if it is above a predetermined value even if there is a fluctuation, and a black threshold (a degree) that determines that everything is black if the reflectance from the black part is below a predetermined value. concentration) and the threshold value (a degree)
The present invention relates to a method for creating multi-gradation image data for an image reading device that divides the interval into a predetermined number of gradations and reads it as a multi-gradation image signal.

(Prior Art and Background) The amount of light reflected from each pixel of an image written on a document is calculated by using an aggregate of photoelectric conversion elements such as a charge-coupled device (CCD) or a solar cell to convert the charge, voltage, and current into each pixel. When reading an image using a so-called multi-gradation image reading device that reads one image as an amount of electricity and represents it in multiple gradations. When reading with , minute fluctuations can be seen in the photoelectric conversion output with each reading. This is thought to be because the surface of the paper, which is the composition of the material of the manuscript, has irregularities and non-uniform textures of a size that cannot be ignored compared to the pixel points used for reading.
Such fluctuations are often meaningless or can be ignored for the purpose of reading an image, that is, as image information for reading.

Similarly, even if photoelectric conversion information G is obtained from the black part read from the original, density information above a certain density, that is, below a certain amount of reflected light, has no practical meaning as information for subsequent image data processing. It can often be ignored.

When attempting to create multi-tone density data by converting the gradations based on the photoelectrically converted raw data in such mechanical image reading.

Since gradations are assigned to densities that have little meaning as image density information, the number of gradations in areas that have more important meaning as density gradation information decreases, and because the data does not have much meaning, The processing circuit that follows (
For example, problems arise in that time is wasted in the compression circuit (for example, a compression circuit) or that the compression ratio is reduced in the compression circuit.

Note that this problem is not so noticeable with the conventionally used 16-gradation quantization gradation, but as the number of gradations is increased, and as the pixel point is made smaller and the reading resolution is increased, these effects become more pronounced. It will appear properly.

(Purpose and Features) In view of the above background, an object of the present invention is to read a document with small changes in reflectance in multiple gradations without affecting the white side of the gradation data, and with The black information has two thresholds for black and white, which are evaluated as black, and the density data between the thresholds is divided into a predetermined number of gradations.

The purpose of the present invention is to obtain multi-gradation data of relatively important parts necessary for subsequent image data processing.1 The features of the present invention are to achieve the above object by means of scanning an original with a pixel mesh, and of scanning a document from the pixel mesh. means for photoelectrically converting the amount of reflected light of the device; means for setting two black and white thresholds for evaluating the amount of reflected light photoelectrically converted by the means; and a predetermined number of thresholds for setting the two thresholds set by the means. In an image reading device having means for quantizing and dividing into multiple gradations, the white threshold is set to the smallest value of the signal obtained by performing multi-pixel spectroelectric conversion of the white of the material of the document. At the same time, the black threshold is set to a signal level corresponding to a predetermined density preset by the device, and white and black quantized gradation signals are uniformly assigned to signal levels above and below the set value. Furthermore, by amplifying the voltage at which the two thresholds for black and white open to a predetermined constant voltage, almost all of the gradation data is configured to correspond to the intermediate density originally required when performing feature determination, etc. That's what I did.

(Embodiment) Figures 1 and 2 are explanatory diagrams of an embodiment of the present invention, with Figure 1 explaining the main part configuration of the implementation circuit, and Figure 2 explaining the operation of each signal in Figure 1. It is something.

Incidentally, although not particularly shown in the drawings, the image on the document is separated into a mesh of unit pixels by a scanning system of an optical character reader or a facsimile machine that is well known and performs main and sub-scanning of the document, and the image of each pixel is calculated for each pixel. It is assumed that control means for controlling the procedure or operation of a part that photoelectrically converts the amount of reflected light to read out an analog image data signal and the following data conversion processing part is provided, and in this way, the image data read out from the original is It is assumed that an image signal is input to the circuit of FIG. 1 as an image input. Also, the signals G, H, and I in the figure.

It is assumed that K and J are control signals, respectively, and F and L are intermediate image signals, respectively, and the control signals are procedurally controlled by the above-mentioned control means.

In addition, if you want to supplement about one circuit component, 1゜2.9
is a comparator, 3.5.6 is a multiplexer inch, 4 is a gradation discrimination circuit, and 7 is a capacitor CD, which is a sample hold that stores the white level created based on the signal from the white background of image input A. A high impedance amplifier 8 forms the circuit, a variable gain amplifier 8 sets the signal level and analog amplitude of the image signal input to the gradation discrimination circuit 4 based on the value of the comparator 9, and 10 is the output of the comparator 9. Voltage generation (control) that generates a compensation voltage accordingly to control the gain setting of the variable gain amplifier.
The circuits 11 and 12 are voltage standards, 11 is the reference side of the comparator 2 and is used to set a threshold for determining the image signal as black, and 12 is used to set the full scale output level of the variable gain amplifier 8. It is something.

In the above configuration, the image signal A is input to one side of the comparator 1.2 and is also connected to the bottom side of the switch 3 and the bottom side of the switch 6. A comparison voltage B, which is the lowest level of the material to be read, is input to the other input of the comparator 1, and this voltage B is also connected to sides c and d of the switch 3. Further, a comparison voltage C predetermined by a voltage reference 11 as a voltage corresponding to black information of a certain density is input to the other input of the comparator 2, and this voltage C is also connected to the a side of the switch 3. The outputs of comparators 1 and 2 are the control input terminals e and f of switch 3.
It is connected to the. The control input terminal g of switch 3 is Hi.
When it becomes gh, the signal of the d terminal appears at the common terminal regardless of the states of the other control input terminals e and f. Also, when the control input terminal g is LOW, depending on the state of the control input terminals e and f, when e and f are both Low, the signal at the B terminal is generated, and when e is High and f is I-OW, the signal at the C terminal is generated.

When e is Low and r is High, it operates so that the signal at the a terminal appears at the common terminal.

The initialization signal G changes from Low to High at the same time as reading of the original starts, and is a signal that initializes the comparison voltage B and the voltage generation circuit 10. When G is LOW, the switch 5 is set to the h side, and when it is High, the switch 5 is set to the i side. At the same time as it falls to the side, the voltage generating circuit 1o generates a voltage such that the gain of the variable gain amplifier 8 becomes 1. The sampling signal H is a signal for sampling the lowest level when scanning the fabric, and when it is H or Low, it is set to the switch 6bj side.
When it happens, it falls to the side. DI is a diode for pulling, CD is a capacitor that holds the lowest level, and 7 is a buffer amplifier with a gain of 1 and high input impedance that works with the above DI and above CD to create a sample and hold circuit that stores the lowest level of the image signal. It is. Note that the output of the switch 3 is input to the variable gain amplifier 8,
It is assumed that the DC level of the variable gain amplifier 8 has been adjusted in advance so that it outputs OV when the comparison voltage C is input. The output of the variable gain amplifier 8 is input to the gradation discrimination circuit 4 and is also connected to the comparator 9 at the same time. The other input of the comparator 9 is the gradation discrimination circuit 4.
A voltage corresponding to the full-scale level of the pestle is input. The output of the comparator 9 is input to the voltage generation circuit 10, and the output of the voltage generation circuit 10 is input to the gain control terminal of the variable gain amplifier 8. The gain setting signal (2) is normally Low and becomes High when the gain is set, and is input to the voltage generating circuit 10 and the control input terminal g of the switch 3. J is a discrimination start signal, and when this is High, the gradation discrimination circuit 4 performs discrimination.

Considering the case of reading an image signal number from a document, normally there is no image at the edge of the document, so the first line or several lines are sampled and the lowest level in the texture area is taken as the comparison voltage B. First, when the original comes, the initialize signal G goes low.
The reset voltage 5 that goes from ow to High goes to i (111 and initializes the comparison voltage B (+V) L, and then goes Low after the voltage generation circuit 10 generates a voltage that makes the gain of the variable gain amplifier 8 1. Then switch 5 turns to h again.
Fall to the side. When the initialization signal becomes Low, the sampling signal H changes from Low to High, and the switch 6 falls to the side and begins sampling the lowest level of the fabric. When sampling is completed, the sampling signal H becomes LOW and the switch 6 falls to the j side again. At this time, the comparison voltage B is the lowest level voltage of the fabric. Next, the gain setting signal ■ changes from LOW to HIGH, the switch 3 outputs the signal at the d terminal (comparison voltage B), and at the same time the voltage generating circuit 10 controls the gain of the variable gain amplifier 8 until the comparator 9 outputs a matching signal. Change the voltage at the terminal. As a result, the output level of the variable gain amplifier 8 changes, and when the comparator 9 finally outputs a match signal, the voltage generating circuit 10 holds the output voltage at that time and outputs the ready signal K. When the ready signal K is output, the gain setting signal (2) becomes Low, and the discrimination start signal J becomes High from the secondary line, and the gradation discrimination circuit 4 starts discrimination. When the image signal A after the discrimination start signal J becomes High is higher than the comparison voltage C and lower than the comparison voltage B, the outputs of the comparators 1 and 2 are output, and both E are L o
Since the switch 3 is turned to the b side, the image signal A is input to the variable gain amplifier 8 as it is. Image signal A
When the voltage becomes higher than the comparison voltage B, the output of comparator 1 becomes H.
i g h, switch 3 falls to the C side and the comparison voltage B
(the lowest level of the dough) is input to the variable gain amplifier 8. Furthermore, when the image signal A becomes lower than the comparison voltage C, the output E of the comparator 2 becomes l (high), and the switch 3 may be configured to cause the voltage at point C (voltage at point C) to follow the input document, but generally speaking Since the black level can be considered to be almost constant regardless of the density of the original material, there is no practical problem even if it is fixed as in this embodiment.According to the embodiment of the present invention, minute changes in the reflectance of the original material can be Even if there is, all the fabric parts are replaced with a constant voltage (voltage at point B), so the gradation data of the fabric part will be a constant value even if it is read with multiple gradations.It is also constant for black information of a certain density or higher. Since it is replaced with a voltage (voltage at point C), the gradation data becomes a constant value.

Furthermore, since the voltage of the fabric part (voltage at point B) is sampled for each document and amplified to the full scale level of the gradation discrimination circuit, there is no problem even if the fabric density (reflectance) differs, and the gradation data is not wasted. It disappears.

As a result, all of the gradation data corresponds to the intermediate density of the document, which has the effect of increasing the number of gradations in the originally required density area.

(Effects) As explained above, according to the present invention, even if minute changes in reflectance are read, the level on the white side of the fabric is always a constant value, and black images with a certain density or higher, which may not be necessary depending on the purpose of use, can be used. All density information is converted to the black level, and the gradation of the intermediate density level sandwiched between these two levels is always amplified to the amplitude of the full scale range and input to the input side of the gradation discrimination circuit. Since almost all of the gradation data obtained corresponds to the intermediate density of the original, it is possible to read out the gradation in detail at the intermediate density, which is important for subsequent image processing, without increasing the number of gradations. Produces a distinctive effect.

[Brief explanation of the drawing]

1 and 2 are explanatory diagrams of one embodiment of the present invention, in which FIG. 1 shows the main part of the constituent circuit, and FIG. 2 is for explaining the operation. In the figure, 1' and 2.9 are comparators, 3.5.6 is a multiplexer 1' notch, 4 is a gradation discrimination circuit, 8 is a variable gain amplifier, 10 is a voltage generation circuit, and 7 is combined with a CD and DI. A high-impedance amplifier that forms a sample-and-hold circuit that holds the lowest value. 41

Claims (1)

[Scope of Claims] Means for sequentially line-scanning a document image by image, and means for photoelectrically converting the amount of reflected light from each pixel. means for setting two threshold values, black and white, for evaluating the amount of reflected light photoelectrically converted by said means; and quantization division into a predetermined number of multiple gradations between the two threshold values set by said means. In an image reading device that provides means for The white threshold is set to the smallest value of the signal obtained by performing multi-pixel spectroelectric conversion of the white material of the document, and the black threshold is set to a predetermined density preset by scanning. The signal level is set to a certain level, and white and black quantized gradation signals are uniformly assigned to signals above and below the set value, and the voltage between the two black and white thresholds is further amplified to a predetermined constant voltage. A method for creating multi-gradation image data.