JPS62204665A - Picture input deice - Google Patents

Abstract

PURPOSE:To attain high accuracy (high S/N) and high resolution (multi-bit) by providing a subtractor subtracting a background density data and a multiplier multiplying a reciprocal to an output signal of the subtractor. CONSTITUTION:A digital density signal PD from an A/D converter 7 is inputted to a subtractor 13 and subtracted by a background density data BGD from the 1st correction data memory 11. Thus, a signal PD 1 having a density being the subtraction of the density in reading only other parts of a read object face from the read input density of the read part of the read object face is outputted from the subtractor 13. Then the signal PD 1 is inputted to a multiplier 14, where the signal is multiplied with a data MULD from the 2nd correction data memory 12. Thus, a density signal PD 2 with high accuracy added with a multiple factor and subject to correction of most of density read error caused by uneven lighting of the read part is outputted from the multiplier 14.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Application of the Invention] The present invention relates to an improvement in an image input device that reads image information on a surface to be read, such as a drawing, and outputs it as a digital signal.

[Background of the invention]

In an image input device, in order to improve the S/N of an obtained image signal, it is necessary to suppress noise caused by uneven illumination of a reading portion of a surface to be read (for example, a drawing). Conventional techniques for this purpose can be classified into the following four types. (1) Use a special light source such as a linear lamp. (2) Install a light shielding plate or the like in the optical path. (3) Install multiple light sources and adjust their arrangement and amount of light emitted. (4) Correcting the influence of illumination unevenness by electrical signal processing of image signals.

The above-mentioned conventional technologies (1), (2), and (3) require special parts, making them difficult to obtain and reliable, and requiring extremely long adjustment times to obtain optimal illumination. There are problems in that it is difficult to control, resulting in high costs, and that it is not possible to cope with changes in light source characteristics over time, and it is not possible to maintain a sufficient S/N.

Therefore, the above-mentioned prior art (4) is used as a means to solve these problems, and this will be explained below with reference to the drawings. FIG. 4 is a block diagram of an image input device used in facsimile machines, etc., which uses a CCD line sensor as a photodetector. In FIG. Here, it is illuminated by a pair of light sources 2.degree. 2, and the image information of FIG. 1 is inputted to a photodetector 4 via a lens 3. Photodetector 4 is shown in drawing 1
An electrical signal having a magnitude corresponding to the density of the image information described in the analog amplifier 5. is output. A/D
The signal is converted into a digital density signal PD via a converter 7. This signal PD is compared with the output signal (threshold value data) THD of the threshold value memory 10 by the data comparator 8, and the results are sequentially binary coded and outputted to the subsequent device via the signal output circuit 9. Note that the light source 2, lens 3,
The components of the photodetector 4 and analog amplifier 5 are collectively referred to as a photodetection device 6.

Here, when a CCD line sensor is used as the photodetector 4, the photodetection cells C, -Cn arranged in a -column can be used to detect a certain range (this (referred to as the reading section) is photoelectrically converted at one time. Therefore, it is necessary to uniformly illuminate this reading portion, but in reality it is difficult to uniformly illuminate, and the S/N of the digital density signal PD decreases.

FIG. 5 schematically shows the value of the digital density signal PD corresponding to the output signal of the photodetection cell CI-C of the photodetector 4. This corresponds to the case where the main line QIy Q2+Q3 is input. Reading portion end (photodetection cell C).

and Cn side), the illumination is dark, so drawing 1 there
The density of the area where there is no line Ql+Q3 is increasing, and if a certain threshold is used to perform binary coding of 1 or 0 depending on the presence or absence of line information, Line Ql+
Q3 becomes thicker and tends to generate noise. Therefore, in the prior art shown in FIG. 4, the threshold data THD, which changes in accordance with the density of the ground part of the drawing (background density data BGDI or BOD5 in FIG. 5), is used as the comparison reference value. By storing threshold value data THD (two types of threshold data THDI and THD2 are illustrated in FIG. 5) in the preliminary threshold value memory 10 so as to provide it to the comparator 8, unevenness in illumination and drawing This suppresses noise caused by the color of the background.

However, although the conventional technology shown in FIG. The following problems arise when applying this method to an image input device such as an image processing device that requires inputting the density of an image as faithfully as possible. In other words, since image processing devices require high resolution, the number of signal bits handled is approximately 8 bits, which is more than 4 bits more than in the case of facsimile devices. 8 (comparing two 8-bit input signals to obtain an 8-bit corrected output signal) increases the scale of the hardware, and it becomes difficult to obtain the data (threshold value data THD) to be stored in the threshold value memory 10. Furthermore, when the brightness of the light source 2 changes over time, it is necessary to rewrite and correct the threshold data THD, but as described above, it is difficult to obtain the data THD, so even when rewriting and correcting, the initial Similar difficulties arise when writing data. Therefore, it is possible to reduce the number of signal bits, but this has a problem in that the noise suppression effect is limited and the S/N ratio is significantly reduced.

[Purpose of the invention]

The present invention was made to solve the above-mentioned problems, and it is possible to achieve high precision (high S/N) and high resolution (multi-bit) with a small-scale, therefore small and inexpensive hardware configuration. It is an object of the present invention to provide an image input device that can extremely easily perform both the initial adjustment when the device is completed and the correction adjustment after it has deteriorated over time in order to improve its accuracy, thereby saving labor.

[Summary of the invention]

The device of the present invention stores background density data obtained by reading the ground portion of the surface to be read and the reciprocal value of the value corresponding to this data in advance, and stores the digital density at the time of reading the surface to be read. The above object is achieved by subtracting the background density data from the signal and then multiplying it by the reciprocal value to correct density reading errors caused by uneven illumination in the reading area.

[Embodiments of the invention]

The present invention will be described in detail below with reference to the drawings, but first the principle will be described here.

The inventors reached the following conclusion from the analysis results of optical noise factors related to the image input device.

(1) Illumination unevenness in the reading area of the reading surface can be measured as background density data BGD by reading the ground part of the reading surface in advance, and this is subtracted from the digital density signal PD output by subsequent reading. This can be corrected by

(2) Digital density signal PD is background density data BG
The large portion of D (dark portion of illumination) is increased compared to the small portion of the same data BGD (bright portion of illumination), but it is possible that the increase rate corresponds one-to-one with the background density data BGD. , the effect of illumination unevenness can be almost completely corrected by multiplying the result of subtracting the background density data BGD from the digital density signal PD in (1) above by a ratio uniquely determined by the background density data BGD. be.

The above (1) and (2) will be explained in detail below with reference to FIG. If there is uneven illumination in the reading area of the surface to be read, even if the ground part of the surface to be read is read, the solid line BGD1 or the dotted line BG will be displayed depending on whether the color of the ground part is white or colored.
A background density data BGD curve as shown in D2 is obtained. The reading result of this ground part (background density data BGD) is stored in advance, and the part with actual image information (
Line 11II.

When the background a degree data BGD is subtracted from the digital density signal PD when inputting the area in which Q2+Q3 is drawn, ΔPDI is obtained as the digital density signal PD of the line ff1ltQ2+Q3.

ΔPD2. ΔPD3 is obtained, and the output signal P shown in FIG.
D1 is obtained. At this stage, line 1. Although the concentration of Q3 is larger than that of line Q2, the rate of increase APDIΔPD2' ΔPD3 ΔPD2 is ΔBGDI, ΔBGD2. Since it can be uniquely determined from ΔBGD3, by multiplying the result of the above-mentioned subtraction process (output signal PDI) by the reciprocal of the increase rate (MULD), the uniform density output signal PD2 as shown in FIG. 3 can be obtained. It will be done. If such subtraction and multiplication processing is sequentially performed in real time for each signal (one pixel) from one photodetection cell C, a series of image data in which illumination unevenness is corrected will be obtained.

FIG. 1 is a block diagram showing an embodiment of such an image input device according to the present invention. In FIG. 1, the same reference numerals as in FIG. 4 indicate the same or corresponding parts.

Additionally, 11 is a first correction data memory that stores the background density data BGD, and 12 is a second correction data memory that stores data MULD that is the reciprocal of the increase rate. The address input signals for both memories 11 and 12 are as follows:
Numbers 1 to n (photodetection cell addresses) of the photodetection cells C that are being read are given.

A subtracter 13 subtracts the output data (background density data) BGD of the first correction data memory 11 from the output signal (digital density signal) FD of the A/D converter 7, and the result is (
outputs the signal PCI). 14 is a multiplier, and subtractor 13
The output signal PCI from the second correction data memory 12 is multiplied by the output data MULD, and the result (signal PD
2) is output.

Next, the operation of the above-mentioned apparatus of the present invention will be explained. Since the operation is the same as that shown in FIG. 4 until the digital density signal PD is obtained, the operation after that will be explained here. That is, the digital concentration signal P from the A/D converter 7
D is input to the subtracter 13, and the first correction data memory 11
Based on the background density data BGD subtracted from the background density data BGD, the subtracter 13 outputs a density value obtained by subtracting the density when only the ground portion of the surface to be read is read from the read input density of the reading portion of the surface to be read. A signal PDI (see FIG. 2) is output. This signal PDI is then input to the multiplier 14 and multiplied by the data MULD from the second correction data memory 12. As a result, the multiplier 14 outputs a high-precision density signal PD2 in which most of the density reading errors caused by illumination unevenness in the reading area are corrected by taking into account the increase rate.
(See Figure 3) is output. The output signal PD2 is output to a subsequent image processing device (not shown) through the signal output circuit 9.

If the subsequent device is a facsimile device or the like and requires output data coded into a small number of steps, the output signal PD2 of the multiplier 12 is sent to the constant setter 15. Data comparator 16
The signal may be outputted via the encoding circuit 18 consisting of the signal output circuit 17 and the signal output circuit 17.

In this case, the constant setter 15 is a circuit that selects the encoding characteristic curve and supplies it to the data comparator 16.
6 encodes the signal PD2 according to the encoding characteristic curve.

Note that the memories 11 and 12 have already been written with multiple sets of data BGD and data MULD obtained by reading in advance the ground portions of multiple types of surfaces to be read that are scheduled to be used, and any one of these sets can be selected. RO that can output
It may be composed of M. Alternatively, several such ROMs may be prepared and one of them may be selectively used depending on the background color of the surface to be read. Furthermore, the memory 11°I2 is a readable/writable memory, for example, R.
It is configured with AM, reads the ground part of the surface to be read in advance each time a reading operation is performed, or at a certain period of time, and records the data BGD.
, MULD may be rewritten using a microcomputer (not shown) or the like.

Further, in the above embodiment, the data BGD and MULD are stored in separate memories 11 and 12, respectively.
It goes without saying that one memory may be used and the data BGD and MULD may be stored in different storage areas therein.

〔Effect of the invention〕

As described above, in the present invention, the background density data obtained by reading the ground part of the surface to be read and the reciprocal value of the value corresponding to this data are written in advance in the memory, and the After subtracting the background density data from the digital density signal of Compared to conventional devices that use comparators to correct
Therefore, it is possible to achieve high precision with a small and inexpensive hardware configuration.
High resolution can be measured. In addition, it is corrected using background density data obtained by reading the ground part of the surface to be read in advance and the reciprocal value of the value according to this data, and can be used for initial adjustment when the device is completed or correction adjustment after aging. Both are extremely simple.
It is easy to perform and saves labor. Especially memory
By configuring AM and rewriting the data BGD and MULD in a microcomputer or the like each time a reading operation is performed or every fixed period, the above-mentioned labor saving becomes more remarkable.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing one embodiment of the device of the present invention, and FIG.
The figure is a schematic diagram of the output signal of the subtracter in Figure 1, Figure 3 is a schematic diagram of the output signal of the multiplier, Figure 4 is a block diagram of the conventional device, and Figure 5 is the digital density in the image input device. It is a schematic diagram of a signal. 1... Drawing (reading target surface), 4... Photodetector, 7...
...A/D converter, 11.12...1st. Second correction data memory, 13... subtracter, 14... multiplier, B
GD: Background density data; MULD: Data that is the reciprocal of the value corresponding to the background density data.

Claims (1)

[Claims] 1. A photodetector that reads the image information of the surface to be read at one time in a certain range and converts it into an electrical image signal of a size corresponding to the density of the image information, and converts the signal from this photodetector into a digital signal. In an image input device comprising an A/D converter for converting the image into a digital image, background density data obtained by reading the ground portion of the surface to be read and an inverse value of a value corresponding to this background density data are written. a subtracter that subtracts the background density data from the output signal of the A/D converter when reading the surface to be read, and a subtracter that multiplies the output signal of the subtracter by the reciprocal value. An image input device comprising: a multiplier.