JPH027772A - Shading correction circuit for picture reader - Google Patents

Abstract

PURPOSE:To attain proper correction over the entire original density by applying logarithmic transformation by a logarithmic amplifier and applying shading correction while using a logarithmic quantity as it is. CONSTITUTION:With a switch 4 closed, a reference white picture is read by an image sensor 1, subject to logarithmic transformation by a logarithmic amplifier 2, transformed into a digital logarithmic data by an A/D converter 3 and written in a RAM 5 as a reference data. With the switch 4 opened, the original picture is read and transformed into a logarithmic digital data, inputted to a comparison limit circuit 6, compared with the correction data stored in the RAM 5, when the data is less than the correcting data or below, no correction is applied and if the data is larger than the correction data, the data is limited so as to be equal to the correction data and the result is outputted to an addition/subtraction circuit 7. Thus, no large change in the density data is caused due to a read error of the read sensor and the shading correction is applied properly.

Description

[Detailed Description of the Invention] <Industrial Application Field> Facsimile, image scanner, digital duplex Tf,
This invention relates to improvements in shading correction circuits used in image reading devices such as rocks.

<Conventional technology> Like Uni? Image reading devices for RI heavy equipment include:
An image sensor such as a CCD (charge-coupled device) is used as a reading sensor, and the image signal obtained by scanning using an image sensor is
An error called shading occurs. This is because it is difficult to apply uniform sensitivity to the entire area of one line of the document, and it is difficult to manufacture the image sensor's receiving element to have the same sensitivity. - A shading correction circuit is installed in the shell to correct such errors. This shading correction requires a white reference signal obtained by reading a white image.

FIG. 2 is a block circuit diagram of a conventional image reading device disclosed in Japanese Patent Publication No. 63-52575. In the figure, 1 is an image sensor using a CCD;
20 is a switch, and 21 is an A/D that converts the analog image signal input from the image sensor into a digital signal.
Converter, 22 is A/D
A reciprocal converter that outputs a reciprocal value that is a value divided by the output of the converter 21, 23 an averaging circuit that averages a plurality of reciprocal values for one light receiving element of the image sensor, and 24 an averaging circuit. 23 output, I! 25 is a D/A conversion 2: ÷ which converts the digital data stored in the RAM 24 into analog data, and 26 is an analog converter that multiplies the original data by a correction value. A compensator 27 is an A/D converter that converts the haW analog data into digital data and outputs the digital data.

The operation of this circuit is as follows. First, switch 20
is kept closed, one document serving as a correction reference value is read, and the correction coefficient is recorded in the RAM 24 through processing by the A/D converter 21, reciprocal converter 22, and perspiration equalization circuit 23. The order is performed for all the received light i7-. Next, when actually reading the image, leave the switch 20 open and set the correction value corresponding to each light receiving element to R.
The signal is output from the AM 24, multiplied by the analog himimeter 26, and then converted into digital data by the A/D converter 27.

<Problem to be solved by the invention> By the way, the output of the CCD read from the sensor and the density of the original are not proportional, and human visual characteristics have a logarithmic relationship (
(See Figure 3). Therefore, a slight change in the amount of light read by the sensor 6 causes a large change in the density difference due to the nature of human vision. However, if the image data is Δ/D converted and then logarithmically converted to suit human visual characteristics, the relationship with the original M density will be 5 bits higher than the accuracy of A/D conversion.
As shown in Figure rjS4, there was a problem in that g, in areas with high document density, the document density changed greatly due to slight fluctuations in the reading sensor, making it impossible to reproduce smooth multi-gradation recording.

The present invention has focused on this point and has been made for the purpose of providing a shading correction circuit that can perform appropriate correction with a simple circuit configuration.

Means for Solving the Problems> In order to achieve the purpose stated above, the present invention provides an A/D
A logarithmic amplifier is installed in the front stage of the converter to linearize the output from the readout signal and the density, and the image data output from the A/D converter is converted to the own standard output from the A/D converter. If the image data is thicker than the reference value, it is limited to the reference value, and shading correction is performed by adding and subtracting using an adder/subtractor while keeping the logarithmic quantity.

Effect> Since shading correction is performed with the logarithm unchanged at 1 after logarithmic conversion using the monthly amplifier, the entire original density area is distorted by 1 r, and appropriate correction is performed, reducing the reading sensor's reading error by 3. ? You can keep the size small.

<Example> Next, one embodiment shown in FIG. 1 will be described.

In the figure, 1 is an image sensor for reading using COD, 2 is a logarithmic amplifier that logarithmically converts the image of the analog signal read by the image sensor 1, : (1 University of Tokyo, 1 number amplifier 2! iF Al1) converter, 4 is a switch, which converts analog data into digital data;
5 is a RAM in which correction values for aging correction can be written and read out at any time; 6 determines whether the read image data is larger or smaller than its own standard value; if it is larger, the image data is set as its own standard value. 7 is an addition/subtraction circuit for performing shading correction.

The circuit of the embodiment has the above configuration and operates as follows.

First, with the switch 4 closed, the image sensor 1 reads a white image as a reference, and the white reference analog data is logarithmically converted by the logarithmic amplifier 2, and then the A/D
The converter 3 converts analog logarithmic data into digital logarithmic data, which is written into the RAM 5 as reference data. Next, the original image is read with the switch 4 open, and similarly to the above, it is processed by the logarithmic amplifier 2 and the A/D converter 3 and converted into logarithmic digital data, which is input to the comparison and restriction circuit 6. The comparison and restriction circuit 6 compares the read image data and the correction data stored in the RAM 5, and if the image data is less than the correction data, outputs the image data as is, and if the image data is larger than the correction data, it outputs the image data as is. For example, the image data is limited to be equal to the correction data and output to the addition/subtraction circuit 7.

Since both the image data and the correction data are logarithmic quantities, the above processing can be performed without the need for Himi: By performing this on all light receiving elements, it is possible to obtain shading corrected data.
The relationship between image processing data and document density is such that each digital data is arranged on a straight line at equal intervals as shown in Figure 5, so large changes in density differences due to sensor reading errors are suppressed. It will be done.

<Effects of the Invention> As is clear from the embodiments described above, the present invention logarithmically transforms the output of the reading sensor before A/D converting it and compares the image data and correction data in a logarithmic state.
Shading correction is performed by adding and subtracting while keeping f1. Therefore, the density data does not change greatly depending on the reading of the reading sensor, and shading correction is performed properly. Also, since a multiplier is not required as in the conventional case, circuit deterioration is simplified. This makes it possible to reduce costs and further speed up the correction process.

[Brief explanation of the drawing]

1 is a block circuit diagram of a shading correction circuit of an image reading device according to an embodiment of the present invention, and FIG.
A block circuit diagram of a conventional shading correction circuit. FIG. 3 is a diagram showing the relationship between normalized CCD output and original density. FIG. 4 is a diagram showing the relationship between image processing data and original density in the conventional example. is a relationship diagram between image processing data and document density in Example t; 1...reading sensor, 2...logarithmic amplifier 3.
...A/D converter, 4...Switch/chi 5...RAM, 6...Comparison/limit circuit 7...Addition/subtraction circuit output Kn Sharp Corporation

Claims (1)

[Claims] 1. A reading sensor that reads the document and an A/ that converts the analog output from this reading sensor into a digital quantity.
A logarithmic amplifier is provided between the D converter and the output from the reading sensor and the density is linearized, and the image data output from the A/D converter is connected to the white reference output from the A/D converter as well. A shading correction for an image reading device characterized in that when image data is compared with a value, if it is larger than a reference value, the image data is limited to the reference value, and the shading correction is performed by adding and subtracting the logarithmic amount. circuit.