JPH06245065A - Picture reader - Google Patents

Abstract

(57) [Abstract] [Purpose] Compensate for erroneous correction that occurs during shading correction, and suppress image quality deterioration during image binarization. [Structure] In front of a shading corrector 15, an interpolation filter 13 having dot levels before and after a white reference signal as an interpolation value, and whether the level of the current dot is apart from the interpolation value by a certain value or more is determined. And a comparator 14 for selectively outputting the reference level. [Effect] The erroneous detection of the white reference signal due to dust or dirt can be compensated by the interpolation data, and the shading correction can always be normally performed.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image reading apparatus for reading an original image by photoelectric conversion and correcting shading to obtain an electric image signal.

[0002]

2. Description of the Related Art FIG. 2 shows the structure of a conventional image reading apparatus. In FIG. 1, 1 is a CCD (charge coupled device) image reading unit, and 2 is an A / D
A converter (analog / digital converter), 3 is a shading correction device, 4 is an MTF (modulation transfer function) correction device, 5 is a binarization processing unit, and 6 is a dither matrix.

Next, the operation of the above conventional example will be described. The image data read by the CCD image reading unit 1 is quantized into a digital signal by the A / D converter 2. In order to compensate for the shading distortion that occurs during this reading, white reference data is read in advance and stored in a memory immediately before image reading, and shading correction is performed by the shading corrector 3 based on this reference data. After that, in order to improve the sharpness of the image as an image quality improving process for this digital image data, the MTF corrector 4 performs M
Perform TF correction. The corrected data is converted into monochrome binary data by the binarization processing unit 5 using the dither matrix 6. There are various binarization methods for tone reproduction of a grayscale image, but here, the dither method is used to represent halftones. In the dither method, various different thresholds are set in advance, and this is compared with the density level of the input image to output one of the binary levels of black and white. As the processing, the image data subjected to the MTF correction processing is compared with the data read by the dither matrix 6 in the binarization processing unit 5 to convert it into binary data. In this way, a sharp image signal whose shading has been corrected can be obtained.

[0004]

However, in such a conventional configuration, dust and dirt adhering to the reference plate are mixed in the data of the white reference plate read in advance of the original image for shading correction. In this case, the reference reading level is lowered by that portion, and the shading correction of the original image is performed with this level as the reference. As a result, when a document including a large number of halftone levels is read, it may be higher than the reference dip level, and when binarizing the gradation expression, it is processed as a white line in the sub-scanning direction. Therefore, there is a problem that the image quality is deteriorated.

The present invention solves such a conventional problem, compensates for the drop of the reference signal due to the adhesion of dust or dirt on the white reference plate, eliminates the image deterioration due to the shading correction of the image, and also in the binarization process. An object is to provide an image reading device capable of obtaining a good image.

[0006]

In order to achieve the above object, the present invention provides an interpolation filter which uses the levels of dots before and after a white reference signal as a correction value as a preprocessing for reading the white reference signal and writing it in a memory. By comparing the interpolation value with the level of the current dot, if there is a certain level or more, it is judged that there is a deposit on the white reference plate, and the interpolation value level is selected as the reference level. It is designed to compensate for false positives.

[0007]

According to the present invention, with the above configuration, it is possible to compensate for the inappropriate correction in the shading correction and realize the binary gradation reproducibility without degrading the image quality.

[0008]

Embodiments of the present invention will now be described with reference to the drawings.
explain about. FIG. 1 is an image in one embodiment of the present invention.
1 illustrates a configuration of a reading device. In FIG.
Reference numeral 11 denotes an electric signal obtained by reading an original image and photoelectrically converting it.
CCD image reading unit for processing, 12 is an analog signal
A / D converter to convert to digital signal, 13 read
Interpolation using dot levels before and after the white reference plate
Filter, 14 is the interpolation value of the interpolation filter 13 and the current dot
The levels of and compare if they are more than a certain value apart
A comparator for judging and selecting and outputting a reference level, 15 is a comparison
CCD image reading based on the reference level output from the instrument 14.
Shading correction of the image read by the sampling section 11
The shading corrector 16 to increase the sharpness of the image quality.
MTF compensator for correction, 17 is the corrected data
Binary processing unit for obtaining binary data from the
Dither used to reproduce halftones during the conversion process
It is a matrix. The correction filter 13 has four registers.
Data 19, 20, 21, 22 and four multipliers 23, 2
Consists of 4, 26, 26 and one adder 27
There is.

Next, the operation of the above embodiment will be described. The image data read by the CCD image reading unit 11 is quantized into a digital signal by the A / D converter 12. Next, in order to compensate for shading distortion that occurs at the time of reading, an appropriate reference level is set by the interpolation filter 13 and the comparator 14 based on the white reference data read in advance immediately before image reading, and the shading corrector is set based on this reference level. The reading characteristic is corrected by 15. After that, as the image quality improving process of the digital image data, the MTF corrector 16 increases the sharpness of the image by making the edge part sharp by using the edge extraction filter part as a parameter. This MTF processed data is 2
The binarization processing unit 17 uses the dither matrix 18 to perform monochrome 2
Converted to value data.

Next, the processing of the interpolation filter 13, which is a preprocessing for shading correction of image data, will be described in detail. Prior to reading the original image, an interpolation filter using the dot levels before and after as the interpolating value as a preprocessing for writing the white reference signal output from the CCD image reading unit 11 when the reflected light from the black and white reference plate is received into the memory. 1
Interpolation data is obtained by 3.

For example, if the data of two dots before and after is used as the filter component, the interpolation data of the current dot can be obtained by the following equation. F '(i) = g (i-2) * F (i-2) + g (i-1) * F (i-1) + g (i + 1) * F (i + 1) + g (i + 2) * F (i + 2) (Where F (i ± k) is the signal level ± k dots away from the current dot, g (i ± k) is the filter coefficient, and F ′ is
(I is the interpolated signal level) The difference between the value derived by the above equation and the signal level F '(i) of the current dot is obtained and compared by the comparator 14 as in the following equation. A> F '(i) -F (i) (where A is the allowable level difference)

When the tolerance value A of the level difference is larger than the difference between the interpolation data of the current dot and the current dot, it is determined that dust or dirt is attached to the white reference plate, and the interpolation data is set as the reference level of the current dot. , And write it to the memory as a new reference level for that dot. Based on the data written in this memory, shading correction of the document read image is performed.

As a result, when shading correction is performed, when the reference signal from the white reference plate, which is an element of shading correction, is written in the memory, level fluctuations due to dust and dirt adhering to the white reference plate are caused by normal signals before and after. It is possible to compensate for this, shading correction can be appropriately performed, and binarized gradation reproducibility can be realized without degrading image quality.

Further, by increasing the number of parameters before and after the interpolation filter component, it is possible to obtain an interpolation value with improved accuracy.

[0015]

As is apparent from the above-described embodiment, the present invention has an interpolation filter having interpolated dot levels before and after the shading corrector, and a current dot level higher than a certain value with respect to the interpolated value. Since it is equipped with a comparator that selects and outputs the reference level based on whether it is far away, it is possible to convert the level to a level close to the normal reference signal level even if the level changes due to dust or dirt adhering to the white reference plate. Therefore, good shading correction can be performed.
Therefore, when binarizing the gradation reproduction of the grayscale image, the image quality deterioration due to the error data of the reference signal is compensated,
A good binary image can be extracted.

[Brief description of drawings]

FIG. 1 is a block diagram showing the configuration of an image reading apparatus according to an embodiment of the present invention.

FIG. 2 is a block diagram showing a configuration of a conventional image reading device.

[Explanation of symbols]

 11 CCD image reading unit 12 A / D converter 13 Interpolation filter 14 Comparator 15 Shading corrector 16 MTF corrector 17 Binarization processing unit 18 Dither matrix 19, 20, 21, 22 Register 23, 24, 25, 26 Multiplication Vessel 27 adder

Claims (1)

[Claims] 1. A CCD image reading unit for reading an original image, an A / D converter for converting the read image data from an analog signal to a digital signal, a shading corrector for correcting reading characteristics, and an image sharpness An image reading apparatus including an MTF corrector for increasing the tone and a binarization processing unit for performing binarization by extracting the gradation of a grayscale image, and reading a document image before the shading corrector. Prior to the above, there is an interpolation filter that uses the dot level before and after the white reference signal output from the CCD image reading unit when the reflected light from the white reference plate is received as an interpolation value, and the level of the current dot with respect to the interpolation value. An image reading apparatus provided with a comparator that determines whether or not a distance is equal to or more than a value and selectively outputs a reference level.