JPH10173874A - Image processor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To prevent the excessive elimination of the highlight of a pattern by detecting a feature value that specifies a certain original type in a first scan mode and performing the conversion of density, i.e., the elimination of texture in response to the detected feature value. SOLUTION: An original type decision circuit 8 decides the type of an original in a pre-scan mode before the main scanning is performed, and a texture level detection part 9 detects and decides the texture level of the original. A correction part 10 corrects the texture level based on the decision result of the circuit 8. Then the part 10 adds the offset to the texture level in response to the output of the circuit 8. An original of a binary structure, i. e., a dot original including the on/off states of characters or dots has a small problem even though it has the considerable rough texture. When it's decided that no characters nor dots are included, it's supposed that a pattern consists some special material and the control is required to evade the extremely rough texture of the original, i.e., not to add the excessive offset to the texture level of the original.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a copier, a facsimile, and the like, and more particularly, to an image processing apparatus for reproducing an image by removing a background according to a document.

[0002]

2. Description of the Related Art In a color copying machine, it is preferable not to apply toner to the background portion in order to eliminate the fogging phenomenon of the background portion. For this reason, when the density level is below a certain threshold value, the signal is set to 0 (no toner is applied) and sent to the printer. However, if this threshold value is set to an excessively large value, the highlight portion in the picture will jump too much. To prevent this, some techniques for automatically setting the threshold value with high accuracy have been proposed.

As a conventional method of forming an image by removing the background, for example, as in the invention described in Japanese Patent Publication No. 6-36553, the entire document is scanned once and the white peak of the document is scanned. A method of determining the background level and information level from the histogram of the black peak, and then determining the quantization slice level using the previously determined background level and information level to perform image formation (and background removal simultaneously). However, thresholds and the like are not switched depending on the document.

As another method, there is a background removal method in an image processing apparatus previously filed by the present applicant.
This is a method in which the background removal level is changed in accordance with the frequency of the character area pixels, photographic paper area pixels, and halftone area pixels. By the way, the printed original is turned on / off of dots called so-called halftone dots.
Since the information is turned off, the projection information remains in the data read by the scanner. In contrast, photographic paper (or a picture drawn with paint or a picture drawn with crayon) has its data continuously changed after being read by a scanner. If background removal is performed on both of these characteristics with a certain threshold value, the former information is likely to remain even with a somewhat higher threshold value, but the latter does not have such a thing, and image quality degradation is remarkable. This is shown in FIG.

[0005] This shows the difference in highlight portion reproduction when the background is removed with the same threshold value in the case of a print original and the case of a photographic paper original. FIG. 7A shows a state of “remaining as information” in a portion where the density level is larger than a threshold value TH0 represented by a dotted line (a shaded portion in the drawing). FIG. 7 (b) shows a case where the density level is smaller than the threshold value TH0 represented by a dotted line, and "does not remain as information" in the case of a photographic paper original. In the area of F, the jump is noticeable.

[0006]

SUMMARY OF THE INVENTION Accordingly, an object of the present invention is to provide a color copying machine which detects a characteristic amount which specifies a certain type of original at the time of a first scan, and performs density conversion, that is, background removal in accordance with the characteristic amount. Accordingly, an object of the present invention is to provide a device that does not remove unnecessary highlights of a pattern.

[0007]

According to a first aspect of the present invention, in an apparatus for processing an image by scanning an original a plurality of times by an image reading apparatus, an original type determining means for determining the type of the original and a background of the original are provided. A background level detecting unit for detecting a level, a background level correcting unit for correcting the background level, and an image density converting unit for performing density conversion of an image based on the value after the background level correction, and determining a type of the document The document type determining means and the background level detecting means for detecting the background level of the document operate at the time of the first scan, and change the correction in accordance with the type of the document.

According to the second aspect of the present invention, the type of the document is determined based on the frequency of character area pixels or halftone area pixels in the document.

According to the third aspect of the present invention, when it is determined that the original does not include the character area, the correction is performed so that the background is not further removed.

According to the fourth aspect of the invention, when it is determined that the original does not include the halftone dot region, the correction is performed so that the background is not removed more.

In the invention according to claim 5, when it is determined that the original does not include the character area and the halftone area, the correction is performed so that the background is not further removed.

According to a sixth aspect of the present invention, there is provided an apparatus for reproducing an image by scanning a document a plurality of times by an image reading device and forming a plate of a plurality of colors, wherein the type of the document is determined. Determining means and a background level detection means for detecting the background level of the document, a background level correction means for correcting the background level and an image density conversion means for performing density conversion of the image based on the value after the background level correction, The document type determination means for determining the type of the document and the background level detection means for detecting the background level of the document operate at the time of the first scan, and in a device that changes the correction according to the type of the document, the black plate generation is performed simultaneously at the first scan. I decided to do it.

[0013]

Embodiments of the present invention will be described below in detail with reference to the drawings. In particular, the present invention detects the background level and determines the document type at the time of the first scan when operating the color copier, and determines the background level of the document detected according to the result of the document type determination after the second scan. By performing the correction, the background can be optimally removed for each document, and the image can be reproduced. In order to realize this at low cost, the document type determination is performed based on two frequencies: the frequency of character pixels and the frequency of halftone pixels.

FIG. 1 shows an outline of the present invention. The present invention relates to an apparatus for processing an image by scanning an original a plurality of times with an image reading apparatus, comprising: an original type determining unit for determining the type of the original; a background level detecting unit for detecting a background level of the original; A document type determining means for determining the type of the document, and a background for detecting the background level of the document, comprising: a background level correcting means for performing correction; and an image density converting means for performing density conversion of the image based on the value after the background level correction. The level detecting means operates at the time of the first scan, and changes the correction in accordance with the type of the document.

A color image digital signal R, G, B which is linear with respect to reflectance from an image input unit 1 such as a scanner.
(8 bits each) is input. Next, logarithmic conversion is performed in the logarithmic conversion unit 2 as pre-processing for performing color correction in the subsequent stage,
A color image digital signal R, G, B data with a linear density is generated. The color correction / black correction unit 3 considers the characteristics of the image input device and the image output device and converts the RGB signal into a complementary color YM.
Convert to C signal. As a color correction method, linear approximation, so-called masking method, tetrahedral interpolation method, triangular prism interpolation method, and the like have been proposed, and color correction is performed using these methods. At the same time, the K signal is calculated by inputting min (C, M, Y) and performing an appropriate table conversion as shown in FIG.

Next, the UCR section 4 performs under color removal from the YMC signal to generate a signal y (m, c) = Y (M, C) -K. However, the K signal in the UCR unit 4 is assumed to be through. These signals are converted by the density conversion unit 5 so as to reduce the density of the highlighted portion. Details of this density conversion will be described later. Then, the halftone processing section 6 performs dither processing and density conversion processing suitable for the characteristics of the output device, and sends the data to an image output section 7 such as a printer.

Also, prior to the main scan, the document type is determined by the document type determination circuit 8 at the time of pre-scanning, and the background level of the document is detected by the background level detector 9 to determine the background level. Then, the correction unit 10 performs background level correction according to the determination result of the document type.

[0018]

Embodiments of the present invention will be described below. First, the operations of the document type determination circuit 8, the background level detection unit 9, the correction unit 10, and the density conversion unit 5 of the present invention will be described below in detail based on the embodiments shown in the drawings.
The document type determination method shown in FIG. 1 performs processing on a green (G) signal. The determination process may be performed alone or may be performed in parallel to make a comprehensive determination.
An example of the comprehensive judgment will be described later.

Next, a description will be given of character / non-character document determination (binary output) based on the character region pixel frequency. A pixel that is a part of a character (hereinafter referred to as a character pixel) is detected, and it is determined whether or not the target document contains a character based on the count value. This is shown in FIG. Here, Japanese Patent Laid-Open No.
The invention disclosed in Japanese Patent No. 92957 is used. FIG. 3 shows a block diagram of the character image determination device. After the detection of the character pixels by the character pixel detection circuit 10, the character pixels are counted for each block of a predetermined size (for example, 8 × 8). The count value of the character block and the threshold value are compared and determined by the comparator 12 to determine the target document. Alternatively, character pixels in all images may be counted, and the target document may be determined based on the counted value.

The printing / non-printing original determination (binary output) based on the halftone dot pixel frequency will be described. A pixel which is a part of a halftone dot (hereinafter, referred to as a "halftone pixel") is detected, and whether or not the target original is a halftone original, that is, a print original is determined based on the count value. The detection of the halftone pixels here is described in
Utilizing the invention disclosed in JP-A-115988. However, in the present invention, a pixel corresponding to a halftone pixel here is described as an "extremum pixel".

FIG. 4 is a block diagram showing a print / non-print original determination process. After the halftone pixel detection circuit 13 detects halftone pixels, halftone pixels are counted for each block of a predetermined size (for example, 8 × 8). The count value of the halftone dot block and the threshold value are compared and determined by the comparator 12 to determine the target document. Alternatively, halftone pixels in all images may be counted, and the target document may be determined based on the counted value.

The background level detector 9 sets max (r, g,
b) Detect the background level (th) in the document from the signal.
Here, the minimum value [max (r, g,
b), min (r, g, b)]. This is shown in FIG. In FIG. 5, ImD is an area where image data is present, an area BLS surrounded by a dotted line is a background level detection range, and MinP is a minimum value point. In this embodiment, an example in which a minimum value is found in a part of a cloud is shown. Show.

As shown in the embodiment of FIG. 5, the reason why the detection area is limited to the inside of the image rather than the whole image is that "the peripheral data may not be a document due to positional deviation of the paper" or " Shading correction is not fully performed in the periphery in the scanning direction. "

The correction unit 10 adds an offset to the background level (th) according to the output of the document type determination circuit 8. A manuscript with few problems even if the background is drastically skipped is a manuscript having a binary structure. More specifically, the document is a halftone original including characters and dots on / off. Therefore, in the present embodiment, when it is determined that neither a character nor a halftone dot is included, it is assumed that a picture made of some other material is included, and the background is not further skipped, that is, an offset is not added much. Perform control.

For example, two cases composed of a combination of the above-described document type determinations will be described as follows. (1) When neither character nor photographic paper is included, for example, δ = 2 (in 8 bits of density signal) (2) In other cases, for example, δ = 10 (in 8 bits of density signal) This is shown in FIG. Based on the output (th + δ) from the correction unit 10, the density conversion unit 5 converts the image data.

Next, a second embodiment will be described. The above is a case where the mode of the color copying machine is not particularly considered. An example in which the mode of the color copying machine is considered will be described below. For example, many current color copiers copy a document by inputting the type of the document in advance, such as "text / print paper mode", "text / print mode", or "picture mode". Therefore, "character / photographic paper mode"
Is selected, since a halftone dot, that is, a printed document does not originally enter, there is no need to activate the above-described halftone dot determination, and the offset level may be controlled by the presence or absence of a character area. (1) When no character is included, for example, δ = 2 (at 8 bits of density signal) (2) At other times, for example, δ = 10 (at 8 bits of density signal)

Next, a third embodiment will be described. Similarly,
When the "picture mode" is selected, since the character original does not originally enter, the above-described character determination does not need to be activated, and the offset level may be controlled by the presence or absence of the halftone dot area. (1) When a halftone dot is not included, for example, δ = 2 (at 8 bits of density signal) (2) At other times, for example, δ = 10 (at 8 bits of density signal)

The above is a method in which the original type is determined and the background level is detected during the pre-scan, and the background is removed during the subsequent main scan. And later CM
A method of removing the background from the Y plate is also possible.

This is because if the conversion shown in FIG. 2 is performed, the K-plate does not enter the highlight (up to a certain point).
A method in which the background is not removed from the plate may be used.
Of course, if it is desired to remove the background to a higher level, the background may be removed from the K-plate signal by threshold processing at a desired level.

Although the present invention has been described using a color copying machine as an example,
For example, the present invention may be applied to a system in which a color scanner is connected to a host computer. That is, a prescan is performed, the type of the document is determined from the G signal from the scanner, the background level of the document is detected, and the image conversion of the present invention is performed on the RGB signal. Further, the present invention is not limited to a color signal, but can be applied to, for example, a monochrome facsimile. Further, the present invention can be widely applied to not only a color copying machine but also an information processing apparatus, an office automation apparatus, and the like, as long as the background removing apparatus in the image processing apparatus that satisfies the gist of the present invention.

[0031]

According to the first aspect of the present invention, since the original type is automatically determined and the correction amount (+ δ) after the background level detection is changed based on the result, a uniform offset is applied. Compared to the case, the background can be optimally removed without unnecessary removal of highlights of the document.

According to the second aspect of the present invention, the type of the original is determined based on the frequency of the character area pixels or the halftone area pixels in the original, and the correction amount (+ δ) after the background level detection is determined based on the result. Because of the change, compared to the case where a uniform offset is applied, the text document and the print document are made to skip the background (the character document wants to skip the highlight portion. The highlight remains even if the print document is slightly skipped). ), And other pattern highlights can be removed optimally without unnecessary removal.

According to the third aspect of the invention, when it is determined that the original does not include characters, correction is performed so that the background is not further removed, so that a reproduced image in which other picture highlight portions are more preserved is obtained. be able to.

According to the fourth aspect of the present invention, when it is determined that the original does not include a halftone dot, correction is performed so that the background is not further removed. Obtainable.

According to the fifth aspect of the present invention, when it is determined that the original contains neither characters nor halftone dots, correction is performed so that the background is not further removed. Images can be obtained.

According to the sixth aspect of the present invention, since the image recognition and the background level detection are performed while forming the K plate, the image can be reproduced at a high speed without performing the press-can.

[Brief description of the drawings]

FIG. 1 is a diagram showing a schematic configuration of a color copying machine to which the present invention is applied.

FIG. 2 is a diagram showing a state of K-plate generation.

FIG. 3 is a diagram showing an embodiment of a character image determination device.

FIG. 4 is a diagram illustrating an embodiment of a print image determination apparatus.

FIG. 5 is a diagram showing a state of background level detection.

FIG. 6 is a diagram showing density conversion.

FIG. 7A is a diagram illustrating a difference in highlight portion reproduction when background removal is performed at the same threshold value (in the case of a print original);
FIG. 4B is a diagram illustrating a difference in highlight portion reproduction when the background is similarly removed at the same threshold value (in the case of a photographic paper original).

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Image input part 2 Logarithmic conversion part 3 Color correction / black generation part 4 UCR part 5 Density conversion part 6 Halftone processing part 7 Image output part 8 Document type judgment part 9 Background level detection part 10 Correction part 11 Character pixel detection circuit 12 , 15 Counter circuit 13, 16 Comparator

Claims (6)

[Claims] 1. An apparatus for processing an image by scanning an original a plurality of times with an image reading apparatus, an original type determining means for determining the type of the original, a background level detecting means for detecting the background level of the original, and the background level And a document type determining unit for determining the type of the document, and detecting a background level of the document. An image processing apparatus characterized in that the background level detection means operates at the time of the first scan and changes the correction in accordance with the type of the document. 2. The image processing apparatus according to claim 1, wherein the determination of the document type is made based on the frequency of character area pixels or halftone area pixels in the document. 3. The image processing apparatus according to claim 1, wherein when the original is determined not to include a character area, the correction is performed so that the background is not removed more. 4. The image processing apparatus according to claim 1, wherein when the original is determined not to include a halftone area, the correction is performed so that the background is not further removed. 5. The image processing apparatus according to claim 1, wherein when the original is determined to include neither a character area nor a halftone area, the correction is performed so that the background is not further removed. 6. An apparatus for reproducing an image by scanning a document a plurality of times by an image reading device to form a plate of a plurality of colors, comprising: a document type determining unit for determining a type of the document; and a background of the document. A background level detecting unit for detecting a level, a background level correcting unit for correcting the background level, and an image density converting unit for performing density conversion of an image based on the value after the background level correction, and determining a type of the document The document type judging means and the background level detecting means for detecting the background level of the document operate at the time of the first scan, and simultaneously perform black plate generation at the time of the first scan in a device for changing the correction according to the type of the document. Copying equipment.