JP2000115529A - Picture reproducing device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To faithfully reproduce an original by judging whether the original attached importance to highlight reproduction. SOLUTION: A dot area detecting circuit 5 detects the dot picture pattern area of the original by using picture data from a γtransforming circuit 4. A highlight area detecting circuit 6 compares the density of each picture element of picture data and a previously fixed threshold with each other to detect a picture element of a density level lower than the threshold to set the picture element to be an active picture element. An AND circuit 7 executes such processing that a picture element under consideration becomes an active picture element at the time of being in a dot picture area and being a low density picture element by signals from the circuit 5 and the circuit 6. A counter 8 counts the active picture element outputted from the circuit 7. A circuit 9 compares the counted value of the counter 8 and the previously fixed threshold with each other and in the case that the counted value is larger than the threshold value, judges that the original attaches importance to highlight reproduction.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image reproducing apparatus such as a digital copying machine or a facsimile apparatus, and more particularly to a method for visualizing an image of a document read by a scanner with toner.
The present invention relates to reproducibility of a highlight portion of a picture and a phenomenon in which so-called "fogging" in which toner adheres to a white background portion of a document does not occur, which is incompatible with image processing, as much as possible.

[0002]

2. Description of the Related Art In a digital copying machine or the like, in order to reproduce a high-quality image, it is desired to increase the reproducibility of a highlight portion of the image while generating so-called "fog" in which toner adheres to a white background portion. There is a request that it is not.
On the other hand, there is a background removal process that removes the background according to the original and reproduces an image as a device that achieves both of them as much as possible by image processing. This background removal processing detects the background level of a read document and changes the correction amount applied to the background level according to the document type mode, as disclosed in Japanese Patent Application Laid-Open No. 8-307687. Is applied to the background level, and the density of the image is converted based on the corrected value of the background level, so that the background can be optimally removed without removing the highlight portion of the picture of the document more than necessary.

[0003]

The background removal processing performed on the detected background level by changing the correction amount performed on the background level in accordance with the document type mode as described above does not include the highlight portion of the picture pattern so much. Although a sufficient effect can be exerted on a manuscript, the tendency of skipping too much on a manuscript containing a very large number of highlights in a picture was strong. In the case of such a manuscript, a "fogging"
Even if there is a slight occurrence, it is better to reproduce the highlight of the low density part as much as possible,
The image is preferred by the user.

The present invention has been made in view of the above-mentioned points, and a document which emphasizes highlight reproduction by a simple process based on the area of a picture highlight area in the document or the ratio of the highlight area to the picture area. It is an object of the present invention to provide an image reproducing apparatus capable of judging whether or not highlight reproduction is important and reproducing the area more faithfully to the original without significantly increasing the cost. It is.

[0005]

An image reproducing apparatus according to the present invention digitally reads an original and removes the background by predetermined processing on input image data to digitally reproduce an image. And a document type recognizing means for determining whether or not the input image data includes a picture area and emphasizes highlights.

An image reproducing apparatus according to a second aspect of the present invention is an image reproducing apparatus which digitally reads a document and removes a background by predetermined processing from input image data to digitally reproduce an image. Background level detection means for detecting the background level of the document, document type recognition means for determining whether or not the input image data includes a picture area and which emphasizes highlight reproduction, and document type recognition means And a correcting means for controlling a level at which the background level detected by the background level detecting section is removed.

The document type recognizing means may detect a pixel or a block corresponding to a highlight of a picture, compare the counted value with a predetermined threshold value, and determine whether or not to emphasize the highlight reproduction.

The document type recognizing means detects a pixel or a block corresponding to a picture, detects a pixel or a block corresponding to a highlight of the picture, and compares each count value with a predetermined threshold to reproduce a highlight. It is desirable to determine whether or not importance is to be placed on.

The above-mentioned pattern is a halftone dot pattern area of printing, a pattern area of printing paper, or an area surrounded by a non-white background.

[0010]

DESCRIPTION OF THE PREFERRED EMBODIMENTS An image reproducing apparatus according to the present invention has an image input section, a document type recognition section, and an image processing section. The document type recognition unit determines whether or not the input image data includes a picture region and is a document that emphasizes highlighting. The gamma conversion circuit, the halftone dot region detection circuit, the highlight region detection circuit, It has an AND circuit, a counter, and a judgment circuit.

The image input section reads a document and outputs a multi-valued digital signal according to the response. The γ-conversion circuit converts a linear reflectance signal output from the image input unit into a linear density signal. The halftone dot area detection circuit detects a halftone dot pattern area of the document using the image data output from the gamma conversion circuit. Highlight area detection circuit is γ
The density of each pixel of the image data output from the conversion circuit is compared with a predetermined threshold, a pixel whose density level is lower than the threshold is detected, and the pixel is set as an active pixel. The AND circuit performs a process based on signals from the halftone area detection circuit and the highlight area detection circuit so that when the target pixel is a halftone image area and a low density pixel, the pixel becomes an active pixel. The counter counts active pixels output from the AND circuit. The determination circuit compares the count value of the counter with a predetermined threshold value. If the count value is larger than the threshold value, the determination circuit determines that the document emphasizes highlight reproduction and outputs the document to the image processing unit.

[0012]

FIG. 1 is a block diagram showing the configuration of an embodiment of the present invention. As shown in the figure, the image reproducing apparatus has an image input unit 1, a document type recognition unit 2, and an image processing unit 3.
The image input unit 1 includes a scanner or the like, and reads a document digitally and outputs image data. The document type recognizing unit 2 determines whether or not the input image data includes a picture region and is a document that emphasizes highlighting. The gamma conversion circuit 4, the halftone region detecting circuit 5, the highlight region It has a detection circuit 6, an AND circuit 7, a counter 8, and a judgment circuit 9. The γ conversion circuit 4 converts a linear reflectance signal output when an image is read by the image input unit 1 into a linear density signal. The dot area detection circuit 5 detects a dot picture area of the document based on a signal output from the γ conversion circuit 4. The highlight area detection circuit 6 detects a low density area of the document based on a signal output from the γ conversion circuit 4. The AND circuit 7 includes a halftone picture area of the original document detected by the halftone area detection circuit 5 and a low density area of the original from the low density area of the original detected by the highlight area detection circuit 6 and the low density area of the original. Is detected. The counter 8 counts active pixels or blocks at the output of the AND circuit 7.
The determination circuit 9 determines whether or not the document includes a picture area and emphasizes highlights based on the count value of the counter 8.
The image processing unit 3 performs various types of image processing on the image data input from the image input unit 1 and performs background removal based on the determination result output from the determination circuit 9 to digitally reproduce the image.

An operation for judging whether or not the original read by the image input unit 1 by the original type recognition unit 2 of the image reproducing apparatus configured as described above includes a picture area and is an original emphasized on highlighting. Will be described.

The image input unit 1 reads an original using a CCD or the like, and outputs a multi-valued digital signal according to the response.
For example, an 8-bit signal is output at a linear reflectance of 400 dpi. The γ-conversion circuit 4 converts the linear reflectance signal output from the image input unit 1 into a linear density signal. When the table conversion is performed by the γ conversion circuit 4, the attribute of the signal does not need to be strictly linear in density, and the signal may be appropriately nonlinearly converted so as to be convenient for the subsequent determination processing. The dot area detection circuit 5 detects the halftone picture area of the document using the image data output from the γ conversion circuit 4. The detection of the halftone dot pattern area of this manuscript is performed, for example, by using the IEICE Transactions (D-II Vol. J75-D-II)
No. 1 pp. 39-47, January 1992), a halftone dot area detection process using peak information of halftone dots, as shown in an image area separation method for mixed images of characters / patterns (halftone dots, photographs). The overall determination of the edge area detection processing using the continuity of black pixels and white pixels determines whether the area is a character area, a halftone dot area, or a picture area indicating a photographic area. That is, the halftone picture area of the original is determined by a method in which the area determined as the non-halftone area by the halftone area detection and the edge area detected by the edge area detection is set as the character area, and the other areas are determined as the picture area. Is determined. The highlight area detection circuit 6 compares the density of each pixel of the image data output from the γ conversion circuit 4 with a predetermined threshold value, detects a pixel whose density level is lower than the threshold value, Is an active pixel. When detecting the low density area of the original, if the original is a halftone dot, the signal is locally rich and undulating. As a preprocessing, a masking process of about 5 × 5 is performed by a smoothing filter. Should be applied. The AND circuit 7 includes a halftone area detection circuit 5 and a highlight area detection circuit 6
Is processed in such a manner that when the target pixel is a halftone image area and a low density pixel, the pixel becomes the active pixel. The counter 8 counts active pixels output from the AND circuit 7. The determination circuit 9 compares the count value of the counter 8 with a predetermined threshold value. If the count value is larger than the threshold value, the determination circuit 9 determines that the document is of importance for highlight reproduction and outputs it to the image processing unit 3.

The pixels which are halftone image areas and low density pixels are counted from the image data thus read, and the frequency of highlighting of the picture is detected. It is possible to determine with high accuracy whether or not a document should be reproduced with emphasis on writing.

In the above embodiment, when the target pixel is a halftone image area and a low-density pixel, it is determined as an active pixel by the signals from the halftone area detection circuit 5 and the highlight area detection circuit 6 as the active pixels. The case where it is determined from the numerical values whether or not the document emphasizes highlight reproduction has been described.At the same time, the pixel corresponding to the picture is detected and the pixel corresponding to the highlight of the picture is detected, and each count value and a predetermined threshold value are determined. A comparison may be made to determine whether emphasis is placed on highlight reproduction.

The configuration of the second embodiment is shown in the block diagram of FIG. As shown in FIG. 2, the document type recognition unit 2a includes a gamma conversion circuit 4, a halftone dot area detection circuit 5, a highlight area detection circuit 6, an AND circuit 7, a counter 81, a counter 82, and a determination circuit 91. The counter 81 counts the number m of blocks of the halftone picture area of the document detected by the halftone area detection circuit 5. The counter 82 counts the number n of active pixels output from the AND circuit 7. The determination circuit 91 determines from the count value m of the counter 81 and the count value n of the counter 82 whether or not the document emphasizes highlight reproduction.

The document type recognition section 2a configured as described above
The operation for determining whether or not the original read by the image input unit 1 includes a picture area and emphasizes highlights will be described. The image input unit 1 reads a document and outputs a multi-valued digital signal. The linear reflectance signal output from the image input unit 1 is converted into a linear density signal by the γ conversion circuit 4 in a table, and the halftone dot area detection circuit 5 uses the image data output from the γ conversion circuit 4 to convert the original. Is detected. On the other hand, from the image data output from the γ-conversion circuit 4, the highlight area detection circuit 6 detects a pixel whose density level is lower than the threshold, and sets that pixel as an active pixel. The signals from the dot area detection circuit 5 and the highlight area detection circuit 6 are input to an AND circuit 7,
When the pixel of interest is a halftone image area and a low density pixel, the D circuit 7 performs a process to be an active pixel.
Then, the counter 81 counts the number m of blocks in the halftone picture area of the document detected by the halftone area detection circuit 5, and the AND circuit 7
The counter 82 counts the number n of active pixels output from the counter and sends it to the determination circuit 91. The determination circuit 91 compares the number m of blocks in the halftone dot pattern area of the original sent from the counter 81 with a predetermined threshold Th1, and when the number m of blocks in the halftone dot pattern area is larger than the threshold Th1, It is determined that a dot pattern is included, and when the ratio n / m of the number m of blocks to the number n of active pixels counted by the counter 82 is larger than a predetermined threshold Th2, a highlight portion is displayed in the halftone dot pattern. It is determined that the area occupied by the document is large, and the document is determined to emphasize the highlight reproduction.

Since the ratio between the halftone dot pattern area and the pattern highlight is detected from the read image data, the original is reproduced with emphasis on the pattern highlight when reproducing the image from the read image data. It can be determined with high accuracy whether or not the original should be printed.

In each of the embodiments described above, the case where the halftone picture area of the original is detected by the halftone area detection circuit 5 is described. When the photographic paper is the original, for example, the halftone dot using peak information of halftone dots is used. The desired judgment is not made from the characteristics of the algorithm for judging the character area or the halftone picture area by the comprehensive judgment of the dot area detection processing and the edge area detection processing using the continuity of black pixels and white pixels. Therefore, when the original is photographic paper, the halftone dot area detection circuit 5 shown in FIG.
As shown in the block diagram of FIG.
1, the photographic paper picture area detection circuit 51 stores the signal after the γ conversion in a necessary line buffer, and determines whether or not the pixel of interest is a photographic paper picture area. It is preferable that the pattern portion of the photographic paper is flat in terms of local data, and that an intermediate density is detected in order to distinguish it from other areas. Therefore, the photographic paper design area detection circuit 51 determines whether or not the target pixel 512 is a photographic paper design area by using a determination mark 511 as shown in FIG. 4, for example. The determination mark 511 shown in FIG. 4 is a 5 × 5 pixel mark when the density x of all pixels in the 5 × 5 pixel mask satisfies α <x <β with respect to predetermined density parameters α and β. Attention pixel 5 at the center of the mask
12 is determined as the picture area of the photographic paper. Here, the density parameter α is set to include the highlight area. By detecting the picture area of the photographic paper in this way, when reproducing an image from the read image data, it is possible to determine with high accuracy whether or not the document is a document to be reproduced with emphasis on the pattern highlight.

In each of the above embodiments, the case where the judgment processing is performed in pixel units has been described. However, for example, a low density block may be detected as a representative of the average value of 4 × 4 pixels, and the judgment processing may be performed in block units.

In each of the above embodiments, the case where a halftone dot pattern area of a manuscript and a picture area of a photographic paper are detected has been described. For example, when a grandchild copy called a so-called generation manuscript is input as a manuscript, The desired judgment cannot be made from the characteristics of the algorithm. Therefore, as shown in the block diagram of FIG. 5, a non-white background area detection circuit 52 is provided in place of the halftone area detection circuit 5, and the non-white background area detection circuit 52 stores the signal after γ conversion in a necessary line buffer, It is determined whether the block of interest is a picture area. The so-called pattern portion of the generation original is also a pattern around the portion, and is a region surrounded by a non-white background region in terms of processing. For this reason, as shown in FIG. 6, A orthogonal to the target block 521 composed of, for example, 4 × 4 pixels is used.
Region 522, B region 523, C region 524, and D region 5
The pattern portion of the generation original is detected by using the determination mask 526 having 25. This determination mask 526
Each of the regions 522 to 525 includes 16 blocks, and the A region 522, the B region 523, the C region 524, and the D region
If there is at least one non-white block in any of the regions 525, the block of interest 521 is determined as a picture region.
By detecting a picture area of a so-called generation document in this way, when reproducing an image from read image data, it is determined with high accuracy whether the document is a document to be reproduced with emphasis on the pattern highlight. be able to.

Although the above embodiment has been described with reference to a case where a photographic paper picture area detection circuit 51 and a non-white area detection circuit 52 are provided in place of the dot area detection circuit 5 of the image reproducing apparatus shown in FIG. Instead of the dot area detection circuit 5 of the image reproducing apparatus shown in FIG. 2, a photographic paper picture area detection circuit 51 and a non-white background area detection circuit 52 may be provided.

Next, as described above, the document type recognition unit 2
A description will now be given of an image reproducing apparatus which removes the background and reproduces the image of the document based on the result of determining whether or not the document is a document to be reproduced with emphasis on the pattern highlight.

FIG. 7 is a block diagram showing the configuration of an image reproducing apparatus for removing background and reproducing an image of a document based on the result of judging whether or not the document is a document to be reproduced with emphasis on picture highlighting. As shown in FIG. 7, the image reproducing apparatus includes an image input unit 1, a background level correction unit 11, an image processing unit 3, and an image output unit 12. The background level correction unit 11
The document type recognizing unit 2, the background level detecting circuit 13, and the correcting circuit 14 shown in (1) above are used, and when the image input unit 1 performs pre-scanning, it is determined whether or not the document should be reproduced with emphasis on the pattern highlight. The background level detection circuit 13 includes the image input unit 1
Detects the background level Th of the document read during the prescan. The correction circuit 14 corrects the background level Th of the document detected by the background level detection circuit 13 based on the result of the document type recognition unit 2 determining whether or not the document should be reproduced with emphasis on the pattern highlight. The image processing unit 3 includes a smoothing filter 31
Γ conversion circuit 32, edge enhancement filter 33, density conversion circuit 34, and halftone processing circuit 35. The smoothing filter 31 smoothes the image data of the document read by the image input unit 1 for suppressing moiré and removing noise. The γ conversion circuit 32 converts a linear reflectance signal output from the smoothing filter 31 into a linear density. The edge enhancement filter circuit 33 performs edge correction for image data output from the γ conversion circuit 32 and edge enhancement processing for sharply reproducing characters. The density conversion circuit 34 outputs a correction signal (Th +
According to a), the background fog of the image data output from the edge enhancement filter circuit 33 is suppressed. Halftone processing circuit 3
Reference numeral 5 creates pseudo halftone data suitable for the image output unit 12 based on the image data output from the density conversion circuit 34 and in which background fog is suppressed. The image output unit 12 includes a halftone processing circuit 35
Prints out the image data created in.

When image data of a document read by the image input unit 1 is printed out by the image reproducing apparatus configured as described above, first, when the document is prescanned by the image input unit 1, a background level detection circuit is used. Reference numeral 13 detects the background level Th in the document. When the background level is detected and the background level Th is larger than a predetermined value, it is determined that there is no background and the background level Th = 0 is output. The detection of the background level Th is described in, for example, JP-A-8-30768.
As described in Japanese Patent Publication No. 7, the minimum value of the density in the document is detected and set as the background level Th. On the other hand, the document type recognition unit 2 determines whether or not the document to be read is a document to be reproduced with emphasis on the pattern highlight. The correction circuit 14 corrects the background level Th detected by the background level detection circuit 13 using information on whether or not the document should be reproduced with emphasis on the picture highlight determined by the document type recognition unit 2. For example, if the document emphasizes highlights, the background level Th is not corrected. If the document does not emphasize highlights, the background level T is corrected.
h is corrected to a corrected background level (Th + a). When the background level Th is corrected, the background level Th is corrected to (Th-b) if the document emphasizes the highlight, and the background level Th is corrected to (Th + a) if the document does not emphasize the highlight. You may.

In this state, the original is read by the image input unit 1 and a multi-valued digital signal is output. The smoothing filter 31 of the image processing section 3 performs a filtering process to suppress the occurrence of moire in the output image signal and to remove noise mixed in when digitizing the image signal, and sends the result to the γ conversion circuit 32. As this filtering processing, for example, an average value of the density of each point in an arbitrary square or rectangular area centered on a certain point is obtained and used as a new density of the smoothed image. For example, FIG.
As shown in (a), a filtering process is performed by a spatial filter M (i, j) in which a weight coefficient is changed according to a distance from a point of interest. The coefficients for the filtering process are changed according to the characteristics of the image input unit 1 and the image output unit 12 and the image characteristics to be obtained. The γ conversion circuit 32 converts the sent image signal from a substantially linear reflectance to a substantially linear density according to the printer characteristics of the image output unit 12, and sends the converted image signal to the edge enhancement filter 33. The edge emphasis filter 33 is used, for example, as shown in FIG. 8B to correct the blur of the edge of the image or to reproduce the character image sharply.
The filtering process is performed by a differential filter g (m, n) having a coefficient as shown in FIG. The coefficients for this filtering process are also changed according to the characteristics of the image input unit 1 and the image output unit 12 and the image characteristics to be obtained. The density conversion circuit 34 receives the corrected background level signal from the correction circuit 14 in order to suppress the background fog of the transmitted image signal, and for example, converts the input image data and the primary image of the converted image data as shown in FIG. A conversion table is created, optimal density conversion is performed according to the background level of the original to be read and whether or not highlighting is emphasized, and an image signal whose density has been converted is sent to the intermediate processing circuit 35. The halftone processing circuit 35 performs a binarization process on the transmitted image signal by an error diffusion method or a dither method. If the image output unit 12 accepts a multi-level signal, processing is performed using a multi-level error diffusion method or a multi-level dither method. The processed image signal is output to the image output unit 12.
To print out and reproduce the image of the original.

As described above, it is determined with high precision whether or not the original is an original to be reproduced with emphasis on picture highlights. If it is determined that the original is to be reproduced with emphasis on picture highlights,
Since the amount of background removal is suppressed and the highlight of the picture is reproduced from a lower density, it is possible to reproduce a high-quality image which is preferable for the user.

In each of the above-described embodiments, the "0/1" determination is made as to whether or not the target document is a reproduction emphasized by highlighting. For example, multi-step control is performed in accordance with the output value from the counter 8 shown in FIG. You may go. By performing the multi-step control in this manner, the background level correction value (T
Control can be performed such that the larger the count value, the smaller the correction coefficient a of h + a) becomes.

Each of the above embodiments can be applied to a color image reproducing apparatus. When applied to a color image reproducing apparatus as described above, scanning is repeated a plurality of times.
In the case of the so-called face-at-a-time method, it is only necessary to determine whether or not the original is a document to be reproduced with emphasis on pattern highlights by pre-scanning. It is also possible to determine whether or not the document is a document to be reproduced, and control the background removal processing after the second version. In the latter case, the copying speed can be increased as a whole because the prescan is not performed.

[0031]

As described above, the present invention reproduces an image by counting pixels that are halftone image areas and low density pixels from the read image data and detecting the highlight frequency of the picture. At times, it can be determined with high accuracy whether or not the original is an original to be reproduced with emphasis on picture highlights.

Further, by detecting the ratio between the halftone dot pattern area and the pattern highlight from the read image data,
When reproducing an image from read image data, it is possible to determine with high accuracy whether or not the original is an original to be reproduced with emphasis on picture highlighting.

Further, it is determined with high precision whether or not the original is an original to be reproduced with emphasis on the pattern highlight. If it is determined that the original is to be reproduced with emphasis on the pattern highlight, the background removal amount is determined. Is suppressed and the highlight of the picture is reproduced from a lower density, so that a high-quality image preferable to the user can be reproduced.

Pixels or blocks corresponding to a halftone picture area of printing, a picture area of printing paper or an area surrounded by a non-white background are detected, and a pixel or block corresponding to a highlight of a picture is detected. Since it is determined whether or not to emphasize highlight reproduction by comparing the count value with a predetermined threshold value, various originals can be determined with high accuracy, and a high-quality image can be reproduced and printed.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of an embodiment of the present invention.

FIG. 2 is a block diagram showing a configuration of a second embodiment.

FIG. 3 is a block diagram showing a configuration of a third embodiment.

FIG. 4 is an explanatory diagram illustrating a determination mark according to a third embodiment.

FIG. 5 is a block diagram showing a configuration of a fourth embodiment.

FIG. 6 is an explanatory diagram illustrating a determination mark according to a fourth embodiment.

FIG. 7 is a block diagram showing a configuration of a fifth embodiment.

FIG. 8 is an explanatory diagram showing a configuration of a spatial filter.

FIG. 9 is an explanatory diagram of a primary conversion table between input image data and converted image data.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Image input part 2 Document type recognition part 3 Image processing part 4 γ conversion circuit 5 Halftone area detection circuit 6 Highlight area detection circuit 7 AND circuit 8 Counter 9 Judgment circuit 11 Background level correction section 12 Image output section 13 Background level Detection circuit 14 Correction circuit 31 Smoothing filter 32 γ conversion circuit 33 Edge enhancement filter 34 Density conversion circuit 35 Halftone processing circuit 51 Photographic paper picture area detection circuit 52 Non-white background area detection circuit

 ──────────────────────────────────────────────────続 き Continuing on the front page (72) Inventor Tokiko Miyagi 1-3-6 Nakamagome, Ota-ku, Tokyo F-term in Ricoh Company (reference) 5C076 AA01 AA26 AA27 BA06 CA10 5C077 MP02 MP06 MP08 PP15 PP25 PP27 PP43 PP52 PP54 PP65 PP68 RR14 TT06

Claims (7)

[Claims] 1. An image reproducing apparatus for digitally reading an original and removing background by predetermined processing on input image data and digitally reproducing an image, comprising a picture area for input image data. An image reproducing apparatus comprising: a document type recognizing unit that determines whether a document is emphasized in highlight. 2. An image reproducing apparatus for digitally reading a document and removing background by predetermined processing on input image data and digitally reproducing an image. A background level for detecting a background level of the input document. A detecting unit, a document type recognizing unit that determines whether or not the document includes a pattern area with respect to the input image data and emphasizes highlight reproduction, and a background level detecting unit based on a determination result of the document type recognizing unit. An image reproducing apparatus, comprising: a correction unit that controls a level at which the detected background level is used to perform background removal. 3. The document type recognizing means detects a pixel or a block corresponding to a highlight of a picture, compares the count value with a predetermined threshold value, and determines whether to emphasize the highlight reproduction. 3. The image reproducing device according to 1 or 2. 4. The document type recognizing means detects a pixel or a block corresponding to a picture, detects a pixel or a block corresponding to a highlight of the picture, and compares each count value with a predetermined threshold value for highlighting. The image reproducing apparatus according to claim 1, wherein it is determined whether or not importance is placed on reproduction. 5. The image reproducing apparatus according to claim 1, wherein the picture is a halftone picture area for printing. 6. The image reproducing apparatus according to claim 1, wherein the picture is a picture area of a printing paper. 7. The image reproducing apparatus according to claim 1, wherein the picture is an area surrounded by a non-white background.