JP4073877B2 - Image processing method, image processing apparatus, image forming apparatus, and computer program - Google Patents

Description

  The present invention is applied to a digital copying machine, a facsimile machine, and the like. In order to improve the image quality of a recorded image, an image signal obtained by scanning a document is identified with a character / character identified using the feature amount of each pixel. Image processing method and image processing for identifying printed photographs (halftone dots), photographic paper photographs (continuous tone), and printer output photographs with respect to photographs / graphics in the input image data using halftone dots and background background areas The present invention relates to an apparatus, an image forming apparatus, and a computer program.

  Conventionally, digital copiers and facsimiles are obtained by reading a document with a character portion, photo portion or halftone dot portion, or a document with such portions mixed with a CCD (Charge Coupled Device) image sensor or the like. The processed image signal is subjected to image processing for improving the image quality of the recorded image. As this image processing, whether the read image is a character, a photograph, or a halftone dot is identified, and appropriate processing is performed according to the identification result.

  As this kind of image identification and document type method, only a line drawing is obtained from an image signal obtained by reading a document as in the image type determination method disclosed in Japanese Patent Application Laid-Open No. 8-251406. Consisting of only halftone, consisting only of continuous tone, consisting of line drawing and halftone, consisting of line drawing and continuous tone, consisting of halftone and continuous tone, consisting of line drawing, halftone and continuous tone, There is a method for determining which of the above.

In addition, as a document type determination method for determining whether a document is output from an ink jet printer other than characters, photographs, and halftone dots, a document disclosed in Patent Document 2 (Japanese Patent Laid-Open No. 10-126631). As in the type determination method, it is determined from the image signal obtained by reading the document whether or not the document is a document output by an ink jet printer based on the feature amount related to the color reproduction area and the feature amount related to the texture. There is a way.
JP-A-8-251406 (published September 27, 1996) Japanese Patent Laid-Open No. 10-126631 (published on May 15, 1998)

  Incidentally, the input image data includes a printer output photograph output by a printer in addition to a printed photograph (halftone area) and a photographic paper photograph (continuous tone area). The characteristics are different from a printed photograph consisting of dots and a continuous-tone photographic paper photograph. For this reason, in order to perform appropriate image processing when the input image data includes a printer output photo, it is necessary to appropriately determine that the image is a printer output photo.

  However, in the image identification method disclosed in Patent Document 1, a continuous tone image, that is, a photographic paper photograph can be determined, but an original (printer output) output by a printer having different characteristics from the photographic paper photograph. Therefore, there is a problem in that it is impossible to perform appropriate processing on the input image data including the printer output photo.

  In addition, although the image identification method disclosed in Patent Document 2 can identify a printer photo document output by an ink jet printer, it uses a feature amount related to a texture peculiar to a document output by an ink jet printer. There arises a problem that it is not possible to properly identify a document output by a printer of this type, for example, a laser beam printer or an inkjet printer having different halftone processing, resolution, and dot reproducibility.

  The present invention has been made in view of the above-described problems, and an object of the present invention is to identify characters identified by using feature values of each pixel with respect to input image data read by an image input device such as a scanner. -Uses halftone dots and background background areas to properly identify printed photographs (halftone dots), photographic paper photographs (continuous tone), and printer output photographs for photographs / graphics included in the input image data. By providing an image processing method that can be performed, an image processing device, an image forming apparatus including the image processing device, and a computer-readable recording medium on which a program for performing each procedure of the image processing method is recorded is there.

  In order to solve the above-described problems, an image processing apparatus according to the present invention includes an image type recognition unit that extracts a feature amount from input image data and identifies an input image type. The means includes halftone pixel detection means for determining whether or not each pixel of the input image data or a block composed of a plurality of pixels belongs to a halftone dot area, and each pixel or a plurality of pixels of the input image data. A background background pixel detection means for determining whether or not a block belongs to a background background area, and a pixel or block in the input image data other than the background background pixel or background background block detected by the background background pixel detection means In accordance with the ratio of halftone pixels or halftone blocks detected by the halftone pixel detection means, the input image data is marked with halftone dots. Photos, is characterized in that photographic paper photograph of continuous tone, and the determining photographic type determining means either a printer output photograph output by the printer is contained is provided.

  In general, the image characteristics of a printer output photograph are intermediate between the characteristics of a printed photograph composed of halftone dots and the image characteristics of a continuous tone photographic paper photograph. The printer output photograph can be easily determined. This difference in image characteristics can be determined by the ratio of halftone pixels or halftone blocks to pixels or blocks in the input image data other than background background pixels or background background blocks in the input image data.

  Therefore, as in the above configuration, the input image data has a halftone dot or halftone block ratio to the input image data according to the ratio of the halftone pixel or halftone block to the pixel or block in the input image data other than the background background pixel or background background block. Appropriately determine the type of photo included in the input image if it is determined whether a printed photo consisting of dots, a continuous-tone photographic paper photo, or a printer output photo output by a printer is included. Can do.

  If the above ratio is set so that it can be used to judge printer output photos with different image characteristics, such as inkjet printers, laser beam printers, halftone processing, resolution, and dot reproducibility, etc., printer output Even printer output photos of different systems can be easily identified from other output photos (printed photos, photographic paper photos, etc.) other than printer output photos.

  Further, the image processing apparatus includes a character pixel detection unit that determines whether each pixel of the input image data belongs to a character area, and the photograph type determination unit includes a character pixel or a character block detected by the character pixel detection unit, By the ratio of the halftone pixel or halftone block detected by the halftone pixel detection means to the pixel or block in the input image data other than the background background pixel or background background block detected by the background background pixel detection means, It may be determined whether the input image data includes a printed photograph composed of halftone dots, a continuous-tone photographic paper photograph, or a printer output photograph output by a printer.

  In this case, by adding the character pixel detection means, it is possible to identify the character pixel or character block and the pixel or block in the input image data other than the background background pixel or background background block. That is, since it is possible to use the ratio of halftone pixels or halftone blocks in only a photograph or figure that is originally intended to be excluded, it is possible to determine whether or not a printer output photograph is included with higher accuracy.

  Further, a pixel or block in the input image data other than the background background pixel or background background block detected by the background background pixel detection means, or a character pixel or character block detected by the character pixel detection means and the background A labeling unit for labeling a photo candidate area composed of pixels or blocks in the input image data other than the background background pixel or background background block detected by the background pixel detection unit; For each photo candidate area labeled by the labeling means, the photo candidate area is represented by a halftone dot or halftone block detected by the halftone pixel detection means relative to the pixels or blocks in the area. Print photos consisting of dots, continuous-tone photographic paper photos, output with a printer It may be determined which of the printer output pictures were.

  In this case, for pixels other than background background pixels or blocks other than background background pixels, or pixels other than character pixels and background background pixels, or blocks other than character blocks and background background blocks (photo candidate areas) in the input image data By labeling each photo candidate area, it is possible to individually identify whether each photo candidate area is a printed photo (halftone dot), a photographic paper photo (continuous tone), or a printer output photo. . That is, even when a print photo and a printer output photo are present in one input image data, it is possible to recognize the type of each photo.

  The image processing apparatus having the above configuration may be provided in the image forming apparatus.

  In this case, it is possible to apply image processing in consideration of the characteristics of the printer output photo to the input image data including the printer output photo or the pixels and blocks to which the printer output photo belongs. High-quality output of image data including output photos is possible.

  If an image processing program for causing a computer to function as each unit of the image processing apparatus having the above configuration is used, each unit of the image processing apparatus can be easily realized by a general-purpose computer.

  The image processing program is preferably recorded on a computer-readable recording medium.

  Accordingly, the image processing apparatus can be easily realized on the computer by the image processing program read from the recording medium.

  The image processing apparatus according to the present invention, as described above, has a ratio of halftone pixels or halftone blocks to pixels or blocks in the input image data other than background background pixels or background background blocks in the input image data. Judgment of the type of photo included in the input image data by determining whether the input image data includes a half-dot print photo, continuous-tone photographic paper photo, or printer output photo output by a printer There is an effect that can be performed appropriately.

[Embodiment 1]
An embodiment of the present invention will be described as follows. In this embodiment, an example in which the image processing apparatus of the present invention is applied to a color image forming apparatus will be described.

  As shown in FIG. 2, the color image forming apparatus according to the present embodiment includes a color image input apparatus 1, an image processing apparatus 2, and a color image output apparatus 3.

  Further, the color image forming apparatus is connected to a color image input apparatus 1, an image processing apparatus 2, and a color image output apparatus 3, and a setting button for setting an operation mode of the image forming apparatus (for example, a digital copying machine) An operation panel 4 is provided that includes a display unit including a numeric keypad and a liquid crystal display.

  The color image input device 1 is composed of, for example, a scanner unit, and a CCD (Charge Coupled Device) converts a reflected light image from an original as an RGB (R: red, G: green, B: blue) analog signal. Read.

  The color image output device 3 is a device that outputs a result of predetermined image processing performed by the image processing device 2.

  The image processing apparatus 2 includes an A / D (analog / digital) conversion unit 11, a shading correction unit 12, a document type automatic discrimination unit 13, an input tone correction unit 14, a color correction unit 15, and a black generation and under color removal unit 16. , A spatial filter processing unit 17, an output gradation correction unit 18, a gradation reproduction processing unit 19, and a region separation processing unit 20.

  The A / D converter 11 converts an analog signal read by the color image input device 1 into a digital signal.

  The shading correction unit 12 performs shading correction for removing various distortions generated in the illumination system, the imaging system, and the imaging system of the color image input apparatus 1.

  The document type automatic discrimination unit 13 is an image processing system employed in the image processing apparatus 2 such as a density signal for an RGB signal (RGB reflectance signal) from which various distortions have been removed by the shading correction unit 12. This makes it possible to determine whether the input document image is a text document, a printed photo copy document, a photographic paper photo, or a combined text / print photo document. To do. That is, the document type automatic determination unit 13 outputs information related to the entire document. Details of the automatic document type discrimination unit 13 will be described later.

  The input tone correction unit 14 performs image quality adjustment processing such as removal of background area density and contrast based on the determination result of the document type automatic determination unit 13.

  The color correction unit 15 performs color correction to remove color turbidity based on spectral characteristics of CMY (C: cyan, M: magenta, Y: yellow) color materials including unnecessary absorption components in order to realize faithful color reproduction. The processing is performed. The image signal subjected to the color correction processing in the color correction unit 15 is transferred to the black generation and under color removal unit 16 and the region separation processing unit 20.

  The region separation processing unit 20 separates the color-corrected image signal into a character, halftone dot, or photo (other) region for each pixel based on the determination result of the document type automatic determination unit 13. To do. This region separation processing unit 20 sends a region identification signal indicating which region a pixel belongs to to the black generation and under color removal unit 16, the spatial filter processing unit 17, and the gradation reproduction processing unit 19 based on the separation result. Is output.

  In other words, the region separation processing unit 20 outputs, for each signal (each pixel) from which the document is read, information indicating whether the pixel belongs to a character, halftone dot, or photograph.

  The black generation and under color removal unit 16 performs black generation processing for generating a black (K) signal from the CMY three-color signals after color correction, while subtracting the K signal obtained by black generation from the original CMY signal. Undercolor removal processing for generating a new CMY signal is performed. As a result of these processes (black generation process / under color removal process), the CMY three-color signal is converted into a CMYK four-color signal.

  The spatial filter processing unit 17 performs spatial filter processing using a digital filter and corrects spatial frequency characteristics, thereby preventing blurring and graininess deterioration of the output image.

  The output tone correction unit 18 performs output tone correction processing for converting a signal such as a density signal into a halftone dot area ratio that is a characteristic value of the color image output device 3.

  The gradation reproduction processing unit 19 performs gradation reproduction processing (halftone generation processing) in which an image is finally divided into pixels and each gradation is reproduced.

  It should be noted that the image region extracted as a black character or in some cases a color character by the region separation processing unit 20 is high in the sharpness enhancement processing in the spatial filter processing unit 17 in order to improve the reproducibility of the black character or the color character. The amount of frequency frequency enhancement is increased. At the same time, in the halftone generation process, a binarization process or a multi-value process on a high resolution screen suitable for high frequency reproduction is selected.

  On the other hand, with respect to the region determined as a halftone dot by the region separation processing unit 20, the spatial filter processing unit 17 performs a low pass filter process for removing the input halftone dot component. At the same time, in the halftone generation process, a binarization process or a multi-value process is performed on the screen with an emphasis on gradation reproducibility.

  Further, for the region separated into photographs by the region separation processing unit 20, binarization or multi-value processing is performed on the screen with an emphasis on gradation reproducibility.

  For the area determined to be a photograph by the area separation processing unit 20, the spatial filter processing unit 17 performs a filtering process in which smoothing and sharpening are mixed in order to remove noise generated at the time of capturing an image and to make the subject clear. Applied. At the same time, in the halftone generation process, a binarization process or a multi-value process is performed on the screen with an emphasis on gradation reproducibility.

  As described above, the image data subjected to the above-described processes is temporarily stored in a storage unit (not shown), read at a predetermined timing, and input to the color image output device 3. The above processing is performed by a CPU (Central Processing Unit). The color image output device 3 outputs image data onto a recording medium (for example, paper). Examples of the color image output device 3 include a color image forming device using an electrophotographic method or an ink jet method. It is not limited.

  As described above, in the image processing apparatus 2 configured as described above, each image processing unit (the black generation and under color removal unit 16, the spatial filter processing unit 17, the tone reproduction process) that sequentially processes each pixel of the image data. When the outputs of the document type automatic discrimination unit 13 and the area separation processing unit 20 are input to the unit 19), each image processing pixel is “text & photo (printer output photo) document”. The processing is performed based on information such as “character pixels in” or “photo pixels in text & photo (photographic paper photo) document”.

  Next, image processing (image type recognition processing) in the document type automatic discrimination unit 13 which is a feature point in the present embodiment will be described.

  First, as shown in FIG. 1, the document type automatic discrimination unit 13 includes a character pixel detection unit 21, a background background pixel detection unit 22, a halftone pixel detection unit 23, a photo candidate pixel detection unit 24, and a photo candidate. The pixel count unit 25, the halftone pixel count unit 26, and the photo type determination unit 27 are configured. FIG. 8 is a flowchart showing the flow of processing in the automatic document type discrimination section 13.

  The character pixel detection unit 21 outputs an identification signal as to whether or not each pixel of the input image data exists in the character edge region. For example, as the processing of the character pixel detection unit 21, the input image data (f (0,0) to f (2,2) stored in the block memory as shown in FIG. For example, the following convolution calculation processing results S1 and S2 using filter coefficients as shown in FIGS. 3B and 3C are used.

  When S is larger than a preset threshold value, the pixel of interest (coordinate (1, 1)) in the input image data stored in the block memory is identified as a character pixel existing in the character edge region. By applying the above process to all the pixels of the input image data, the character pixels in the input image data can be identified.

  The background background pixel detection unit 22 outputs an identification signal as to whether or not each pixel of the input image data exists in the background background area. For example, as the processing of the background background pixel detection unit 22, a density histogram representing the frequency of each pixel density value (for example, G signal of RGB signal) of the input image data as shown in FIGS. There is something used.

A specific processing procedure in the background background pixel detection unit 22 will be described below with reference to FIGS.
Procedure 1: Detect maximum frequency (Fmax).
Procedure 2: When Fmax is smaller than a preset threshold value (THbg), it is assumed that no background background area exists in the input image data.
Procedure 3: When Fmax is equal to or greater than a preset threshold value (THbg), the pixel density values close to the pixel density value (Dmax) to be Fmax, for example, the frequencies Fn1 and Fn2 for the pixel density values of Dmax-1 and Dmax + 1 are used. When the sum of Fmax, Fn1, and Fn2 (shaded portion in FIG. 4A) is larger than a preset threshold, it is assumed that a background background area exists in the input image data.
Step 4: When a background background area exists in Step 3, a pixel having a pixel density value near Dmax, for example, a pixel density value from Dmax-5 to Dmax + 5 is identified as a background background pixel existing in the background background area. .

  Further, the density histogram may be a simple density histogram using density divisions (for example, pixel density values of 256 gradations divided into 16 density divisions) instead of the pixel density values. Alternatively, the luminance Y may be obtained by the following formula and a luminance histogram may be used.

  The halftone pixel detection unit 23 outputs an identification signal as to whether or not each pixel of the input image data exists in the halftone dot region. For example, as the processing of the halftone pixel detection unit 23, the input image data (f (0,0) to f (4,4)) stored in the block memory as shown in FIG. Some of them use the following adjacent pixel difference value sum Busy and maximum density difference MD for the pixel density value).

  Here, the Busy and the MD are used for identifying whether or not the pixel of interest (coordinates (2, 2)) is a halftone dot pixel existing in the halftone dot area.

  In the two-dimensional plane with the Busy and the MD as axes, as shown in FIG. 5B, the halftone pixels have a distribution different from the pixels (characters and photographs) existing in other areas. By performing threshold processing using the boundary line (dotted line) shown in FIG. 5B on the Busy and the MD obtained for each target pixel of the image data, each target pixel becomes a halftone dot area. Is identified.

  An example of the threshold processing is shown below.

  By applying the above process to all the pixels of the input image data, it is possible to identify halftone pixels in the input image data.

  The photograph candidate pixel detection unit 24 outputs an identification signal as to whether or not each pixel of the input image data exists in the photograph candidate pixel area. For example, the character pixels identified by the character pixel detection unit 21 in the input image data and the pixels other than the background background pixels identified by the background background pixel detection unit 22 are identified as photo candidate pixels.

  For example, FIG. 6A shows an example of input image data, and FIG. 6B shows a character pixel identified by the character pixel detector 21 for the input image data shown in FIG. A photograph composed of a character area composed of the above, a background ground area composed of the background ground pixels identified by the background ground pixel detection unit 22, and a photograph candidate pixel identified by the photograph candidate pixel detection unit 24 An example of a halftone dot area constituted by a candidate area and a halftone dot identified by the halftone pixel detection unit 23 is shown.

  The photograph candidate pixel counting unit 25 counts the number of photograph candidate pixels identified by the photograph candidate pixel detection unit 24. If the photo candidate pixel is a pixel other than the character pixel and the background background pixel, the photo candidate pixel number Np is expressed as follows using the total pixel number Na, the character pixel number Nt, and the background background pixel number Nb in the input image data. It is calculated by the following formula.

Np = Na-Nt-Nb
The total number of pixels can be obtained, for example, by counting the output signal of the character pixel detection unit regardless of the identification result.

  The halftone pixel counting unit 26 counts the number of halftone pixels identified by the halftone pixel detection unit 23.

  The photograph type determination unit 27 outputs the photograph candidate area from a printed photograph (halftone dot), a photographic paper photograph (continuous tone), or a printer output photograph (laser beam printer, inkjet printer, thermal transfer printer, etc.) (Photo). For example, as shown in FIGS. 6C, 6D, and 6E, the determination is made by the following conditional expression using the photograph candidate pixel number Np, the halftone pixel number Ns, and preset thresholds THr1 and THr2. Is what you do.

  Examples of the threshold value include THr1 = 0.7, THr2 = 0.3, and the like.

  The determination result may be output in pixel units, area units, or document units. Further, in the above processing example, the type determination target is only a photograph, but it is also possible to target document components other than characters and background backgrounds, such as graphics and graphs.

  In addition, the photo type determination unit 27 does not perform the determination of a printed photo, a print output photo, or a photographic paper photo, but compares the ratio of the halftone pixel number Ns to the photo candidate pixel number Np with a preset threshold value. Based on the above, control may be performed so as to switch processing contents such as color correction processing and spatial filter processing.

  As shown in FIG. 2, the result of the determination of the document type by the automatic document type determination unit 13 is the result of the color correction unit 15 connected to the subsequent stage of the automatic document type determination unit 13 and the black generation under color removal. Are output to the unit 16, the spatial filter processing unit 17, the gradation reproduction processing unit 19, and the region separation processing unit 20. Here, the document type determination result indicates that the region with the document is a photo candidate region. If the color correction unit 15, the spatial filter processing unit 17, and the gradation reproduction processing unit 19 determine that each of the photo types (printed photo, photographic paper photo, printer output photo) is as follows, Different processing is performed.

  An example of processing applied to the photo candidate area when the photo type determination unit 27 determines that the photo is a printed photo is shown below.

  Color correction process: In order to faithfully reproduce the color of a printed photograph, a process considering the color material of the printed photograph is performed.

  Spatial filter processing: Smoothing processing is performed to reduce moire (interference fringes) generated in a printed photograph. An example of the filter coefficient in this case is shown in FIG.

  Gradation reproduction processing: When performing dither processing, moire is strongly generated in proportion to the size of the matrix, and therefore dither processing using a relatively small matrix is performed.

An example of processing applied to the photo candidate area when the photo type determination unit 27 determines that the photo is a photographic paper photo is shown below.
Color correction processing: In order to faithfully reproduce the color of the photographic paper photograph, processing is performed in consideration of the color material of the photographic paper photograph.

  Spatial filter processing: Edge enhancement processing for correcting image blur is performed. An example of the filter coefficient in this case is shown in FIG.

  Gradation reproduction processing: Since moire does not occur, dither processing using a relatively large matrix that emphasizes gradation is performed.

  An example of processing applied to the photo candidate area when the photo type determination unit 27 determines that the photo is a printer output photo is shown below.

  Color correction processing: In order to faithfully reproduce the color of the printer output photograph, processing is performed in consideration of the color material used in the printer.

  Spatial filter processing: smoothing / enhancement mixed filter processing is performed to suppress graininess caused by dots constituting the printer output photograph while correcting blur of the image. An example of the filter coefficient in this case is shown in FIG.

  Gradation reproduction processing: Since moire does not occur, dither processing using a relatively large matrix that emphasizes gradation is performed.

  In the above, it is not determined whether or not a photo is included, but the above-described processing may be performed after determining whether or not a photo is included. In this case, when the photo candidate pixel count Np / Na with respect to the total number of pixels is larger than a predetermined threshold THpa (for example, 0.05), it can be determined that the photo area is included.

  Further, when the output of the document type automatic discrimination unit 13 is a discrimination result of a component other than the photo type, for example, a document component (character, photo, text & photo, etc.), the input tone correction unit 14 The processing content of the color removal unit 16 is switched as follows.

When the document type automatic determination unit 13 determines that the document is a character document,
Input gradation correction: In order to make the character easier to read, gradation correction is performed to lower the density of the highlight portion corresponding to the background.
Black generation under color removal: In order to eliminate the color blur of the black character portion, the black generation rate and the under color removal rate are increased.

When the automatic document type determination unit 13 determines that the document is a text & photo document or photo document
Input gradation correction: Performs near-linear gradation correction that can faithfully reproduce the gradation in the highlight portion of the photograph.
Black generation and under color removal: Processing is performed with an appropriate black generation rate and background removal rate that enables color reproduction faithful to the original document.

  Here, the flow of the image type recognition process in the document type automatic discrimination unit 13 having the above configuration will be described below with reference to the flowchart shown in FIG.

  First, based on RGB density signals converted from RGB signals (RGB reflectance signals) from which various distortions have been removed by the shading correction unit 12, character pixel detection processing (step S1), background background pixel detection processing (Step S2), halftone pixel detection processing (Step S3) is performed simultaneously. Here, the character pixel detection process is performed in the character pixel detection unit 21 described above, the background background pixel detection process is performed in the background background pixel detection unit 22 described above, and the halftone pixel detection process is performed as described above. Since it is performed in the pixel detection unit 23, the details of these processes are omitted.

  Next, photo candidate pixel detection processing is performed based on the processing result in the character pixel detection processing and the processing result in the background background pixel detection processing (step S4). The photograph candidate pixel detection process here is performed in the above-described photograph candidate pixel detection unit 24, and therefore the details of the process are omitted.

  Subsequently, based on the processing result in the photograph candidate pixel detection process, a process of counting the number of photograph candidate pixels Np is performed (step S5). The photo candidate pixel count processing here is performed in the above-described photo candidate pixel counting unit 25, and thus the details of the processing are omitted.

  In parallel with the processing from steps S1 to S5, processing for counting the number of halftone pixels Ns is performed based on the result of the halftone pixel detection processing in step S3 (step S6). Since the halftone pixel count processing here is performed in the halftone pixel counting section 26 described above, the details of the processing are omitted.

  Next, based on the photograph candidate pixel number Np obtained in step S5 and the halftone pixel number Ns obtained in step S6, the ratio of the halftone pixel number Ns to the photograph candidate pixel number Np, that is, Ns / Np is calculated. (Step S7).

  Subsequently, it is determined from the Ns / Np obtained in step S7 whether it is a printed photograph, a printer output photograph, or a photographic paper photograph (step S8).

  Since the processes in steps S7 and S8 are performed in the above-described photo type determination unit 27, details of the processes are omitted.

  In the above description, character pixel information is used as a parameter when performing image type recognition processing. Thus, in the image type recognition, if the information of the character pixel is used, the recognition accuracy can be improved.

  However, when the ratio of character pixels included in the document is small, that is, in the case of a document whose main is a photograph, the recognition accuracy of the photo type is recognized out of the image type recognition without using the character pixel information. Has little effect.

  Therefore, the automatic document type determination unit 30 as shown in FIG. 9 may be used in which the character pixel detection unit 21 is omitted from the automatic document type determination unit 13 shown in FIG.

  That is, as shown in FIG. 9, the document type automatic determination unit 30 includes a background background pixel detection unit 22, a halftone pixel detection unit 23, a photo candidate pixel detection unit 31, a photo candidate pixel count unit 25, A halftone pixel count unit 26 and a photo type determination unit 27 are included. The members having the same reference numerals as those of the automatic document type discrimination unit 13 shown in FIG. 1 have the same functions, and thus the description thereof is omitted.

  Further, the flow of processing in the automatic document type discrimination section 30 is substantially the same except that step S1 for performing character pixel detection processing is not performed in the flowchart shown in FIG.

The photograph candidate pixel detection unit 31 outputs an identification signal as to whether or not each pixel of the input image data exists in the photograph candidate pixel region, similarly to the photograph candidate pixel detection unit 24 shown in FIG. Therefore, since the pixel information is not used, the photo candidate pixel detection process is performed so that pixels other than the background background pixel identified by the background background pixel detection unit in the input image data are identified as photo candidate pixels. Is called.
[Embodiment 2]
Another embodiment of the present invention will be described as follows. In addition, about the member which has the same function as the said Embodiment 1, the same code | symbol is attached and the description is abbreviate | omitted.

  The color image forming apparatus according to the present embodiment includes an automatic document type determination unit 40 shown in FIG. 10 instead of the automatic document type determination unit 13 shown in FIG. 1 of the first embodiment.

  As shown in FIG. 10, the automatic document type discrimination unit 40 includes a character pixel detection unit 21, a background background pixel detection unit 22, a halftone pixel detection unit 23, a photo candidate pixel detection unit 24, and a photo candidate pixel. It comprises a labeling unit 41, a photo candidate pixel count unit 42, a halftone pixel count unit 43, and a photo type determination unit 27. FIG. 12 is a flowchart showing a flow of processing in the automatic document type discrimination section 40.

  As shown in FIG. 11A, the photo candidate pixel labeling unit 41 is composed of photo candidate pixels identified by the photo candidate pixel detecting unit 24 for input image data having a plurality of photo units. By performing a labeling process on a plurality of photo candidate areas, as shown in the photo candidate area (1) and the photo candidate area (2) shown in FIG. The photo candidate area is identified as a different area. Here, the photo candidate area is (1), and the other is (0), and the labeling process is applied in units of one pixel. Details of the labeling process will be described later.

  Unlike the photo candidate pixel counting unit 25 of the first embodiment, the photo candidate pixel counting unit 42 counts the number of pixels for a plurality of photo candidate areas labeled by the photo candidate pixel labeling unit 41, respectively. It is.

  The halftone pixel counting unit 43 counts the number of pixels for the halftone dot area identified by the halftone pixel detection unit 23 for each photo candidate area labeled by the photo candidate pixel labeling unit 41. is there. For example, as shown in FIG. 11B, the halftone dot pixel counting unit 43 sets the halftone dot area (halftone dot area (1)) existing in the photo candidate area (1) and the number of pixels Ns1 and the photo candidates. The number of pixels Ns2 constituting the halftone dot area (halftone dot area (2)) existing in area (2) is counted.

  The photo type determination unit 27 determines whether each of the photo candidate areas is a printed photo (halftone dot), a photographic paper photo (continuous tone), or a printer output photo. For example, as shown in FIGS. 11 (c) and 11 (d), the determination is made according to the following conditional expression using the photograph candidate pixel count Np, the halftone dot pixel count Ns, and preset thresholds THr1 and THr2. is there.

  In FIG. 11C, the photo candidate area (1) satisfies the condition 1 and is determined as a print photo, and the photo candidate area (2) is determined as the print output photo area because the condition 2 is satisfied. Further, in FIG. 11D, the photo candidate area (1) satisfies the condition 3 and is therefore determined as a photographic paper photograph, and the photo candidate area (2) is determined as the print output photo area because the condition 2 is satisfied. .

  Here, the flow of the image type recognition process in the document type automatic discrimination unit 40 having the above configuration will be described below with reference to the flowchart shown in FIG.

  First, based on the RGB density signal converted from the RGB signal (RGB reflectance signal) from which various distortions have been removed by the shading correction unit 12 (see FIG. 1), the character pixel detection process (step S11), Background background pixel detection processing (step S12) and halftone pixel detection processing (step S13) are performed simultaneously. Here, the character pixel detection process is performed in the character pixel detection unit 21 described above, the background background pixel detection process is performed in the background background pixel detection unit 22 described above, and the halftone pixel detection process is performed as described above. Since it is performed in the pixel detection unit 23, details of these processes are omitted.

  Next, photo candidate pixel detection processing is performed based on the processing result in the character pixel detection processing and the processing result in the background background pixel detection processing (step S14). The photograph candidate pixel detection process here is performed in the above-described photograph candidate pixel detection unit 24, and therefore the details of the process are omitted.

  Subsequently, a labeling process is performed on the detected photograph candidate pixels (step S15). Details of the labeling process will be described later.

  Subsequently, a process of counting the number of photograph candidate pixels Np is performed based on the processing result in the labeling process (step S16). Since the photo candidate pixel count processing here is performed in the above-described photo candidate pixel counting section 42, the details of the processing are omitted.

  In parallel with the processing of steps S11 to S16, processing for counting the number of halftone pixels Ns is performed based on the result of the halftone pixel detection processing in step S13 (step S17). Since the halftone pixel count processing here is performed in the halftone pixel counting section 43 described above, the details of the processing are omitted.

  Next, based on the photograph candidate pixel number Np obtained in step S16 and the halftone pixel number Ns obtained in step S17, the ratio of the halftone pixel number Ns to the photograph candidate pixel number Np, that is, Ns / Np is calculated. (Step S18).

  Subsequently, from the Ns / Np obtained in step S18, it is determined whether it is a printed photograph, a printer output photograph, or a photographic paper photograph (step S19).

  Since the processes in steps S18 and S19 are performed in the above-described photo type determination unit 27, the details of the processes are omitted.

  Here, the labeling process described above will be described.

  In general, the labeling process is a process in which the same label is assigned to a block of connected foreground pixels (= 1) and different connected components are assigned different connected components (image processing standard textbook CG-ARTS Association p). 262-268). Various labeling processes have been proposed. In this embodiment, a method using two scans will be described. The flow of this labeling process will be described below with reference to the flowchart shown in FIG.

  First, pixel values are examined in the raster scan order from the upper left pixel (step S21). When the target pixel value is 1, it is determined whether the upper adjacent pixel is 1 and the left adjacent pixel is 0 (step S21). Step S22). Perform the following process.

  Here, in step S22, when the upper adjacent pixel is 1 and the left adjacent pixel is 0, the following procedure 1 is executed.

  Procedure 1: As shown in FIG. 14 (a), when the target pixel is 1, if the pixel adjacent to the processing pixel is 1 and the label (A) is already attached, the processing pixel has the same label ( A) is added (step S23). Then, the process proceeds to step S29, where it is determined whether or not labeling has been completed for all pixels. Here, if all the pixels have been completed, the process proceeds to step S16 shown in FIG. 12, and the number of photograph candidate pixels Np is counted for each photograph candidate area.

  In step S22, if the upper adjacent pixel is 1 and the left adjacent pixel is not 0, it is determined whether the upper adjacent pixel is 0 and the left adjacent pixel is 1 (step S24).

  Here, when the upper adjacent pixel is 0 and the left adjacent pixel is 1 in step S24, the following procedure 2 is executed.

  Procedure 2: As shown in FIG. 14C, when the upper adjacent pixel is 0 and the left adjacent pixel is 1, the same label (A) as the left adjacent pixel is attached to the processing pixel (step S25). Then, the process proceeds to step S29, where it is determined whether or not labeling has been completed for all pixels. Here, if all the pixels are completed, the process proceeds to step S16 shown in FIG. 12, and the number Np of photo candidate pixels is counted for each photo candidate area.

  In step S24, if the upper adjacent pixel is 0 and the left adjacent pixel is not 1, it is determined whether the upper adjacent pixel is 1 and the left adjacent pixel is 1 (step S26).

  Here, in step S26, when the upper adjacent pixel is 1 and the left adjacent pixel is 1, the following procedure 3 is executed.

  Step 3: As shown in FIG. 14B, when the left adjacent pixel is 1 and a different label (B) from the upper adjacent pixel is attached, the same label (A) as the upper adjacent pixel is recorded. At the same time, it is maintained that there is a correlation between the label (B) in the left adjacent pixel and the label (A) in the upper adjacent pixel (step S27). Then, the process proceeds to step S29, where it is determined whether or not labeling has been completed for all pixels. Here, if all the pixels are completed, the process proceeds to step S16 shown in FIG. 12, and the number Np of photo candidate pixels is counted for each photo candidate area.

  In step S26, when the upper adjacent pixel is 1 and the left adjacent pixel is not 1, the following procedure 4 is executed.

  Procedure 4: As shown in FIG. 12 (d), when both the upper and left neighbors are 0, a new label (C) is attached (step S28). Then, the process proceeds to step S29, where it is determined whether or not labeling has been completed for all pixels. Here, if all the pixels are completed, the process proceeds to step S16 shown in FIG. 12, and the number Np of photo candidate pixels is counted for each photo candidate area.

When a plurality of labels are recorded, the labels are unified based on the above rules.

In the above description, character pixel information is used as a parameter when performing image type recognition processing. Thus, in the image type recognition, if the information of the character pixel is used, the recognition accuracy can be improved.

  However, when the ratio of character pixels included in the document is small, that is, in the case of a document whose main is a photograph, the recognition accuracy of the photo type is recognized out of the image type recognition without using the character pixel information. Has little effect.

  Accordingly, the automatic document type determination unit 50 as shown in FIG. 15 may be used in which the character pixel detection unit 21 is omitted from the automatic document type determination unit 40 shown in FIG.

  That is, as shown in FIG. 15, the document type automatic discrimination unit 50 includes a background background pixel detection unit 22, a halftone pixel detection unit 23, a photo candidate pixel detection unit 31, a photo candidate pixel labeling unit 41, and a photo. A candidate pixel counting unit 42, a halftone pixel counting unit 43, and a photo type determination unit 27 are configured. The members having the same reference numerals as those in the automatic document type discrimination section 40 shown in FIG. 10 have the same functions, and thus the description thereof is omitted.

  Further, the flow of processing in the automatic document type discrimination section 50 is substantially the same except that step S11 for performing character pixel detection processing is not performed in the flowchart shown in FIG.

  The photographic candidate pixel detection unit 31 outputs an identification signal as to whether or not each pixel of the input image data exists in the photographic candidate pixel region, similarly to the photographic candidate pixel detection unit 24 shown in FIG. Therefore, since the pixel information is not used, the photo candidate pixel detection process is performed so that pixels other than the background background pixel identified by the background background pixel detection unit in the input image data are identified as photo candidate pixels. Is called.

  In each of the above embodiments, the configuration shown in FIGS. 1 and 10 may be used to determine the type of the entire image, not just the photographic area. In this case, an image type determination unit 51 is provided after the photo type determination unit 27 (see FIGS. 16 and 17). In the image type determination unit 51, the ratio Nt / Na of the number of character pixels Nt with respect to the total number of pixels Na, the ratio (Np−Ns) / Na of the difference between the number of photo candidate pixels Np and the number of halftone pixels Ns with respect to the total number of pixels Na, The ratio Ns / Na of the halftone dot number Ns to the total number of pixels Na is obtained and compared with predetermined threshold values THt, THp, THs, and the type of the entire image is determined based on the result of the photo type determination unit 27. I do. For example, if the ratio Nt / Na of the number of character pixels Nt to the total number of pixels Na is equal to or greater than the threshold value and the result of the photo type discrimination unit 27 is a printer output photo, it is determined that the original is a mixed document of characters and printer output photos Is done.

  Although the method for automatically recognizing the document type has been described above, the above processing may be performed when the user selects the character / photo mode or the photo mode representing the document type from the operation panel. good.

  Further, the document type determination method according to the present invention may be realized as software (application program). In this case, a printer driver incorporating software that realizes processing based on the document type discrimination result can be provided in the computer or printer.

  As an example of the above, processing based on the document type discrimination result will be described below with reference to FIG.

  As shown in FIG. 18, the computer 101 includes a printer driver 102, a communication port driver 103, and a communication port 104. The printer driver 102 includes a color correction unit 105, a spatial filter processing unit 106, a gradation reproduction processing unit 107, and a printer language translation unit 108. The computer 101 is connected to a printer (image output device) 109, and the printer 109 outputs an image according to image data output from the computer 101.

  The image data generated by executing various application programs in the computer 101 is processed by the color correction unit 105, the spatial filter processing unit 106, and the gradation reproduction processing unit 107 based on the document type determination result. Made. In this case, the color correction unit 105 includes black generation and under color removal processing.

  The image data subjected to the above processing is converted into a printer language by the printer language translation unit 108 and input to the printer 109 via the communication port driver 103 and a communication port (for example, RS232C / LAN) 104. The printer 109 may be a digital multifunction machine having a copy function and a fax function in addition to the printer function.

  The present invention can also record an image processing method for performing document type determination processing on a computer-readable recording medium in which a program to be executed by a computer is recorded.

  As a result, it is possible to portablely provide a recording medium on which a program for performing an image processing method for determining an original type and performing an appropriate process based on the result is recorded.

  The recording medium may be a non-illustrated memory, for example, a program medium such as a ROM because processing is performed by a microcomputer, and a program reading device as an external storage device (not illustrated) is provided, and the recording medium is stored therein. It may be a program medium that can be read by being inserted.

  In any case, the stored program may be configured to be accessed and executed by the microprocessor, and the program is read out, and the read program is stored in a program storage area (not shown) of the microcomputer. A method of downloading and executing the program may be used. In this case, it is assumed that the download program is stored in the main device in advance.

  Here, the program medium is a recording medium configured to be separable from the main body, such as a tape system such as a magnetic tape or a cassette tape, a magnetic disk such as a flexible disk or a hard disk, and a CD-ROM / MO / MD / DVD. Semiconductors such as optical discs, IC cards (including memory cards) / optical cards, etc., or mask ROM, EPROM (Erasable Programmable Read Only Memory), EEPROM (Electrically Erasable Programmable Read Only Memory), flash ROM, etc. It may be a medium that carries a fixed program including a memory.

  In this case, since the system configuration is capable of connecting a communication network including the Internet, the medium may be a medium that fluidly carries the program so as to download the program from the communication network. When the program is downloaded from the communication network in this way, the download program may be stored in the main device in advance or installed from another recording medium.

  The recording medium is read by a program reading device provided in a digital color image forming apparatus or a computer system, whereby the above-described image processing method is executed.

  The computer system includes an image input device such as a flatbed scanner, a film scanner, and a digital camera, a computer that performs various processes such as the image processing method by loading a predetermined program, and displays the processing results of the computer. An image display device such as a CRT display / liquid crystal display and a printer that outputs the processing results of the computer to paper. Furthermore, a network card, a modem, and the like are provided as communication means for connecting to a server or the like via a network.

  In addition, if the system configuration is capable of connecting to a communication network including the Internet, the recording medium is preferably a recording medium that fluidly carries the program so as to download the program from the communication network.

  Further, when the program is downloaded from the communication network as described above, it is preferable that the download program is stored in the main device in advance or installed from another recording medium.

  The image processing apparatus of the present invention may have the following configuration.

  That is, the image processing apparatus of the present invention includes an image type recognizing unit that extracts a feature amount from input image data and identifies the type of the input image. The image type recognizing unit includes each of the input image data. A halftone pixel detection means for determining whether a pixel or a block composed of a plurality of pixels belongs to a halftone dot area, and whether a block composed of each pixel or a plurality of pixels of the input image data belongs to a background background area Detected by the halftone dot detection means for the pixels or blocks in the input image data other than the background background pixels or background background blocks detected by the background background pixel detection means. Depending on the ratio of halftone dot pixels or halftone dot blocks, the input image data may be printed photographs (halftone dots), photographic paper photographs (continuous tone), Printer output photos may be provided for determining photographic type determination unit either is contained in (printer photo output in).

  Thus, by utilizing the fact that the characteristics of various printer output photographs are located between the characteristics of the printed photograph and the characteristics of the photographic paper photograph, the halftone pixel and background background pixel information obtained by the existing pixel attribute detection means It is possible to determine whether or not a printer output photograph is included by a relatively simple method using. In particular, when it is desired to determine whether a printer output photograph is included with a simpler configuration, since the ratio of character pixels in the input image is often low, the identification accuracy is almost the same even without character pixel information. Utilizing the fact that it can be maintained to the extent, the construction means can be simplified by omitting the character pixel detection means.

  Also, the photo type determination means does not identify the photo area of the input image data as a printed photo (dot), a photographic paper photo (continuous tone), or a printer output photo, and calculates the ratio of the number of pixels or blocks calculated. Based on the above, it may be controlled to switch the contents of the image processing.

  Here, the block for performing the attribute determination is not limited to the rectangular area, and may have any shape.

  The image processing apparatus having the above-described configuration further includes character pixel detection means for determining whether each pixel of the input image data belongs to a character area, and the photo type determination means is detected by the character pixel detection means. A character pixel or character block and a halftone dot pixel detected by the halftone pixel detector for a pixel or block in the input image data other than the background background pixel or background background block detected by the background background pixel detector or Depending on the ratio of halftone dots, it may be determined whether the input image data includes a printed photograph (halftone dot), a photographic paper photograph (continuous tone), or a printer output photograph.

  As a result, when the identification accuracy of the printer output photo is more important than simplification of the configuration means, an input image other than the character pixel or character block and the background background pixel or background background block can be obtained by adding the character pixel detection means. Pixels or blocks in the data can be identified, that is, the ratio of halftone pixels or halftone blocks in only the photos or figures that are originally intended to be excluded, excluding characters, can be used for more accurate printer output photos. Can be determined.

  Further, a pixel or block in the input image data other than a background background pixel or background background block detected by the background background pixel detection means, or a character pixel or character block detected by the character pixel detection means and the background Labeling means for labeling a photo candidate area composed of pixels or blocks in the input image data other than the background background pixels or background background blocks detected by the background pixel detection means, and the photo type determination means For each labeled photo candidate area, the photo candidate area is a printed photo (halftone dot) according to the ratio of the halftone pixel or halftone block detected by the halftone pixel detection means to the pixels or blocks in the area. ), Photographic paper photo (continuous tone), or printer output photo. It may be.

  As a result, for pixels other than background background pixels or blocks other than background background blocks, or pixels other than character pixels and background background pixels, or blocks other than character blocks and background background blocks (photo candidate areas) in the input image data By labeling each of the above candidate photo areas, it is possible to individually identify whether the photograph is a printed photograph (halftone dot), a photographic paper photograph (continuous tone), or a printer output photograph. Become. That is, even when a print photo and a printer output photo are present in one input image data, it is possible to recognize the type of each photo.

  When the image processing apparatus having the above-described configuration is provided in an image forming apparatus, the following effects can be obtained.

  Image data that includes the printer output photo by applying image processing that considers the characteristics of the printer output photo to the input image data that includes the printer output photo or the pixels and blocks to which the printer output photo belongs. High-quality output is possible.

  The image processing method in the above image processing apparatus has the following steps.

  That is, in an image processing method including an image type recognition step that extracts a feature amount from input image data and identifies the type of the input image, the image type recognition step includes a block composed of each pixel or a plurality of pixels of the input image data. A halftone dot pixel detecting step for determining whether or not a pixel belongs to a halftone dot area, and a background background pixel for determining whether or not a block comprising each pixel or a plurality of pixels of the input image data belongs to a background background area A halftone pixel or halftone block detected in the halftone pixel detection step for a pixel or block in the input image data other than the background background pixel or background background block detected in the detection step and the background background pixel detection step; Depending on the ratio of the input image data, a print photograph (halftone dot), a photographic paper photograph (continuous tone), and a printer output photograph (pre- It is contemplated and a determining photographic type determination step of determining one is included in the output photographic) in data.

  In addition, a character pixel detection step for determining whether each pixel of the input image data belongs to a character area, the photo type determination step includes the character pixel or the character block detected in the character pixel detection step, According to the ratio of the halftone dot pixel or halftone block detected in the halftone pixel detection step to the pixel or block in the input image data other than the background background pixel or background background block detected in the background background pixel detection step, It may be possible to determine whether the input image data includes a printed photograph (halftone dot), a photographic paper photograph (continuous tone), or a printer output photograph.

  Further, a pixel or block in the input image data other than the background background pixel or background background block detected in the background background pixel detection step, or the character pixel or character block detected in the character pixel detection step and the background A labeling step for labeling a photo candidate area composed of pixels or blocks in the input image data other than the background background pixel or background background block detected in the background pixel detection step, and the photo type determination step For each labeled photo candidate area, the photo candidate area is a printed photo (halftone dot) according to the ratio of the halftone pixel or halftone block detected in the halftone pixel detection step to the pixels or blocks in the area. ), Photographic paper photo (continuous tone), or printer output photo. It is conceivable to.

  As an image processing program for operating the above-described image processing apparatus, a computer may function as each of the above-described means.

  In this case, the image processing apparatus can be realized by realizing each unit of the image processing apparatus with a computer.

  The image processing program may be recorded on a computer-readable recording medium.

  In this case, the image processing apparatus can be realized on a computer by an image processing program read from the recording medium.

An image processing apparatus according to the present invention is an image processing apparatus that extracts a feature amount from input image data and performs image processing according to the extracted feature amount.
Halftone pixel detection means for determining whether or not each pixel of the input image data or a block composed of a plurality of pixels belongs to a halftone dot area;
A background background pixel detection means for determining whether a block consisting of each pixel or a plurality of pixels of the input image data belongs to a background background area;
The ratio of the halftone pixel or halftone block detected by the halftone pixel detection means to the pixel or block in the input image data other than the background background pixel or background background block detected by the background background pixel detection means is input. It is characterized by being extracted as a feature amount of image data.

  According to the above configuration, the ratio of halftone pixels or halftone blocks to pixels or blocks in the input image data other than background background pixels or background background blocks in the input image data is extracted as the feature amount of the input image data. Thus, an output image that faithfully reproduces the input image can be obtained.

  The present invention is not limited to the above-described embodiments, and various modifications are possible within the scope shown in the claims, and embodiments obtained by appropriately combining technical means disclosed in different embodiments. Is also included in the technical scope of the present invention.

  The image processing method of the present invention can be applied to a digital color copying machine, but any other apparatus that is required to improve the reproducibility of image data that is input and output by image data. Even applicable. An example of such an apparatus is an image reading apparatus such as a scanner.

1, showing an embodiment of the present invention, is a block diagram illustrating a schematic configuration of a document type automatic determination unit (photo type determination unit) provided in an image processing apparatus. FIG. It is a block diagram which shows schematic structure of the said image processing apparatus. (A)-(c) is explanatory drawing which shows an example of the filter coefficient by which the character pixel detection part with which the said original type automatic discrimination | determination part was equipped is convolved with input image data in order to detect a character pixel. is there. (A) (b) is explanatory drawing which shows an example of the density histogram used in order to detect a background background pixel in the background background detection part with which the said original classification automatic discrimination | determination part was equipped. (A) and (b) are used for calculating feature amounts (adjacent pixel difference value sum, maximum density difference) for detecting halftone pixels in the halftone pixel detection unit provided in the automatic document type discrimination unit. It is explanatory drawing which shows an example of distribution of a block memory and the said feature-value. (A) to (e) are the results of the photo candidate pixel detection unit, the photo candidate pixel count unit, the halftone pixel detection unit, and the halftone pixel count unit provided in the document type automatic discrimination unit for the input image data example. It is explanatory drawing which shows an example. It is explanatory drawing which shows an example of the filter coefficient applied to the various photographs based on the determination result of the said photograph classification determination part. 3 is a flowchart showing a flow of processing of a document type automatic determination unit (photo type determination unit) provided in the image processing apparatus. 14 is a block diagram illustrating a schematic configuration of an automatic document type determination unit (photo type determination unit) included in the image processing apparatus according to another embodiment of the present invention. FIG. 14 is a block diagram illustrating a schematic configuration of an automatic document type determination unit (photo type determination unit) included in the image processing apparatus according to another embodiment of the present invention. FIG. (A) to (e) are a photo candidate pixel detection unit, a photo candidate pixel labeling unit, a photo candidate pixel count unit, and a halftone pixel detection unit provided in the automatic document type discrimination unit shown in FIG. 10 for an example of input image data. It is explanatory drawing which shows an example of the result of a halftone pixel count part. FIG. 11 is a flowchart showing a flow of processing of a document type automatic determination unit (photo type determination unit) shown in FIG. 10. 11 is a flowchart showing a flow of processing of a labeling unit provided in the automatic document type discrimination unit shown in FIG. 10. (A)-(d) is explanatory drawing which shows an example of the processing method of the said labeling part. 14 is a block diagram illustrating a schematic configuration of an automatic document type determination unit (photo type determination unit) included in the image processing apparatus according to another embodiment of the present invention. FIG. 1 is a block diagram illustrating an example of an apparatus to which an image processing apparatus of the present invention is applied, and illustrating a schematic configuration of an original document type automatic determination unit (image type determination unit) provided in the image processing apparatus. FIG. 7 is a block diagram illustrating another example of an apparatus to which the image processing apparatus of the present invention is applied, and illustrating a configuration of an original type automatic determination unit (image type determination unit) provided in the image processing apparatus. It is a block diagram which shows the structure of the said image processing apparatus at the time of implement | achieving the image processing method of this invention as software (application program).

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Color image input device 2 Image processing device 3 Color image output device 4 Operation panel 11 A / D conversion part 12 Shading correction part 13 Document type automatic discrimination | determination part (image type discrimination | determination means)
DESCRIPTION OF SYMBOLS 14 Input gradation correction part 15 Color correction part 16 Black production | generation under color removal part 17 Spatial filter process part 18 Output gradation correction part 19 Tone reproduction process part 20 Area separation process part 21 Character pixel detection part (character pixel detection means)
22 Background background pixel detection unit (background background pixel detection means)
23 Halftone pixel detection unit (halftone pixel detection means)
24 Photo Candidate Pixel Detection Unit 25 Photo Candidate Pixel Counting Unit 26 Halftone Pixel Counting Unit 27 Photo Type Determination Unit 30 Document Type Automatic Determination Unit (Image Type Determination Unit)
31 Photo candidate pixel detection unit 40 Document type automatic determination unit (image type determination unit)
41 Photo candidate pixel labeling section (labeling means)
42 Photo Candidate Pixel Count Unit 43 Halftone Pixel Count Unit 50 Document Type Automatic Discrimination Unit (Image Type Discrimination Unit)
51 Image Type Determination Unit 101 Computer 102 Printer Driver 103 Communication Port Driver 104 Communication Port 105 Color Correction Unit 106 Spatial Filter Processing Unit 107 Tone Reproduction Processing Unit 108 Printer Language Translation Unit 109 Printer Na Total Number of Pixels Np Number of Photo Candidate Pixels Ns Ratio Nt Number of character pixels

Claims (10)

In an image processing apparatus including an image type recognition unit that extracts a feature amount from input image data and identifies the type of the input image,
The image type recognition means includes
Halftone pixel detection means for determining whether or not each pixel of the input image data or a block composed of a plurality of pixels belongs to a halftone dot area;
Background background pixel detection means for determining whether or not each pixel of the input image data or a block composed of a plurality of pixels belongs to the background background area;
From the ratio of the halftone pixel or halftone block detected by the halftone pixel detection means to the pixel or block in the input image data other than the background background pixel or background background block detected by the background background pixel detection means , Determines whether the image characteristics of a printed photograph consisting of halftone dots, the image characteristics of a continuous-tone photographic paper photograph, or the image characteristics of a printer output photograph output by a printer, and the input image data is determined from the determined image characteristics. An image processing apparatus comprising: a photo type determining unit that determines a type of a photo included in the image. Character pixel detection means for determining whether each pixel of the input image data belongs to a character area,
The photograph type determination means includes a character pixel or a character block detected by the character pixel detection means, a pixel in the input image data other than the background background pixel or background background block detected by the background background pixel detection means, or Based on the ratio of halftone pixels or halftone blocks detected by the halftone pixel detection means to the block, the image characteristics of a printed photograph composed of halftone dots, the image characteristics of a continuous-tone photographic paper photograph, and the printer output output by a printer The image processing apparatus according to claim 1, wherein the image processing apparatus determines which of the image characteristics is a photograph, and determines a type of the photograph included in the input image data from the determined image characteristics . Further, a pixel or block in the input image data other than the background background pixel or background background block detected by the background background pixel detection means, or a character pixel or character block detected by the character pixel detection means and the background Labeling means for labeling a photo candidate area composed of pixels or blocks in the input image data other than the background background pixels or background background blocks detected by the background pixel detection means,
The photograph type determination means determines the halftone dot from the ratio of the halftone pixel or halftone block detected by the halftone pixel detection means to the pixel or block in the area for each photo candidate area labeled by the labeling means. The image characteristics of the print candidate area , the image characteristics of the continuous-tone photographic paper photograph, and the image characteristics of the printer output photo output by the printer are determined. The image processing apparatus according to claim 1 , wherein the type of the image processing apparatus is determined.   An image forming apparatus comprising the image processing apparatus according to claim 1. In an image processing method including an image type recognition step of extracting a feature amount from input image data and identifying the type of the input image,
In the image type recognition process,
A halftone pixel detection step of determining whether or not each pixel of the input image data or a block composed of a plurality of pixels belongs to a halftone dot area;
A background background pixel detection step for determining whether each pixel of the input image data or a block composed of a plurality of pixels belongs to a background background area;
From the ratio of the halftone pixel or halftone block detected in the halftone pixel detection step to the pixel or block in the input image data other than the background background pixel or background background block detected in the background background pixel detection step , Determines whether the image characteristics of a printed photograph consisting of halftone dots, the image characteristics of a continuous-tone photographic paper photograph, or the image characteristics of a printer output photograph output by a printer, and the input image data is determined from the determined image characteristics. An image processing method comprising: a photo type determination step for determining a type of a photo included in the image. A character pixel detection step of determining whether each pixel of the input image data belongs to a character area;
The photo type determination step includes a character pixel or a character block detected in the character pixel detection step and a pixel in the input image data other than the background background pixel or background background block detected in the background background pixel detection step. Based on the ratio of halftone pixels or halftone blocks detected in the halftone pixel detection step to the block, the image characteristics of a printed photograph composed of halftone dots, the image characteristics of a continuous-tone photographic paper photograph, and the printer output output by a printer 6. The image processing method according to claim 5, wherein it is determined which of the image characteristics of the photo is determined, and the type of the photo included in the input image data is determined from the determined image characteristics . Further, a pixel or block in the input image data other than the background background pixel or background background block detected in the background background pixel detection step, or a character pixel or character block detected in the character pixel detection step and the background A labeling step for labeling a photo candidate area composed of pixels or blocks in the input image data other than the background background pixel or background background block detected in the background pixel detection step,
The photo type determination step includes a halftone dot from a ratio of halftone pixels or halftone blocks detected in the halftone pixel detection step to pixels or blocks in the area for each photo candidate area labeled by the labeling step. The image characteristics of the print candidate area , the image characteristics of the continuous-tone photographic paper photograph, and the image characteristics of the printer output photo output by the printer are determined. The image processing method according to claim 5 , wherein the type of the image is determined. In the photo type determination method for determining the photo type included in the input image data,
From the ratio of halftone pixels or halftone blocks contained in the input image data to the background background pixels or pixels other than the background background blocks contained in the input image data, or the continuous tone, The image characteristics of the photographic paper photograph and the image characteristics of the printer output photograph output by the printer are determined, and the type of the photograph included in the input image data is determined from the determined image characteristics A photo type determination method characterized by the above.   An image processing program for operating the image processing apparatus according to any one of claims 1 to 3, wherein the image processing program causes a computer to function as each of the means.   A computer-readable recording medium on which the image processing program according to claim 9 is recorded.

Images