JP2009302761A - Image processor - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To perform adaptive processing to an edge portion of characters and line drawing, in an image processor for determining characters and a line drawing part with respect to an inputted image. <P>SOLUTION: An image processor uses a method for determining an area of arbitrary width adjacent to an inner side or an outer side of character edges by extracting an inner edge area and an outer edge area of edges of characters and line drawing and expanding the inner edge area in the outer direction of the edges, and the outer edge area in the inner direction of the edges. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to an image processing apparatus, and more particularly to an image processing apparatus that detects a character / line drawing portion of an input image and performs adaptive image processing on the portion.

  As this type of apparatus, an electrophotographic recording apparatus is generally known.

  Such a recording apparatus includes a photoelectric conversion element such as a CCD in a reading unit, and converts the image on the original into an electric signal by scanning the photoelectric conversion element. Then, a latent image is formed on the photosensitive drum by driving a laser or LED based on the electric signal obtained by the reading, and the toner is developed at a place where a potential difference is generated by the latent image, and this is recorded. It is transferred and reproduced on a recording medium such as paper.

  In the above-described apparatus, when the read image is reproduced on paper due to image blurring due to lens aberration, CCD aperture, or the like during reading, or deterioration of the reproduced image due to process factors such as transfer / fixing. In addition, there is a problem that the MTF of the image is deteriorated.

  In order to correct the image quality deterioration of the character part due to the MTF deterioration as described above, many characters and line drawing parts are determined and image processing (for example, edge enhancement etc.) optimal for the determined part is proposed. (For example, Patent Document 1).

  As another conventional example, Patent Document 2 can be cited.

In the following description, as described above, characteristic detection / determination and attribute assignment for each image unit such as pixels in the original image are performed for each image unit according to the image area determination and the determination result of the image area determination. The adaptive processing is called image area processing, and both are collectively referred to as image area separation or image area separation processing.
Japanese Unexamined Patent Publication No. 04-142880 JP 2005-176222 A

  However, the deterioration of MTF in electrophotography is very complicated, and even if only the edge of the character / line drawing part is determined and subjected to the optimum processing, good image quality cannot be obtained.

  In the conventional method, there are a method that can determine only one or two pixels of the edge, or a method that determines a large area. However, such a method requires a very large amount of memory, and is very costly. It was.

  For example, when it is desired to print a black character area only with black toner, a process in which only one or two pixels of the edge portion of the black character can be determined as a character cannot perform a desired process on the entire black character area. Moreover, a slightly thick character has a problem that it looks like a pseudo contour when processing extremely different between the inner side and the edge portion of the character (hereinafter referred to as edge edging).

  In order to solve such a problem, Patent Document 2 emphasizes / extracts a desired spatial frequency using a digital filter, and determines the result by using two threshold values (positive and negative), thereby obtaining an inner edge. The area inside the character is determined by enlarging the inner edge signal in the opposite direction to the outer edge signal by using a means for obtaining the signal and the outer edge signal and integrating the signals in the determined areas. A method has been proposed.

  However, in the image forming apparatus that performs image area separation, the problem that occurs around the edge is not limited to the inside of the edge as described above.

  For example, if the color of the character is higher than the color of the surrounding background part, such as in a document where black characters are placed on a gray background, applying an emphasis filter to the edge of the character will affect the effect. Thus, there is a problem that an undershoot as shown in FIG. 2A is generated in the base region in contact with the outside of the edge, and the density is reproduced at a low level (hereinafter referred to as white spots).

  In addition, a color output device that reproduces a plurality of colors has a problem that has the opposite effect. As shown in FIG. 2D, a high density of magenta (hereinafter referred to as “M”) and yellow (hereinafter referred to as “Y”) with high density in a background of low density cyan (hereinafter referred to as “C”). If the character and background color are reproduced with different hues, such as in a document with red characters, the emphasis filter is applied to the edge of the character and the background touches the outside of the edge. As shown in FIG. 2C, there is a problem that overshoot occurs in the region and the density is reproduced high (hereinafter referred to as reverse enhancement).

  According to the present invention, in an image forming apparatus that performs image area separation, adaptive determination is performed on each of the determined areas by appropriately determining the areas inside and outside the edge peripheral area, and This is to solve not only a problem that occurs in the area adjacent to the inside of the edge, such as a border, but also a problem that occurs in an area adjacent to the outside of the edge, such as white spots or reverse enhancement. An image processing apparatus of the present invention for solving such a problem has the following configuration shown in FIG.

  That is, in an image processing apparatus that performs character / line drawing determination on an input image signal, filter processing means for performing filter processing on input image data, and an image filtered by the filter processing means as a threshold value Outer edge clipping processing means for obtaining an outer edge area by clipping in, an inner edge clipping processing means for obtaining an inner edge area by clipping the filtered image with a threshold value, and the outer edge area in the character internal direction Area extension processing means for extending the inner edge area in a character outward direction, a character edge area and a character using the outer edge area and the inner edge area expanded by the area extension processing means as a criterion. Non-character edge area adjacent to the outside of the edge area and inside the character edge area Flag determination unit determines the non-character edge region adjacent, and having a city.

  As described above, according to the present invention, in an image forming apparatus that performs image area separation, it is possible to extract inner and outer areas adjacent to edges of characters, line drawings, and the like. By performing adaptive image area processing, it is possible to solve the problems of image quality degradation such as white spots and reverse enhancement that occur around edges of characters and line drawings due to image area separation.

  Next, details of the present invention will be described in accordance with the description of the embodiments.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

  A configuration example of an electrophotographic color MFP (Multi Function Printer / Multi Function Peripheral) will be described as an example of an image processing apparatus applicable as an embodiment of the present invention. Note that this embodiment can be applied not only to an electrophotographic color MFP but also to an inkjet or other monochrome and color image forming apparatus.

  First, the configuration of an electrophotographic color MFP (Multi Function Peripheral) will be described with reference to the block diagram of the MFP shown in FIG.

  In the figure, an input image processing unit reads a paper document or the like with an image reading device such as a scanner, and performs image processing on the read image data.

  The NIC (Network Interface Card) unit passes image data (mainly PDL data) input using the network to the RIP unit, and transmits image data and device information inside the MFP to the outside via the network. To do. The RIP part is a part that decodes the input PDL (Page Description Language) data and develops the RIP (Raster Image Processor).

  Next, the input image data is sent to the MFP control unit. The MFP control unit plays a role of traffic control for controlling input data and output data.

  Also, the image data input to the MFP control unit is temporarily stored in the memory unit. The stored image data is temporarily stored or called as necessary.

  The output image processing unit performs image processing for printing and sends it to the printer unit.

  The printer unit feeds a sheet and sequentially prints the image data created by the output image processing unit on the sheet. The printed sheet is sent to the post-processing unit, where sheet sorting processing and sheet finishing processing are performed.

  Furthermore, the operation unit is for selecting the above-described various flows and functions and instructing operations. As the resolution of the display device of the operation unit is increased, the image data in the memory unit is previewed. If it is OK after confirmation, it can be used for printing.

  As described above, the MFP has various functions and usage methods, examples of which are shown below.

A) Copy function: Input image processing unit → Output image processing unit → Printer unit
B) Network scan: Input image processing unit → NIC unit
C) Network print: NIC unit → RIP unit → output image processing unit → printer unit
D) Box scan function: Input image processing unit → Output image processing unit → Memory unit
E) Box print function: Memory part → Printer part
F) Box reception function: NIC unit → RIP unit → output image processing unit → memory unit
G) Box transmission function: Memory part → NIC part
H) Preview function: memory unit → operation unit Next, the input image processing unit will be described.

  FIG. 4 is a block diagram showing the flow and configuration of image data in the scanner input unit.

  The image read by the color scanner is converted into an electrical signal by the CCD sensor. This CCD sensor is an RGB 3-line color sensor and is input to the A / D converter as image data of R (Red), G (Green), and B (Blue).

  In the A / D converter, after gain adjustment and offset adjustment, each color signal is converted into 8-bit digital image data.

  The shading correction unit corrects the variation in sensitivity of each pixel of the CCD sensor and the variation in the amount of light of the document illumination lamp for each color using the read signal of the reference white plate.

  Furthermore, since each color line sensor of the CCD sensor is arranged at a predetermined distance from each other, a spatial delay in the sub-scanning direction (paper feed direction) that comes from the CCD RGB arrangement at the color offset unit is a line delay. In addition to correction by the adjustment circuit, a sub MTF (Modulation Transfer Function) correction unit corrects deviations caused by sensor unevenness and the like, and in particular, contrast reduction caused by narrowing the interval between white and black Also corrects blurring of images.

  The input gamma correction unit is a one-dimensional lookup table (LUT) that performs correction so that the exposure amount and the luminance have a linear relationship with respect to each RGB input.

  The input direct mapping unit is a three-dimensional LUT that converts an input RGB signal into an RGB signal in the device to unify the color space. The read color space determined by the spectral characteristics of the RGB filter of the CCD sensor is converted to sRGB. It is a part that converts to a standard color space such as, and can absorb various characteristics such as sensitivity characteristics of a CCD sensor and spectral characteristics of an illumination lamp.

  The main MTF correction unit realizes MTF correction in the main scanning direction by taking a weighted average for the pixel of interest and each of the left and right pixels.

  The image area determination unit includes a character determination unit and an achromatic color determination unit. The former performs pixel determination of a character area or a line drawing outline area from RGB image data, and outputs the result as image area data ZA. In the latter case, image data in the RGB space is converted into image data in a luminance color difference space such as L * a * b * or YUV, and then a low saturation region, a black character, or a color character region is determined, and the result is obtained as image region data ZB. Output as. Here, the image area data ZA and ZB are collectively referred to as image area data.

  Next, the spatial filter unit performs filtering on the luminance component using the image area data. Specifically, edge enhancement is performed on the character area, and different sharpness filters such as smoothing or weak edge enhancement are applied to the photo area.

  The saturation suppression unit performs saturation suppression on an area determined as low saturation from the image area data. In order to represent a black character area with a single color of K (Black), processing is performed with a color difference of zero.

  An ACS (Auto Color Select) unit can count low-saturation pixels from the image area determination result and determine whether the image is a monochrome image or a color image.

  In the BE (background removal) sampling unit, in order to detect the background, pixels in a designated rectangular area are discretely sampled to create a luminance histogram, which is used for background removal at the time of printing.

  Next, a concept and an example of the image area separation processing unit will be described.

  The image area separation process is for extracting a feature of an original image and generating a signal indicating an image area attribute (hereinafter referred to as flag data) in order to perform optimum image processing according to the feature of the image included in the original image. To be done. For example, in an original, various image areas such as a continuous-tone full-color photo area, a black-colored character area, or a halftone dot printing area such as newspaper printing are usually mixed.

  If these are uniformly processed by the same image processing procedure and output, the output image generally does not often have a desirable image quality.

  Therefore, in this system, the attribute of the image data included in the document image is detected using the color image signal input from the input image processing unit 301, and flag data for identifying the attribute is generated. A specific procedure is shown in FIG.

  This figure is a block diagram for explaining an example of an image area attribute extraction method. The input signal 501 is input to the average density calculation 502 and edge enhancement processing 503.

  In the average density calculation 502, an average value AVE of an M × N region (M and N are natural numbers) centered on the target pixel is calculated for the input signal.

  In the edge enhancement processing 503, the edge enhancement processing of the pixel of interest is performed in the M × N region (M and N are natural numbers) by referring to the data of the region around the pixel of interest, and two types of edge enhancement signals EDG1 and EDG2 having different intensities are used. Is calculated.

  The results of average density calculation 502 and edge enhancement processing 503 are input to halftone dot determination 504 and edge determination 505.

  The edge determination 505 will be described with reference to FIG. The results of the average density calculation 502 (AVE) and edge enhancement processing 503 (EDG2) are input to the density difference determination 601. At this time, the magnitude of the difference between the EDG2 output from the edge enhancement processing 503 and the density (AVE) of the surrounding area is determined, and 1 is output if it is larger than the predetermined density difference, and 0 is output if it is smaller.

If AVE-A × EDG2> B, density difference signal = 1
If A × EDG2-AVE> C, density difference signal = 1
Otherwise, density difference signal = 0
(A is a real number, B and C are real numbers or integers)
The density difference signal output from the density difference determination 601 is input to the isolation determination 602, and the isolated points are removed. For example, when the density difference signal is referenced 7 × 7 and there is no density difference signal = 1 at the outermost pixel position, the density difference signal in the inner 5 × 5 region is forcibly set to 0. , To remove isolated points.

  The isolation determination signal output from the isolation determination 602 is input to the correction process 603, and the discontinuous portion is corrected. For example, if the isolation determination signal of the pixel (front / rear, left / right, diagonal) sandwiching the target pixel (3 × 3 center) in the 3 × 3 region is 1, and the target pixel is 0, the value of the target pixel is corrected to 1. To do. As a result, the continuity of the portion lacking the line drawing is increased, and the edge signal 604 is generated with a smooth line drawing. If it is an edge, EDGE = 1 is output.

  The results of the halftone dot determination signal and the edge determination signal 604 output from the halftone dot determination 504 and the edge determination 505 are input to the character determination 506. Here, processing is performed on a pixel-by-pixel basis, and not a halftone dot (AMI = 0) but an edge (EDGE = 1) pixel is considered as a character, and a character signal 507 is generated.

  The above is an example of the image area attribute extraction method, but is not limited thereto.

  When the attribute of the image is detected for each pixel by the above image area separation processing, image processing corresponding to the image attribute is performed in the subsequent input image processing unit.

  Here, for example, the high-frequency component of the image is emphasized for the character region to enhance the sharpness of the character, and the so-called low-pass filter processing is performed for the dot region to remove the moire component peculiar to the digital image. Can be performed. These processes can be switched in units of pixels according to the attribute flag data generated by the image area determination unit.

  Next, the most characteristic part of the present invention will be further described with reference to another drawing.

  FIG. 1 is a block diagram showing the data flow and configuration of the edge determination unit 205 to which the present invention is applied. FIG. 7 schematically shows an image and density characteristics imitating a character edge portion of an input image, a value characteristic obtained for each processing flow, and an intermediate processed image.

  The input image 102 is subjected to a second-order differential filter by the filter processing unit 102. The differentiated image is clipped by the outer edge clipping processing unit 103 and the inner edge clipping processing unit 104 with a predetermined threshold. As a result, as shown at 705, an outer portion (hereinafter referred to as an outer edge region) and an inner portion (hereinafter referred to as an inner edge region) of the edge portion of the character / line drawing are obtained as intermediate flag data.

  The clipped image data is expanded in the area extension processing unit 105 as indicated by reference numeral 706, the outer edge area is expanded in the inner edge direction, the inner edge area is expanded in the outer edge direction, and integration determination is performed. The vicinity outside area 722, the character edge vicinity inside area 723, and the other image area 724 are determined.

  In the region expansion process, for example, the flag data of the target pixel is determined by the number of pixels having flag data in a 5 × 5 area.

  For example, in the case of expansion of the outer edge region, the integration determination is performed according to the following conditions.

The target pixel is the outer edge region. → Outer edge area ・ Other than the above, there are a predetermined number or more of outer edge areas and a predetermined number or more of inner edge areas in the 5 × 5 area around the target pixel. → Outer edge area ・ None of the above. → Non-outer edge region The region expansion processing is not limited to the above example, and a combination or repetition thereof may be performed.

  The flag determination unit determines flag data based on a plurality of types of intermediate flag data obtained for each pixel. FIG. 8 shows a table used when determining flag data from the intermediate flag data of the extended outer edge region and the extended inner edge region. As a result, a character edge region, a character edge peripheral outer region, a character edge peripheral inner region, and other image regions are discriminated.

  FIG. 9A shows the pixel value characteristics of the edge portion of the input image where white spots occur, and FIG. 9B shows the input image of FIG. 9A subjected to edge enhancement processing to which the present invention is not applied. FIG. 9C schematically shows the pixel value characteristics subjected to the edge enhancement processing to which the present invention is applied.

  10A is a pixel value characteristic of an edge portion of an input image where reverse enhancement occurs, and FIG. 10B is an edge enhancement process in which the present invention is not applied to the input image of FIG. 10A. FIG. 10C schematically shows the pixel value characteristics subjected to the edge enhancement processing to which the present invention is applied.

  FIG. 14 shows a one-dimensional table of convolution operations used for edge enhancement performed in FIGS.

  The generated flag data can be used in the subsequent stage of the input image processing unit or the output image processing unit. In the input image processing unit, as shown in FIG. 11, the pixel value of the pixel having the flag data 722 in the outer peripheral area of the character edge is changed to the pixel value of the pixel having the flag data 724 in the image area existing in the 5 × 5 area. The process of replacing with the average value of is performed. As a result, as shown in FIG. 9C and FIG. 10C, the outer area around the character edge is made uniform with the outer image area, so that white spots and reverse enhancement due to overshoot and undershoot are prevented. A good image as shown in FIGS. 2B and 2D can be obtained.

  In this embodiment, an example of improving the image quality of the outer area around the character edge, which is an area outside the character adjacent to the character edge, is shown, but the present invention is not limited to this, and the inner area around the character edge. It is also possible to perform predetermined image processing on the image. Further, the outer and inner regions around the character edge can have an arbitrary width by designing a threshold value or a coefficient for region expansion processing.

  In the first embodiment, the flag data is generated by referring to the two intermediate flag areas of the outer edge area and the inner edge area of the character edge. However, if the edge sharpness is low and there is a width between the inner edge region and the outer edge region, a large memory is required.

  In this embodiment, a mode that does not require a large-capacity memory that can be applied even in such a situation will be described.

  FIG. 12 is a block diagram showing the data flow and configuration of the edge determination unit 205 in this embodiment.

  A difference from the first embodiment is that a second filter processing unit 1208 that performs a first-order differential filter is provided, and a character edge clipping unit 1209 generates a character edge region as an intermediate flag by clipping with a threshold value. This is a point to be referred to when the processing unit 1205 performs extension processing. That is, when there is an intermediate flag generated after finishing each clipping process in the positional relationship as shown in FIG. 13, it is possible to extend the inner and outer edge regions under the following conditions.

The target pixel is the outer edge region. → Outer edge region ・ In addition to the above, the target pixel is a character edge region, and there are a predetermined number or more of outer edge regions in the 5 × 5 area around the target pixel. → Outer edge area ・ None of the above. → Non-outer edge area The image area processing is the same as in the first embodiment.

  In this embodiment, the region expansion processing is performed by combining the processing results of the primary differential filter and the secondary differential filter. However, even in the character edge portion where the edge amount is low, the region expansion is performed even if the reference area is narrow. Thus, the memory capacity can be reduced.

The block diagram which shows the data flow and structure of the edge determination part in the image processing apparatus to which this invention is applied. Examples of white spots and reverse enhancement and output examples to which the present invention is applied Structure of electrophotographic color MFP Flow and configuration of image data in the scanner input section Block diagram for explaining an example of an image area attribute extraction method Edge judgment unit flow Image simulating character edge and density characteristics and processing result Reference table for judgment Pixel value characteristics of an image with white spots and the result of edge enhancement Pixel value characteristics of an image with inverse enhancement and the result of edge enhancement Color replacement according to flag data Data flow of edge determination unit in embodiment 2 Positional relationship between character edge area and inner / outer edge area Convolution table used for edge enhancement

Claims (5)

  In an image processing apparatus that performs character / line drawing determination on an input image signal, filter processing means for performing filter processing on input image data, and clipping an image filtered by the filter processing means with a threshold value Outer edge clipping processing means for obtaining an outer edge area by clipping, an inner edge clipping processing means for obtaining an inner edge area by clipping the filtered image with a threshold value, and extending the outer edge area in the direction of the inside of the character Area extension processing means for extending the inner edge area in a character outward direction, a character edge area and a character edge area using the outer edge area and the inner edge area expanded by the area extension processing means as a criterion Adjacent to the outside of the non-character edge area and the inside of the character edge area Flag determination unit determines a character edge area, the image processing apparatus characterized by having a city.   The image processing apparatus according to claim 1, wherein the filter processing unit includes at least one secondary differential filter.   3. The image according to claim 1, wherein the region expansion processing unit has an image processing system in addition to the filter processing unit, and performs region expansion processing with reference to a result thereof. Processing equipment.   The image processing apparatus according to claim 3, wherein the flag determination unit performs determination with reference to a result of image processing according to claim 3.   Separately from the filter processing means and the outer edge clipping processing means and the inner edge clipping processing means, a second area is obtained by a second filter processing means and a third clipping processing means. The image processing apparatus according to claim 1, wherein region expansion processing is performed with reference to the outer edge region, the inner edge region, and the third region.

Images