JP2013026865A - Image processing apparatus and image processing method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To correct displacement with reduced deterioration in image quality.SOLUTION: An image processing apparatus 1 comprises: a correction value calculation part 23 which acquires the amount of the displacement of an image printed on a sheet by printing processing of image data, divides the image into objects, and calculates the correction value of the displacement in accordance with the acquired amount of displacement with respect to each divided object; and a correction processing part 24 which corrects the image data with the correction value calculated with respect to each object and corrects the displacement object by object.

Description

  The present invention relates to an image processing apparatus and an image processing method.

  When printing with a printer or copier, misalignment may occur in the image printed on the paper due to bending or expansion / contraction of the paper, deviation in the conveyance direction of the paper, curvature or inclination of the printing unit, and the like. Conventionally, it is specified in advance how much misalignment occurs, and the image data is subjected to a correction process for deforming the image so that the position is shifted in the opposite direction to the specified misalignment, thereby correcting misalignment during printing. Offsetting was done.

  In the case of an electrophotographic system, pseudo halftones are often expressed by gradation processing such as screen processing. As such pseudo-halftone image data correction processing, a pixel interpolation method such as a nearest neighbor method (nearest neighbor method) is known. Also, a correction process is disclosed in which an image is shifted in units of pixels in a direction opposite to the position shift at a position where the position shift occurs (for example, see Patent Documents 1 and 2).

JP 2000-253231-A JP 2001-245156 A

As disclosed in Patent Documents 1 and 2, when an image is shifted in units of pixels, a step may occur due to the shift, which causes image quality deterioration. An image including an edge or an outline, such as a character or a figure, is particularly easy to see a step, and the image quality is likely to deteriorate. Even if the image does not include an edge or an outline, a step is easily visible even when the resolution of the printer is not high.
In addition, pixel shifts cause insertion or deletion of pixels, causing noise and unevenness.

  According to Patent Documents 1 and 2, the visibility of the step is lowered by performing smoothing processing on the image data in which the step is generated and adjusting the density near the step. However, in addition to the misalignment correction process, a smoothing process must be performed, which requires time and cost. Steps that do not match the detection pattern are not subject to the smoothing process, so there is a possibility that steps will remain.

  An object of the present invention is to correct misalignment with little image quality degradation.

According to the invention of claim 1,
The amount of displacement of an image printed on paper by image data printing processing is acquired, the image is divided into objects, and the position displacement is determined for each divided object according to the amount of displacement acquired. A correction value calculation unit for calculating the correction value of
A correction processing unit that corrects the image data according to the correction value calculated for each object, and corrects the displacement in units of objects;
An image processing apparatus is provided.

According to invention of Claim 2,
The correction value calculation unit calculates a correction value for the translation amount and deformation amount of each object from the displacement amount of the positional deviation,
For each object, the correction processing unit translates the object with the correction value for the parallel movement amount, deforms the object with the correction value for the deformation amount, or executes both the parallel movement and the deformation. An image processing apparatus according to claim 1 is provided.

According to invention of Claim 3,
The correction processing unit determines whether the deformation amount is large using a threshold,
When it is determined that the deformation amount is not large, the object is translated by the correction value of the parallel movement amount,
3. The image processing apparatus according to claim 2, wherein when the deformation amount is determined to be large, the object is translated by the correction value of the parallel movement amount, and the object is deformed by the correction value of the deformation amount. .

According to invention of Claim 4,
The image processing apparatus according to claim 3, wherein the correction processing unit changes the threshold value according to object attribute information.

According to the invention of claim 5,
The correction value calculation unit acquires displacement amounts of positional deviations of a plurality of detection patterns printed around each object, and calculates a correction value for a parallel movement amount and a deformation amount of each object from the deviation amounts. The image processing apparatus as described in any one of 2-4 is provided.

According to the invention of claim 6,
The image processing apparatus according to claim 1, wherein the correction value calculation unit extracts an object based on typesetting information or layout information of image data, and divides the image into each object. The

According to the invention of claim 7,
The image according to claim 1, wherein the correction value calculation unit extracts a closed region having the same attribute as an object based on attribute information of image data, and divides the image into each object. A processing device is provided.

According to the invention described in claim 8,
The correction value calculating unit detects image edges by processing image data, extracts a closed region in which the detected edges form a contour line as an object, and divides the image into each object. The image processing device according to any one of 7 is provided.

According to the invention of claim 9,
The image processing apparatus according to claim 1, wherein the correction value calculation unit extracts an area designated by a user as an object, and divides the image into each object.

According to the invention of claim 10,
The image processing apparatus according to any one of claims 1 to 9, wherein the misregistration includes color misregistration, misregistration between front and back surfaces, and misregistration.

According to the invention of claim 11,
Obtaining a displacement amount of an image printed on a sheet by a printing process of image data;
Dividing the image into object units, and calculating a correction value for the positional deviation according to the obtained deviation amount for each divided object;
Correcting the image data with the correction value calculated for each object, and correcting the positional deviation in units of objects;
An image processing method is provided.

  According to the present invention, since correction is performed in units of objects, the outline of the object is not disturbed or the image of the object is not lost due to the correction, and it is possible to correct misalignment with little deterioration in image quality.

It is a figure which shows the image processing apparatus which concerns on this Embodiment. (A) An example of color misregistration. (B) An example of a register mark used for color misregistration detection. (C) It is an example of a caliper pattern used for the detection of color misregistration. (A) It is an example of position shift of front and back. (B) It is an example of the register mark used for detection of the positional deviation of the front and back. (A) It is an example of the position shift by distortion. (B), (c) It is an example of the register mark used for the detection of the position shift due to distortion. 10 is a flowchart of processing executed by the image processing apparatus when correcting misalignment. It is a flowchart of the process which divides | segments an image into object units. FIG. 5 is a diagram of images before and after misalignment and registration marks arranged at a plurality of positions on a sheet. It is a flowchart of the process which calculates the correction value of the amount of translation and deformation. Shows a square object before and after displacement. The process of positional deviation of the front and back of an 8-sided image is shown. The process of correcting the positional deviation by the parallel movement of the object unit is shown. The process of correcting the positional deviation between the front and back of a large image and a small image only by translation is shown. An image having a large size indicates a process of correcting by translation and deformation. The timing at which the misalignment can be corrected is shown.

  Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 shows an image processing apparatus 1 according to the present embodiment.
The image processing apparatus 1 rasterizes PDL (Page Description Language) data input to the image processing apparatus 1 to generate image data, performs image processing on the image data, and outputs the image data to a printing unit (not shown). The PDL data is data obtained by converting data in an application file format created by an application such as a word processor or spreadsheet into a PDL format by a printer driver or the like. The image processing apparatus 1 can be mounted on a printer, a copier, a controller, or the like.

  As shown in FIG. 1, the image processing apparatus 1 includes a RIP (Raster Image Processor) 11, a memory 12, a screen processing unit 13, a memory 14, a writing processing unit 15, a detection pattern processing unit 21, a deviation detection unit 22, and a correction value. A calculation unit 23, a correction processing unit 24, a control unit 31, a display unit 32, and an operation unit 33 are provided.

  The RIP 11 generates image data by rasterization. The RIP 11 analyzes the command and PDL data and creates a display list and layout information. The display list includes information such as the size, color, and attribute of the object. The layout information includes information on the arrangement of objects on each page. The RIP 11 generates image data in which each object is developed into a bitmap in units of pages according to the display list and layout information. The RIP 11 generates attribute information indicating the attribute of each pixel of the image data based on the display list and outputs it together with the image data.

The memory 12 stores image data and attribute information output from the RIP 11.
The screen processing unit 13 performs screen processing on the image data held in the memory 12 and outputs the image data after the screen processing.

The memory 14 stores the image data output from the screen processing unit 13.
The writing processing unit 15 performs PWM (Pulse Width Modulation) conversion on the image data held in the memory 14. The converted PWM data is output to a printing unit (not shown). The printing unit executes a printing process using the PWM data, and prints an image on a sheet.

  The detection pattern processing unit 21 adds a detection pattern used for detection of misalignment to the screen-processed image data when performing misalignment correction. The memory 14 stores the image data to which the detection pattern is added.

  The deviation detection unit 22 reads an image printed on a sheet by a scanner or the like, and detects a positional deviation of the image. Examples of misregistration include color misregistration, front / back registration misalignment, and misregistration due to image distortion. The misalignment is caused by misalignment of the printing unit, distortion of the sheet itself, bending, misalignment of the sheet conveyance position, and the like.

  Color misregistration occurs when a color image is formed by superimposing images formed by a plurality of color toners. FIG. 2A shows an example of color misregistration, in which a cyan image and a magenta image that should be matched originally are printed with misalignment. As a detection pattern used for detecting such color misregistration, a cross-shaped image called a register mark as shown in FIG. 2B or a line image called caliper as shown in FIG. It is done. Since each color line is formed in each of the main scanning direction and the sub-scanning direction for both the registration marks and the calipers, the shift detection unit 22 can calculate the shift amount of the positional shift from the distance between the lines.

Misregistration occurs when an image is formed on the front and back of a sheet. FIG. 3A shows an example of misregistration, in which the front and back images that should be matched on the front and back are printed with their positions shifted.
As the misregistration detection pattern, registration marks printed on the front surface and the back surface can be used as shown in FIG. The deviation detection unit 22 measures the distances x1, y1, x2, and y2 from the printed paper edge to the registration marks for each of the front surface and the back surface, and detects the differences x1-x2 and y1-y2 as deviation amounts.

  Deviation due to distortion can occur due to the inclination of the paper during printing and distortion of the printing unit such as a polygon mirror. FIG. 4A is an example of a positional shift due to distortion, and an image that should originally be formed in a quadrangle is transformed into a parallelogram and printed. As a detection pattern for misalignment due to distortion, registration marks as shown in FIG. 3B are arranged near the four corners of the paper, printed, and the distance of the printed registration marks from the paper edge is measured. It is possible to detect a displacement due to distortion from the difference between the two. In addition, as shown in FIG. 4B, a plurality of register marks arranged at a plurality of positions on the sheet can be used. The registration mark shown in FIG. 4B is not a mere cross but has a characteristic shape. The vertical and horizontal lines constituting the cross differ in thickness and shape depending on the vertical and horizontal positions, and the points attached to the periphery of the cross differ depending on the position of the registration mark on the paper. According to such a characteristic register mark, the shift detection unit 22 generates a position shift on the sheet not only based on the position shift amount but also the position of the point from the partial information on the sheet. It is possible to specify in which direction the positional deviation occurs from the shape of the line. Therefore, it is convenient when, for example, a paper having a large size is read by a scanner smaller than the paper and a positional deviation is detected.

  If punching is possible after the printing process, a punch hole may be provided at the center of the register mark as shown in FIG. The misalignment detection unit 22 can detect misalignment from the edge position of the register mark notch, and can detect misregistration between the front and back sides in more detail.

The amount of deviation detected by the deviation detector 22 is stored in the storage unit 34 and used. If the quality and size of the paper used by the printing unit is constant, the degree of misalignment is substantially constant. Therefore, the misregistration amount stored in the storage unit 34 can be read and used as necessary. Since it is expected that the degree of misregistration will be different if the paper quality and size are different, the storage unit 34 stores the misregistration amount in association with the quality and size of a plurality of papers and other paper conditions. However, it is preferable to read the amount of deviation according to the paper conditions.
In the present embodiment, an example is shown in which the misalignment detection unit 22 detects misalignment. However, the misalignment amount may be measured separately using a position measuring device such as a digitizer, and the operator can visually determine the misalignment amount. You may observe. In the case of visual observation, the shift amount input by the operator from the operation unit 33 is stored in the storage unit 34 and used.

  The correction value calculation unit 23 divides the image to be printed into object units, and calculates a correction value for the positional deviation detected by the deviation detection unit 22 for each divided object.

  The correction value calculation unit 23 extracts an object based on the layout information or typesetting information of the image data, and divides the image into each object. Both layout information and typesetting information are information on the arrangement in one page. The layout information is information on the arrangement of characters, graphics, and photographs in one page specified by the application software, and the format information is information on the arrangement specified by the printer driver or the like at the time of printing. The typesetting information indicates, for example, the arrangement of images when a plurality of designated images are allocated in one page, or when a plurality of pages of images are reduced and allocated in one page. Since the layout information or the composition information is obtained from the PDL data, the correction value calculation unit 23 can obtain the layout information or the composition information from the RIP 11.

  The correction value calculation unit 23 can also extract a closed region having the same attribute as an object based on the attribute information of the image data, and divide the image into each object. The correction value calculation unit 23 can also process the image data to detect an edge, extract a closed region where the detected edge forms an outline as an object, and divide the image into each object.

  When the control information output from the control unit 31 instructs the correction value calculation unit 23 to extract a region designated by the user as an object, the correction value calculation unit 23 extracts the region as an object and divides the image into each object. .

  The correction value calculation unit 23 calculates a correction value for the parallel movement amount and the deformation amount of each object from the displacement amount output from the displacement detection unit 22, and supplies the correction value to the correction processing unit 24 as a correction value for the displacement. Output.

  The correction processing unit 24 corrects the original image data with the correction value calculated for each object, and corrects the positional deviation in units of objects.

  For each object, the correction processing unit 24 translates the object with the correction value of the parallel movement amount output from the correction value calculation unit 23 or moves the object with the correction value of the deformation amount output from the correction value calculation unit 23. Deform or perform both the translation and the deformation.

  The control unit 31 acquires read data of an image printed on a sheet from the deviation detection unit 22 and outputs the acquired data to the display unit 32. When a region of an object whose position deviation is to be corrected is designated by the user via the operation unit 33, the control unit 31 generates a control signal so as to extract the region as an object, and outputs the control signal to the correction value calculation unit 23.

The display unit 32 displays the data read by the deviation detection unit 22 on the display.
The operation unit 33 generates an operation signal corresponding to a user operation and outputs the operation signal to the control unit 31. A touch panel in which the display unit 32 and the operation unit 33 are integrated may be used.

  The storage unit 34 stores a shift amount of a positional shift of an image printed on a sheet. The amount of misalignment is stored for each condition such as the quality and size of the paper used by the printing unit for the printing process. As the storage unit 34, a RAM, a hard disk, a nonvolatile memory, or the like can be used.

FIG. 5 is a flowchart illustrating a process executed by the image processing apparatus 1 when correcting misalignment.
At the time of correcting the misalignment, the image data output from the RIP 11 is output to the screen processing unit 13 without being corrected by the correction processing unit 24, and a detection pattern is added by the detection pattern processing unit 21. Thereafter, the image data is printed by the printing unit via the writing processing unit 15. When the user reads the printed paper into the scanner of the deviation detection unit 22, the deviation detection unit 22 reads an image printed on the paper and analyzes the obtained read data to detect the amount of deviation of the positional deviation. (Step S1). At this time, the deviation detection unit 22 acquires position information on the sheet to which the detection pattern is added from the detection pattern processing unit 21, scans and analyzes only the area around the area to which the detection pattern is added, and detects the position deviation. Efficiency may be improved.

  For example, when register marks shown in FIG. 3B are added as detection patterns in order to detect misregistration between the front and back sides, the misalignment detection unit 22 analyzes the read data and detects the register marks printed on the front and back surfaces of the paper. The positions (x1, y1) and (x2, y2) of the sheet edge are specified, and the difference (x1−x2, y1−y2) is detected as the shift amount.

  The correction value calculation unit 23 acquires the amount of displacement from the displacement detection unit 22, and divides the image to be printed into object units (step S2). As described above, since the positional deviation generated in the printing unit is generally constant when the paper quality and size are constant, the deviation amount obtained in step S1 is stored in the storage unit 34 and stored. The correction value calculation unit 23 may acquire the amount of deviation from the storage unit 34.

  FIG. 6 is a flowchart of the process of dividing into object units. As shown in FIG. 6, the correction value calculation unit 23 extracts one object from the original image data (step T1). Since the object is initially not divided, the correction value calculation unit 23 treats the entire image as one object and extracts the entire image.

  If the original image data includes layout information or typesetting information (step T2), the correction value calculation unit 23 divides the extracted object based on the layout information or typesetting information (step T3). According to the layout information, it is possible to specify a character, graphic, or photographic object arranged in one page, so that the correction value calculation unit 23 divides the image for each object specified by the layout information. Also, according to the typesetting information, the image unit allocated in one page can be specified, so the correction value calculation unit 23 divides the image using the image unit allocated by the typesetting information as an object. After dividing the object, the correction value calculation unit 23 extracts one object from the divided objects (step T11; N, T12). The processes in steps T2 to T10 are repeated for the extracted object.

  Although there is no layout information or typesetting information (step T2; N), if the extracted object has attribute information (step T4; Y), the correction value calculation unit 23 divides the object based on the attribute information ( Step T5). The correction value calculation unit 23 extracts a closed region having the same attribute as an object, and divides it into a character attribute object, a graphic attribute object, and a photo attribute object. After dividing the object, the correction value calculation unit 23 extracts one object from the divided objects (step T11; N, T12). The processes in steps T2 to T10 are repeated for the extracted object.

  When there is no typesetting information, layout information, and no attribute information (step T4; N), the correction value calculation unit 23 processes the extracted object region to detect an edge (step T6). The edge can be detected by a known detection process such as a Laplacian filter process or a Sobel filter process. The correction value calculation unit 23 divides the closed region as an object (step T8) if the object has a closed region where the detected edge forms a contour line (step T7; Y). After dividing the object, the correction value calculation unit 23 extracts one object from the divided objects (step T11; N, T12). The processes in steps T2 to T10 are repeated for the extracted object.

  When there is no closed region in the object where the detected edge forms a contour line, but control information is output from the control unit 31 and an image region for correcting misalignment is designated by the user (step T7; N) , T9; Y), the correction value calculation unit 23 divides the region designated by the user as an object (step T10).

  If there is no area specification by the user (step T9; N), the correction value calculation unit 23 determines whether or not the processing of steps T2 to T10 has been completed for all objects (step T11). If not completed (step T11; N), the correction value calculation unit 23 extracts one object from the unprocessed objects (step T12). The processes in steps T2 to T10 are repeated for the extracted object. When all the processes are completed (step T11; Y), the process proceeds to step S3 in FIG.

The object division may be repeated until there are no more objects that can be divided, or the division may be terminated when the total number of divided objects reaches a predetermined number. Alternatively, the division of the object may be terminated when the size of the divided object reaches a predetermined value or less.
When there is an area designation by the user, the correction value calculation unit 23 may divide only the designated area as an object and end the process. In this case, only the area designated by the user is corrected for positional deviation.

  When the object division is completed, the correction value calculation unit 23 selects one object from the divided objects as shown in FIG. 5 (step S3). The correction value calculation unit 23 acquires a shift amount detected by the detection pattern located closest to the selected object among the shift amounts of the positional shift acquired from the shift detection unit 22 (step S4).

  When there are a plurality of detection patterns, the correction value calculation unit 23 acquires the shift amounts of the plurality of detection patterns printed around the object. For example, as shown in FIG. 7, when the detection pattern processing unit 21 arranges a plurality of registration marks as detection patterns at equal intervals on the paper, the shift detection unit 22 detects the shift amount of the positional shift of each registration mark. The correction value calculation unit 23 may acquire the shift amounts of the six registration marks positioned around each object as the shift amounts of the six objects.

The correction value calculation unit 23 calculates correction values for the parallel movement amount and deformation amount of the object from the acquired deviation amount (step S5).
FIG. 8 is a flowchart of the process for calculating the correction value. As shown in FIG. 8, the correction value calculation unit 23 calculates the position of the center of gravity of the selected object (step U1). Next, the correction value calculation unit 23 calculates the shift amount of the center of gravity position from the acquired shift amount of the detection pattern, and obtains the shift amount as a parallel movement amount (step U2). The correction value calculation unit 23 calculates a deviation amount in the opposite direction to the parallel movement amount as a correction value for the parallel movement amount (step U3).

  When a square object as shown in FIG. 9 is selected, the correction value calculation unit 23 calculates the gravity center position g1 from the position coordinates of the vertices m1, m2, m3, and m4 of the square. Since there are six registration marks around the object as the detection pattern, the correction value calculation unit 23 estimates the position of the object after the positional deviation from the deviation amount of the six registration marks. First, the correction value calculation unit 23 specifies six points m1, m2, m3, m4, m5, and m6 that are the minimum distance from each registration mark on the outline of the selected object. The correction value calculation unit 23 obtains points n1 to n6 obtained by shifting the specified six points m1 to m6 by the amount of registration mark corresponding to each point, and spatially interpolates these six points n1 to n6 to obtain a position shift. Estimated as a later object.

  The correction value calculation unit 23 calculates the centroid position g2 from the position coordinates of the four vertices n1, n2, n3, and n4 of the position-shifted object, and calculates the shift amount from the centroid position g1 to the centroid position g2 as a parallel movement amount. A deviation amount in the opposite direction to the deviation amount is calculated as a correction value for the parallel movement amount. For example, when the shift amount of the center of gravity position is −5 in the main scanning direction and 3 in the sub-scanning direction, the parallel movement amount is −5 in the main scanning direction and 3 in the sub-scanning direction, which is the same as the shift amount. The correction value of the movement amount is 5 in the main scanning direction and -3 in the sub scanning direction.

  The correction value calculation unit 23 calculates the shift amounts of the plurality of positions of the object from the shift amount of the detection pattern, and obtains a shift amount obtained by subtracting the parallel movement amount from the shift amount as a deformation amount (step U4). The correction value calculation unit 23 calculates a deviation amount in the direction opposite to the deformation amount as a deformation amount correction value (step U5). When the object is translated by the amount of translation, the object is corrected to the original position on average as a whole. However, the positional shift does not always occur by the same shift amount at all positions of the object, and the shift amount that cannot be corrected by the parallel movement needs to be corrected by deformation. Therefore, the difference between the shift amount of each position in the object and the translation amount is obtained as the deformation amount, and the shift amount in the opposite direction to the deformation amount is obtained as the correction value.

  In the case of the example of FIG. 9, the correction value calculation unit 23 determines the shift amounts of the six registration marks from the six points m1 to m6 of the object corresponding to each register mark to the shift points n1 to n6. Therefore, the amount of deformation is obtained by subtracting the amount of parallel movement from the amount of shift of each of the six registration marks. For example, if the translation amount is -5 in the main scanning direction and 3 in the sub-scanning direction, and the shift amount of the registration mark on the upper left is -6 in the main scanning direction and 5 in the sub-scanning direction, it corresponds to the registration mark on the upper left. The amount of deformation of the object to the upper left vertex m1 → n1 is −1 in the main scanning direction, 2 in the sub scanning direction, the correction value of the deformation amount is +1 in the main scanning direction, and −2 in the sub scanning direction.

  When the correction values for the parallel movement amount and the deformation amount are obtained, as shown in FIG. 5, the correction value calculation unit 23 determines whether the deformation amount is large using a threshold value (step S6). When a plurality of deformation amounts are obtained, it can be determined that the deformation amount is large when at least one of them exceeds a threshold value. In the example of FIG. 9, six deformation amounts are obtained from the six registration marks. Therefore, when at least one of the deformation amounts exceeds the threshold value, the correction value calculation unit 23 can determine that the deformation amount is large. Note that the method for determining whether or not the deformation amount is large is not limited to this, and when the average of a plurality of deformation amounts exceeds the threshold or when the degree of image processing according to the deformation amount is large (for example, rotation The image processing is necessary, and the amount of deformation may be determined to be large (for example, when the rotation angle is equal to or greater than the threshold).

  When it is determined that the deformation amount is not large (step S6; N), the correction value calculation unit 23 outputs only the correction value of the average movement amount to the correction processing unit 24. The correction processing unit 24 translates the selected object according to the average moving amount correction value output from the correction value calculating unit 23 (step S7).

  Usually, there is a sufficient distance between objects with respect to the correction value, but when the correction value is large and a collision occurs between objects due to parallel movement, the correction value calculation unit 23 has an object with a larger area or an object with a predetermined attribute. What is necessary is just to delete the collision part. Conversely, when an unnatural gap occurs between objects due to parallel movement, the correction value calculation unit 23 may enlarge an object having a larger area or an object having a predetermined attribute. The predetermined attribute is an attribute having a lower priority order. The priority is set higher in the order of the influence on the image quality degradation due to deletion or enlargement, for example, the priority is set higher in the order of character> figure> photograph. According to this priority, if a character object and a photo object collide, the collision part of a photo object with a low priority is deleted.

  On the other hand, when it is determined that the deformation amount is large (step S6; Y), the correction value calculation unit 23 outputs the correction value for the deformation amount to the correction processing unit 24 in addition to the correction value for the average movement amount. The correction processing unit 24 translates the selected object according to the average movement amount correction value output from the correction value calculation unit 23, and deforms the selected object according to the deformation amount correction value (step S8). The deformation is, for example, rotation, enlargement / reduction, coordinate conversion, or the like. The correction may be performed both in parallel movement and deformation, or may be performed only in deformation. The correction only for the deformation is performed when the position of the center of gravity of the object is not moved, but there is a position shift at each position in the object.

  The correction by translation does not change the image of the object itself, so the reproducibility of the image is good. On the other hand, it cannot cope with all of the non-uniform displacement in the object. The accuracy of correction is not high. Correction by deformation can deal with individual positional deviations in the object, so that the correction accuracy is high, but the image of the object is changed.

  In consideration of such characteristics, it is preferable to switch according to the attribute whether the correction is parallel movement or deformation is added. Objects whose attributes are characters and figures include many structures such as outlines and lines where misalignments are easily visible. Therefore, it is preferable to correct the deformation of the deformable element. On the other hand, an object whose attribute is a photograph has many continuous gradations and unclear edges, and a correction effect can be sufficiently obtained by parallel movement without deformation.

  Therefore, when determining the deformation amount in step S6, the correction processing unit 24 changes the threshold value according to the attribute information of the object, and in the case of a character or graphic attribute, the correction of the deformation is easy to be performed. In the case of an attribute, it is preferable to adjust so that only translational correction is performed. For example, the correction processing unit 24 holds two threshold values as threshold values for determination. If the attribute of the object is a character or a graphic, the smaller threshold is used, and if the attribute is a photograph, the larger threshold is used. Is used to determine whether the amount of deformation is large. When the threshold is changed, the determination criterion is switched, and when the attribute is a character or a graphic, it is easy to determine that the amount of deformation is larger than that of the photograph attribute. As a result, in the case of a character or a figure, it is easy to apply deformation correction, and in the case of a photo attribute, only parallel movement correction is easily performed.

  When the processes of steps S3 to S8 have not been completed for all the divided objects (step S9; N), the process returns to step S3, and the processes of steps S3 to S8 are repeated for the unfinished objects. When the processes in steps S3 to S8 are completed for all objects (step S9; Y), the correction of the positional deviation is completed.

FIG. 10 shows an example of misregistration between the front and back sides.
For postcards, business cards, price tags, etc., images of the same size are repeatedly arranged on paper, and the front and back surfaces are often printed in the same arrangement. FIG. 10 shows an example in which eight images of the same size are arranged at equal intervals in two columns and four rows by eight-plane allocation. As shown in FIG. 10, the front image a1 was printed without misalignment, but the back image b11 was rotated counterclockwise during printing, so that the front image a1 and the back image were printed on the printed paper c1. A position shift of the image b12 occurs.

  As shown in FIG. 11, the correction value calculation unit 23 divides the eight images allocated on the sheet as objects as shown in FIG. 11, and the parallel movement amount and the deformation amount of each of the eight objects. A correction value is calculated. Since the displacement amount of the positional deviation is small in the object unit, the correction processing unit 24 determines that the deformation amount is not large, and translates the object according to the translation amount calculated for each object. While the entire back image b11 rotates counterclockwise due to the position shift, the corrected back image b13 is as if the positional relationship of each object is opposite to the position shift due to the parallel movement of each object in the vertical and horizontal directions. The positional relationship is changed as if it is rotating clockwise. Thereafter, when a positional deviation occurs during printing, the back image b13 rotates in the counterclockwise direction, and the front image a1 and the back image b13 substantially coincide with each other on the printed paper c2.

FIG. 12 is an example of misregistration between the front and back when a relatively large size image is included.
As shown in FIG. 12, the front image a1 and the back image b21 have the same image arrangement, and a large image d1 that occupies half of the paper and four small images d2 are arranged below the large image d1. As in FIG. 11, when the image b21 on the back surface is rotated during printing, the large image d1 has a large amount of displacement due to the rotation. Nevertheless, when all the objects of the images d1 and d2 are corrected by translation, the front image a2 and the corrected back image b22 of the printed paper c3 are substantially coincident with each other in the position of the small image d2. However, the large image d1 has a relatively large misalignment and correction is insufficient.

  However, according to the correction processing unit 24, it is determined that the object of the large image d1 has a large deformation amount, and not only the parallel movement but also correction by deformation is performed as shown in FIG. As a result, the front image a2 and the corrected back image b23 of the printed paper c4 are substantially coincident in position on the front and back of the large image d1 and the small image d2. Since the correction accuracy due to deformation is high, the degree of coincidence of a particularly large image d1 is high.

As described above, according to the present embodiment, the image processing apparatus 1 acquires the amount of positional deviation of an image printed on paper by image data printing processing, divides the image into objects, For each divided object, a correction value calculation unit 23 that calculates a correction value for misregistration in accordance with the acquired misregistration amount, and correction processing of the image data using the correction value calculated for each object, The correction processing unit 24 corrects the positional deviation.
Since correction is performed in units of objects, the outline of the object is not disturbed or the image of the object is not lost by the correction, and it is possible to correct misalignment with little deterioration in image quality.

Further, the correction value calculation unit 23 calculates the correction value of the parallel movement amount and deformation amount of each object from the displacement amount of the positional deviation, and the correction processing unit 24 selects the object by the correction value of the parallel movement amount for each object. Either the object is translated, the object is deformed by the correction value of the deformation amount, or both the parallel movement and the deformation are executed.
The correction processing unit 24 determines whether or not the deformation amount is large using a threshold value. If the deformation processing unit 24 determines that the deformation amount is not large, the correction processing unit 24 translates the object according to the correction value of the parallel movement amount, and the deformation amount is large. When the determination is made, the object is translated by the parallel movement amount correction value, and the object is deformed by the deformation amount correction value.

  Since misalignment is easy to be visually recognized due to misalignment between objects, which is a clear pattern, by correcting only the misalignment by parallel movement and reducing misalignment, the position of the object itself can be changed without any change. Deviation correction is possible. When correction by translation is insufficient, correction by deformation is added, so that correction with high accuracy can be performed while suppressing deterioration in image quality.

In addition, the said embodiment is a suitable example of this invention, and is not limited to this.
For example, the timing of correction of misalignment by the correction processing unit 24 is not limited to immediately before screen processing after rasterization by the RIP 11. As shown in FIG. 14, before rasterization, at rasterization, at the time of writing image data to the memory 12, immediately after the screen processing, at the time of writing to the memory 14, or at the time of writing by the writing processing unit 15, it is possible.

When correcting before rasterization, the correction processing unit 24 rewrites data such as layout, font, line width, scaling, and rotation of PDL data according to the correction value of the translation amount and deformation amount, and translates the position of the object. To correct the misalignment.
When correcting during rasterization, rewrite intermediate data such as the display list and layout information generated during rasterization, and change the object layout, font, line width, and scaling to move the object position in parallel, or deform it to shift the position. Correct.

When correction is performed when writing to the memories 12 and 14, the correction processing unit 24 changes the address of each pixel to be written to the memories 12 and 14, and writes the address of the pixel after translation or deformation.
When correction is performed after the screen processing, a periodic pattern formed by the screen processing is shifted due to the correction. Therefore, it is preferable to provide a means for correcting a threshold matrix used for the screen processing to prevent image quality deterioration.

The correction timing may be switched according to the attribute information.
For example, since an image having a photograph attribute originally has a raster structure, it is preferably corrected after rasterization. On the other hand, a character attribute image is held as code data on the PDL data and has a small data amount. Therefore, the processing time required for correction can be shortened by correcting an object having a character attribute before rasterization. However, since it is necessary to appropriately process various conditions for rasterization, the processing content is often packaged, and the processing content of rasterization tends to be unified. Correction of translation can be easily performed by changing layout information, but correction of deformation is not easy by packaging of rasterization. Therefore, in the case of character attributes, parallel movement correction is performed before or during rasterization, and deformation correction is performed after rasterization. In the case of deformation after rasterization, more preferable image processing such as smoothing processing on the edge of a character can be selected.

  Further, the above-described misalignment correction processing may be programmed, and misalignment correction according to the present invention may be realized by cooperation of hardware such as a CPU and the program. As the computer-readable medium of the program, a non-volatile memory such as a ROM and a flash memory, and a portable recording medium such as a CD-ROM can be applied. Further, a carrier wave is also applied to the present invention as a medium for providing the program data via a communication line.

1 Image processing device 11 RIP
12 memory 13 screen processing unit 14 memory 15 writing processing unit 21 detection pattern processing unit 22 correction processing unit 23 correction value calculation unit 24 correction processing unit 31 control unit 32 display unit 33 operation unit

Claims (11)

The amount of displacement of an image printed on paper by image data printing processing is acquired, the image is divided into objects, and the position displacement is determined for each divided object according to the amount of displacement acquired. A correction value calculation unit for calculating the correction value of
A correction processing unit that corrects the image data according to the correction value calculated for each object, and corrects the displacement in units of objects;
An image processing apparatus comprising: The correction value calculation unit calculates a correction value for the translation amount and deformation amount of each object from the displacement amount of the positional deviation,
For each object, the correction processing unit translates the object with the correction value for the parallel movement amount, deforms the object with the correction value for the deformation amount, or executes both the parallel movement and the deformation. The image processing apparatus according to claim 1. The correction processing unit determines whether the deformation amount is large using a threshold,
When it is determined that the deformation amount is not large, the object is translated by the correction value of the parallel movement amount,
The image processing apparatus according to claim 2, wherein when it is determined that the amount of deformation is large, the object is translated by the correction value of the parallel movement amount, and the object is deformed by the correction value of the deformation amount.   The image processing apparatus according to claim 3, wherein the correction processing unit changes the threshold value according to object attribute information.   The correction value calculation unit obtains a displacement amount of a positional deviation of a plurality of detection patterns printed around each object, and calculates a correction value for a parallel movement amount and a deformation amount of each object from the deviation amount. The image processing apparatus as described in any one of 2-4.   The image processing apparatus according to claim 1, wherein the correction value calculation unit extracts an object based on typesetting information or layout information of image data, and divides the image into each object.   The image according to claim 1, wherein the correction value calculation unit extracts a closed region having the same attribute as an object based on attribute information of image data, and divides the image into each object. Processing equipment.   The correction value calculating unit detects image edges by processing image data, extracts a closed region in which the detected edges form a contour line as an object, and divides the image into each object. The image processing apparatus according to any one of claims 7 to 9.   The image processing apparatus according to claim 1, wherein the correction value calculation unit extracts a region designated by a user as an object, and divides the image into each object.   The image processing apparatus according to claim 1, wherein the misregistration includes color misregistration, misregistration between front and back surfaces, and misregistration. Obtaining a displacement amount of an image printed on a sheet by a printing process of image data;
Dividing the image into object units, and calculating a correction value for the positional deviation according to the obtained deviation amount for each divided object;
Correcting the image data with the correction value calculated for each object, and correcting the positional deviation in units of objects;
An image processing method including: